Re: Batch function for Ultimapper and Render Map
Wow Matt,
Thanks for spending time writing the script, I'll test it out together with
the Mapify plugin to see which one could suit my needs
Sorry to ask the same question as before, but based on your Jscript for
Ultimapper I'll just need to replace the function with the Ultimapper
function, add the maps variables ( normals, albedo, depth, AO ) and change
the specified ultimapper property to be what is called Ultimapper thingy
at the end of the script
I'll start experimenting with the SDK editor to see if there are some
usefull informations ;)
Thanks again Matt for the script, amazing how you guys could come out with
a solution within few minutes just writing it down! :)
2014/1/9 Matt Lind ml...@carbinestudios.com
// Jscript – will need some tweaking to be functional
RenderMapSequence( 1, 100, “C:\\tmp\\my_sequence.CURRENTFRAME.tga” );
function RenderMapSequence( FrameStart, FrameEnd, FileName )
{
var oPlayControl = Dictionary.GetObject( “PlayControl” );
// get eligible objects from selection
var aFilterNames = new Array( siPolyMeshFilter, siSurfaceMeshFilter );
var oObjects = SIFilter( null, aFilterNames.join(“,”), true, siQuickSearch
);
If ( !oObjects || oObjects.Count =0 ) {
LogMessage( “nothing selected”, siError );
}
for ( var i = 0; i oObjects.Count; i++ ) {
var oObject = oObjects(i);
var oProperties = oObject.Properties.Filter( “rendermap” );
if ( oProperties.Count = 0 ) {
// rendermap property not found
continue;
}
var oRenderMapProperty = oProperties(0);
for ( var j = FrameStart; j = FrameEnd; j++ ) {
var FrameCurrent = j;
// advance timeline
oPlayControl.Parameters( “Current” ).value = FrameCurrent;
// update the parameter defining output image file name
var ImageFileName = FileName.replace( /CURRENTFRAME/,
FrameCurrent );
oRenderMapProperty.Parameters( “imagefilepath” ).value =
ImageFileName;
// execute specified rendermap property
RegenerateMaps ( oRenderMapProperty.FullName );
}
}
}
*From:* softimage-boun...@listproc.autodesk.com [mailto:
softimage-boun...@listproc.autodesk.com] *On Behalf Of *Nicolas Esposito
*Sent:* Wednesday, January 08, 2014 3:19 PM
*To:* softimage@listproc.autodesk.com
*Subject:* Re: Batch function for Ultimapper and Render Map
The same script functionality to execute every frame and update the output
file could be applied to Ultimapper as well, am I correct?
Sorry but I'm not too familiar with scripting, but looking at the manual
looks nothing super-complicated ;)
2014/1/9 Nicolas Esposito 3dv...@gmail.com
This is gold!
Thanks Matt and Alan ;)
2014/1/8 Alan Fregtman alan.fregt...@gmail.com
You might find Sajjad's *Mapify* plugin useful as it can rendermap a
sequence:
http://www.sajjadamjad.com/plugins.html#Mapify
On Wed, Jan 8, 2014 at 5:42 PM, Matt Lind ml...@carbinestudios.com
wrote:
Set up the rendermap property as usual, then write a few lines of script
to execute rendermap every frame while updating the output file path.
Matt
*From:* softimage-boun...@listproc.autodesk.com [mailto:
softimage-boun...@listproc.autodesk.com] *On Behalf Of *Nicolas Esposito
*Sent:* Wednesday, January 08, 2014 2:40 PM
*To:* softimage@listproc.autodesk.com
*Subject:* Batch function for Ultimapper and Render Map
I remember there was backburner for Softimage once but I suppose has been
removed...
So, I need to use ultimapper and render map to render out each frame of my
animation, and if I remember correctly this was doable with backburner
With Soft 2013 there is a way to achieve that?
I'm trying to replicate the effect done by Blur in this video @1.59
http://www.youtube.com/watch?v=38wh5Fn4WEs
The system or the dynamic tassellation/normal map is what I'm trying to
achieve, but I will need the maps in order to test something else
Cheers
Re: Redshift3D Render
Well I finally got nice reason to get 4 Titans.. mmm 4way SLI and 3 27 mon stereo gaming. Soo nice to have same comp good for both job and gaming. On Thu, Jan 9, 2014 at 3:49 AM, Sebastien Sterling sebastien.sterl...@gmail.com wrote: Call me immature, but i kind of love the idea of my gaming pc also being my render farm :P Sod of Nvidia and your flaby Quadro cards there expensive as fuck and you can't play Crysis on them :) On 8 January 2014 23:13, Tim Leydecker bauero...@gmx.de wrote: For a current list of features available as well as a roadmap, I would like to suggest to just go and give it a free try: https://www.redshift3d.com/get-redshift Yes, actually you don´t even have to commit to spending $100 directly, the Free Beta Trial gives you 30 days of full access to Redshift. A special benefit of this free trial option is that you could actually try out how a bunch of machines would run using redshift in a farm or knot. Reading the docs doesn´t require a login: http://docs.redshift3d.com/Default.html Redshift is a really well balanced renderer and I wholehartedly trust in it´s success. With the above opportunity available it is a good time to test it in your production scenario and wheight it against VRay and Arnold, which are also both very nice plattforms with enough momentum to also be around for quite a while. I am sure Redshift is a valuable addition to that arsenal. Cheers, tim On 08.01.2014 22:44, Emilio Hernandez wrote: Backing up Tim, in the forums there is actually a hair test in Softimage with a simple phong shader. And IMHO it looks nice. 2014/1/8 Tim Crowson tim.crow...@magneticdreams.com mailto: tim.crow...@magneticdreams.com They've stated pretty clearly that Hair and Strand support is the next big thing to come... shouldn't be too long now... -Tim On 1/8/2014 3:36 PM, Daniel Kim wrote: I found some weird result of displacement map with Redshift. Elevation is okay, but sometimes I could see weird connection of UVM. All UVM boders wasn't smooth and I had no idea how to fix it. Arnold has that option though : / But more option I need is... hair @__@ and ICE strand --- Daniel Kim Animation Director Professional 3D Generalist http://www.danielkim3d.com --- On Thu, Jan 9, 2014 at 10:31 AM, Emilio Hernandez emi...@e-roja.com mailto:emi...@e-roja.com wrote: Displacement and bump map are there and they work beautiful. They event implement a scalar change range into the displacement node. 2014/1/8 Tim Crowson tim.crow...@magneticdreams.com mailto: tim.crow...@magneticdreams.com I'll repost what I said in the other thread We started using Redshift back in March and pretty much use it exclusively now. Of course it all depends on the needs of the project (and there are still some real limitations). The RS dev team is top notch though. I'm really excited to see how things will be at the end of this calendar year. Redshift development is progressing at a fantastic rate, and the pricing is very competitive. For facilities, even small ones, it does require that you spend some time considering your hardware and infrastructure, especially if you want to start converting CPU farms for GPU rendering, or augmenting them. Fortunately, Redshift isn't licensed per GPU, but per machine, and that should provide some breathing room. To be honest (and I realize we have many Arnold folks here), here at Magnetic we evaluated our rendering options (MR, vRay, 3Delight, Arnold). I even started working on a Soft-to-Modo pipeline. Among these Arnold was the clear winner. That said, we felt that to be useful for us in production, Arnold was too costly a solution for us, both financially and in render time, /considering the kinds of projects we do/ . Then Redshift came along and despite its infancy, really turned our heads. We cautiously began using it on productions, and it has since proven itself for us. Again, it all depends on what kind of project you're working on! You need to evaluate it for yourself of course, but for smaller houses like us, it allows us to produce better looking content faster, while staying in Softimage. And in this economy, we can't argue with that. -Tim On 1/8/2014 2:53 PM, Paul Griswold wrote: I've been using it along side Arnold for quite a while now. I just finished a project for CES entirely in Redshift. I think Redshift falls more into the category of a VRay competitor rather than Arnold. Redshift isn't open the way Arnold is I don't think they intend it to
Re: rigging in xsi vs maya
Autodesk is doing a lot of development in the area of 3D scan data handling. If you look into what is going on in the area of topology data aquisition for architecture, engineering and the military, there is a shift towards 3D pointcloud data which imho is compareable to what 2D tracking as a concept brought us in the 90s. (Facial recognition and finally image based modeling and camera positional data) It is at hand that the more complex, raw 3D point cloud data will need new and abstracted ways of handling and manipulation, filtering options and adaptive control layers for approximated data. The implication such data for 3D animation brings is that the concept of wheighting a fixed number of vertices to a bone may have to be extended beyond a fixed number of polygons. Unfortunately, taking fall-off based volume wheighting as in it´s current level of finesse may give worse results than before, especially if your shape options for the influence volume are limited to capsules, boxes or spheres. I am a bit worried that the process of riggingwheighting an organic character will become even more frustrating and stiff or at least will need even more steps, like creating an extra controlsurface with a fixed number of points and wrapping it around the high-density data. Such a wrap-deformer takes away control. It´s always the rims and little caveats that need extra care. Cheers, tim On 09.01.2014 02:13, Guillaume Laforge wrote: On Wed, Jan 8, 2014 at 7:55 PM, Luc-Eric Rousseau luceri...@gmail.com mailto:luceri...@gmail.com wrote: In the new future ( not talking about autodesk here) I think workflows will standards will be Gator-like tools to deal with topo changes (point clouds tools as necessary also ptex-based workflows) and katana-like proceduralism for render passes-like workflow. I'm still wondering if a company ( not talking about Autodesk here ) will do anything new like that for our little world. Money for such large dev projects is just not in the animation/vfx world anymore. I'm not sarcastic, just realist. So lets embrace old techs like Maya or XSI. They won't evolve too much but won't disappear before many (many) years. Btw, Katana is not the futur, it is now :).
Re: Redshift3D Render
quick one, can your rendermap with redshift? On Thu, Jan 9, 2014 at 7:39 PM, Mirko Jankovic mirkoj.anima...@gmail.comwrote: Well I finally got nice reason to get 4 Titans.. mmm 4way SLI and 3 27 mon stereo gaming. Soo nice to have same comp good for both job and gaming. On Thu, Jan 9, 2014 at 3:49 AM, Sebastien Sterling sebastien.sterl...@gmail.com wrote: Call me immature, but i kind of love the idea of my gaming pc also being my render farm :P Sod of Nvidia and your flaby Quadro cards there expensive as fuck and you can't play Crysis on them :) On 8 January 2014 23:13, Tim Leydecker bauero...@gmx.de wrote: For a current list of features available as well as a roadmap, I would like to suggest to just go and give it a free try: https://www.redshift3d.com/get-redshift Yes, actually you don´t even have to commit to spending $100 directly, the Free Beta Trial gives you 30 days of full access to Redshift. A special benefit of this free trial option is that you could actually try out how a bunch of machines would run using redshift in a farm or knot. Reading the docs doesn´t require a login: http://docs.redshift3d.com/Default.html Redshift is a really well balanced renderer and I wholehartedly trust in it´s success. With the above opportunity available it is a good time to test it in your production scenario and wheight it against VRay and Arnold, which are also both very nice plattforms with enough momentum to also be around for quite a while. I am sure Redshift is a valuable addition to that arsenal. Cheers, tim On 08.01.2014 22:44, Emilio Hernandez wrote: Backing up Tim, in the forums there is actually a hair test in Softimage with a simple phong shader. And IMHO it looks nice. 2014/1/8 Tim Crowson tim.crow...@magneticdreams.com mailto: tim.crow...@magneticdreams.com They've stated pretty clearly that Hair and Strand support is the next big thing to come... shouldn't be too long now... -Tim On 1/8/2014 3:36 PM, Daniel Kim wrote: I found some weird result of displacement map with Redshift. Elevation is okay, but sometimes I could see weird connection of UVM. All UVM boders wasn't smooth and I had no idea how to fix it. Arnold has that option though : / But more option I need is... hair @__@ and ICE strand --- Daniel Kim Animation Director Professional 3D Generalist http://www.danielkim3d.com --- On Thu, Jan 9, 2014 at 10:31 AM, Emilio Hernandez emi...@e-roja.com mailto:emi...@e-roja.com wrote: Displacement and bump map are there and they work beautiful. They event implement a scalar change range into the displacement node. 2014/1/8 Tim Crowson tim.crow...@magneticdreams.com mailto: tim.crow...@magneticdreams.com I'll repost what I said in the other thread We started using Redshift back in March and pretty much use it exclusively now. Of course it all depends on the needs of the project (and there are still some real limitations). The RS dev team is top notch though. I'm really excited to see how things will be at the end of this calendar year. Redshift development is progressing at a fantastic rate, and the pricing is very competitive. For facilities, even small ones, it does require that you spend some time considering your hardware and infrastructure, especially if you want to start converting CPU farms for GPU rendering, or augmenting them. Fortunately, Redshift isn't licensed per GPU, but per machine, and that should provide some breathing room. To be honest (and I realize we have many Arnold folks here), here at Magnetic we evaluated our rendering options (MR, vRay, 3Delight, Arnold). I even started working on a Soft-to-Modo pipeline. Among these Arnold was the clear winner. That said, we felt that to be useful for us in production, Arnold was too costly a solution for us, both financially and in render time, /considering the kinds of projects we do/ . Then Redshift came along and despite its infancy, really turned our heads. We cautiously began using it on productions, and it has since proven itself for us. Again, it all depends on what kind of project you're working on! You need to evaluate it for yourself of course, but for smaller houses like us, it allows us to produce better looking content faster, while staying in Softimage. And in this economy, we can't argue with that. -Tim On 1/8/2014 2:53 PM, Paul Griswold wrote: I've been using it along side Arnold for quite a while now. I just finished a project for CES entirely in Redshift. I think Redshift falls
Re: Redshift3D Render
Afaik it's in development just like hair fur.quick one, can your rendermap with redshift?On Thu, Jan 9, 2014 at 7:39 PM, Mirko Jankovic mirkoj.anima...@gmail.com wrote: Well I finally got nice reason to get 4 Titans.. mmm 4way SLI and 3 27" mon stereo gaming.Soo nice to have same comp good for both job and gaming. On Thu, Jan 9, 2014 at 3:49 AM, Sebastien Sterling sebastien.sterl...@gmail.com wrote: Call me immature, but i kind of love the idea of my gaming pc also being my render farm :PSod of Nvidia and your flaby Quadro cards there expensive as fuck and you can't play Crysis on them :) On 8 January 2014 23:13, Tim Leydecker bauero...@gmx.de wrote: For a current list of features available as well as a roadmap, I would like to suggest to just go and give it a free try: https://www.redshift3d.com/get-redshift Yes, actually you don´t even have to commit to spending $100 directly, the Free Beta Trial gives you 30 days of full access to Redshift. A special benefit of this free trial option is that you could actually try out how a bunch of machines would run using redshift in a farm or knot. Reading the docs doesn´t require a login: http://docs.redshift3d.com/Default.html Redshift is a really well balanced renderer and I wholehartedly trust in it´s success. With the above opportunity available it is a good time to test it in your production scenario and wheight it against VRay and Arnold, which are also both very nice plattforms with enough momentum to also be around for quite a while. I am sure Redshift is a valuable addition to that arsenal. Cheers, tim On 08.01.2014 22:44, Emilio Hernandez wrote: Backing up Tim, in the forums there is actually a hair test in Softimage with a simple phong shader. And IMHO it looks nice. 2014/1/8 Tim Crowson tim.crow...@magneticdreams.com mailto:tim.crow...@magneticdreams.com They've stated pretty clearly that Hair and Strand support is the next big thing to come... shouldn't be too long now... -Tim On 1/8/2014 3:36 PM, Daniel Kim wrote: I found some weird result of displacement map with Redshift. Elevation is okay, but sometimes I could see weird connection of UVM. All UVM boders wasn't smooth and I had no idea how to fix it. Arnold has that option though : / But more option I need is... hair @__@ and ICE strand --- Daniel Kim Animation Director Professional 3D Generalist http://www.danielkim3d.com --- On Thu, Jan 9, 2014 at 10:31 AM, Emilio Hernandez emi...@e-roja.com mailto:emi...@e-roja.com wrote: Displacement and bump map are there and they work beautiful. They event implement a scalar change range into the displacement node. 2014/1/8 Tim Crowson tim.crow...@magneticdreams.com mailto:tim.crow...@magneticdreams.com I'll repost what I said in the other thread We started using Redshift back in March and pretty much use it exclusively now. Of course it all depends on the needs of the project (and there are still some real limitations). The RS dev team is top notch though. I'm really excited to see how things will be at the end of this calendar year. Redshift development is progressing at a fantastic rate, and the pricing is very competitive. For facilities, even small ones, it does require that you spend some time considering your hardware and infrastructure, especially if you want to start converting CPU farms for GPU rendering, or augmenting them. Fortunately, Redshift isn't licensed per GPU, but per machine, and that should provide some breathing room. To be honest (and I realize we have many Arnold folks here), here at Magnetic we evaluated our rendering options (MR, vRay, 3Delight, Arnold). I even started working on a Soft-to-Modo pipeline. Among these Arnold was the clear winner. That said, we felt that to be useful for us in production, Arnold was too costly a solution for us, both financially and in render time, /considering the kinds of projects we do/ . Then Redshift came along and despite its infancy, really turned our heads. We cautiously began using it on productions, and it has since proven itself for us. Again, it all depends on what kind of project you're working on! You need to evaluate it for yourself of course, but for smaller houses like us, it allows us to produce better looking content faster, while staying in Softimage. And in this economy, we can't argue with that. -Tim On 1/8/2014 2:53 PM, Paul Griswold wrote: I've been using it along side Arnold for quite a while now. I just finished a project for CES entirely in Redshift. I think Redshift falls more into the category of a VRay competitor rather than Arnold. Redshift isn't open the way Arnold is I don't think they intend it to be. I've found it to be extremely fast and
[ICE] B-Spline/NURBS curve compound?
Hey there, is anybody aware of an ICE compound solving B-Spline/NURBS curves? You know, like the existing 'Bezier 4' and 'Bezier 5' compounds but for a curve with an arbitrary number of control points? Thanks in advance, cheers, Martin -- Martin Chatterjee [ Freelance Technical Director ] [ http://www.chatterjee.de ] [ https://vimeo.com/chatterjee ]
Re: [ICE] B-Spline/NURBS curve compound?
there is the 'Piecewise Cubic B Spline' Compound you could use as a starting point. Cheers, Vladimir On Thu, Jan 9, 2014 at 11:21 AM, Martin Chatterjee martin.chatterjee.li...@googlemail.com wrote: Hey there, is anybody aware of an ICE compound solving B-Spline/NURBS curves? You know, like the existing 'Bezier 4' and 'Bezier 5' compounds but for a curve with an arbitrary number of control points? Thanks in advance, cheers, Martin -- Martin Chatterjee [ Freelance Technical Director ] [ http://www.chatterjee.de ] [ https://vimeo.com/chatterjee ]
Re: [ICE] B-Spline/NURBS curve compound?
Cheers Vladimir, that definitely helps! Can't believe I actually overlooked this compound in my search... :) Martin -- Martin Chatterjee [ Freelance Technical Director ] [ http://www.chatterjee.de ] [ https://vimeo.com/chatterjee ] On Thu, Jan 9, 2014 at 11:23 AM, Vladimir Jankijevic vladi...@elefantstudios.ch wrote: there is the 'Piecewise Cubic B Spline' Compound you could use as a starting point. Cheers, Vladimir On Thu, Jan 9, 2014 at 11:21 AM, Martin Chatterjee martin.chatterjee.li...@googlemail.com wrote: Hey there, is anybody aware of an ICE compound solving B-Spline/NURBS curves? You know, like the existing 'Bezier 4' and 'Bezier 5' compounds but for a curve with an arbitrary number of control points? Thanks in advance, cheers, Martin -- Martin Chatterjee [ Freelance Technical Director ] [ http://www.chatterjee.de ] [ https://vimeo.com/chatterjee ]
BUG: referenced models with gear rig swapping animation between characters.
Hi guys, we have a bit of a strange situation here, as you can see by the title... A couple of referenced models with characters are swapping the animation in random scenes, the bug seems to appear when the models are referenced but the moment you localize everything it seems to revert to the original state, so im presuming it has something to do with the deltas going haywire but i have no clue as to how to attack this problem since my knowledge of animation and anything tied to animation is very limited. I was hoping someone might have run into this before and might have a quick fix. Cheers, Ogi.
RE: Rendering ZBrush displacement in Soft
Actually finally I managed to work with displacement. However, displacement is not as detailed as I'd like, but boosting with a bump map, it looks fine. Unfortunately, the midpoint seems to be off, I have to somehow tune it. In ZB I set midpoint to 0.5, because if I set it to 0, It looked as if I have no recesses on the skin. Now what I think is that I have to exaggerate the details to make it work properly with skin shading, but that's another story. It looks like that the details are in the texture, but somehow the model doesn't want to reflect it, maybe I should pump up the subdivision in the displacement tab of geoapprox PPG. Cheers Szabolcs PS. For me, after GoZ is used then I have two issues: 1) I can't Alt-Tab to change between tasks, I have to minimize ZBrush, 2) All playback function, simulation is not available anymore in Softimage Cheers From: softimage-boun...@listproc.autodesk.com [mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Emilio Hernandez Sent: Wednesday, January 08, 2014 4:58 PM To: softimage@listproc.autodesk.com Subject: Re: Rendering ZBrush displacement in Soft I also use GoZ a lot, the only bug that happens for me is that if you send from Softimage to Zbrush, make an UV check before start sculpting. Sometimes they got screwed. So export an OBJ and import in Zbrush do your sculpt and sendback to SI. Make sure your UV's are consistent and flip the U before exporting from Zbrush or your map is going to come up flipped. upside down. Regularly I don't subd the mesh in Zbrush more than 4 subd. As you need to subd your mesh also inside SI the same amount to properly displace the geo when rendering. If you find that you need more subd in your sculpt. I willo go back, subd the mesh to get more polys and send it back to Zbrush, so you can go as high as 8 subd from the original mesh, that is a lot. As far as I remember you can use a 32bit depth image to plug that into the scalar change range node. But this depth in the bitmap has nothing to do with the linear workflow, as the linear workflow is related to gamma correct display and rendering, and this maps are to drive values of units to displace the geometry inwards and outwards and to be interpretated by the render engine Here is video that will help to understand the displacement in SI and MR. https://vimeo.com/29898426 . [http://img694.imageshack.us/img694/8965/erojamailpleca.jpg] 2014/1/8 Cristobal Infante cgc...@gmail.commailto:cgc...@gmail.com You read it linear for sure. What exactly is your problem, are you not getting enough detail?. If this is the problem it can also be the UVs are not big enough for each poly. To be honest the final shader doesn't really matter for the disp, in fact GoZ export with a phong. I personally only use Architectural materials when in MR though. On 8 January 2014 12:45, Szabolcs Matefy szabol...@crytek.commailto:szabol...@crytek.com wrote: OK, next question. If you are working in linear workflow, how would you set up the displacement? At this moment I feed the displacement map as Linear, and scale it down with Change Range node. The displacement map is an OpenEXR 32 bit texture, and I am using the MILA mental Ray shaders. Cheers Szabolcs From: softimage-boun...@listproc.autodesk.commailto:softimage-boun...@listproc.autodesk.com [mailto:softimage-boun...@listproc.autodesk.commailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Cristobal Infante Sent: Wednesday, January 08, 2014 12:08 PM To: softimage@listproc.autodesk.commailto:softimage@listproc.autodesk.com Subject: Re: Rendering ZBrush displacement in Soft GoZ for softimage all the way... On 8 January 2014 09:24, Emilio Hernandez emi...@e-roja.commailto:emi...@e-roja.com wrote: The trick is the scalar change range node in the render tree to match the displacement. So MR will know wich values go up and wich values go down. Take a look at the alpha value when exporting the displacement map from Zbrush. That value should be placed in the scalar change range node in the render tree. So let's say that the alpha value in Zbrush for the exported map is 0.054. You should plug the dm in the scalar change range node and place the minimum value to -0.054 ant the max to 0.054. That way MR will know where is 0.0 value and the max and min displacement. [http://img694.imageshack.us/img694/8965/erojamailpleca.jpg] 2014/1/8 Florian Breg florian.b...@gmail.commailto:florian.b...@gmail.com Have you tried GoZ for softimage? I used it a lot recently and it works like a charm. With the click of a button you get your model imported into SI with all the maps attached and a change range node with the right values added to your displacement. It even flips the maps for you. I think you can download it from the pixologic site. Good Luck, Florian Am 08.01.2014 09:18 schrieb Szabolcs Matefy szabol...@crytek.commailto:szabol...@crytek.com: I'm sure that I'm the source of
Re: BUG: referenced models with gear rig swapping animation between characters.
It seems to be throwing a delta invalid error and shouting at me that the emdl file was created in another version, but we are running the same version of 2013 on every comp INFO : 4152 - Data loaded from file Z:\\Rusco_Rig.emdl was created with build number: 11.0.525.0 - compatibility version: 1100 ... On Thu, Jan 9, 2014 at 11:49 AM, Ognjen Vukovic ognj...@gmail.com wrote: Hi guys, we have a bit of a strange situation here, as you can see by the title... A couple of referenced models with characters are swapping the animation in random scenes, the bug seems to appear when the models are referenced but the moment you localize everything it seems to revert to the original state, so im presuming it has something to do with the deltas going haywire but i have no clue as to how to attack this problem since my knowledge of animation and anything tied to animation is very limited. I was hoping someone might have run into this before and might have a quick fix. Cheers, Ogi.
Re: BUG: referenced models with gear rig swapping animation between characters.
That's just an information that the emdl was created with version 11.0.525.0 - 2013 On Thu, Jan 9, 2014 at 12:11 PM, Ognjen Vukovic ognj...@gmail.com wrote: It seems to be throwing a delta invalid error and shouting at me that the emdl file was created in another version, but we are running the same version of 2013 on every comp INFO : 4152 - Data loaded from file Z:\\Rusco_Rig.emdl was created with build number: 11.0.525.0 - compatibility version: 1100 ... On Thu, Jan 9, 2014 at 11:49 AM, Ognjen Vukovic ognj...@gmail.com wrote: Hi guys, we have a bit of a strange situation here, as you can see by the title... A couple of referenced models with characters are swapping the animation in random scenes, the bug seems to appear when the models are referenced but the moment you localize everything it seems to revert to the original state, so im presuming it has something to do with the deltas going haywire but i have no clue as to how to attack this problem since my knowledge of animation and anything tied to animation is very limited. I was hoping someone might have run into this before and might have a quick fix. Cheers, Ogi.
Re: Redshift3D Render
I just added 2 780ti's to my machine use my 680 just to drive my displays. One of the negatives right now with Redshift (AFAIK) is that it's not entirely optimized for more than 2 cards. Once you get 3-4 cards in place you start seeing diminishing returns. But that might be an nVidia problem, not a Redshift problem. I think the main thing you have to consider is - do you already have a renderfarm that is CPU based and are you willing to migrate to GPU-based rendering? I have rack-mounted machines that just have Intel motheboard graphics no space for a GPU, so for now Redshift is limited to just 2 machines. For larger companies there's a space heat issue as well. Running just the 2 780ti's all night long kept my office nice and toasty warm. You need to have a very large power supply as well as a case motherboard that can fit the cards. It'd be very interesting to see a comparison of space, heat and watts consumed for CPU's vs GPU's for rendering... -Paul ᐧ On Thu, Jan 9, 2014 at 3:39 AM, Mirko Jankovic mirkoj.anima...@gmail.comwrote: Well I finally got nice reason to get 4 Titans.. mmm 4way SLI and 3 27 mon stereo gaming. Soo nice to have same comp good for both job and gaming. On Thu, Jan 9, 2014 at 3:49 AM, Sebastien Sterling sebastien.sterl...@gmail.com wrote: Call me immature, but i kind of love the idea of my gaming pc also being my render farm :P Sod of Nvidia and your flaby Quadro cards there expensive as fuck and you can't play Crysis on them :) On 8 January 2014 23:13, Tim Leydecker bauero...@gmx.de wrote: For a current list of features available as well as a roadmap, I would like to suggest to just go and give it a free try: https://www.redshift3d.com/get-redshift Yes, actually you don´t even have to commit to spending $100 directly, the Free Beta Trial gives you 30 days of full access to Redshift. A special benefit of this free trial option is that you could actually try out how a bunch of machines would run using redshift in a farm or knot. Reading the docs doesn´t require a login: http://docs.redshift3d.com/Default.html Redshift is a really well balanced renderer and I wholehartedly trust in it´s success. With the above opportunity available it is a good time to test it in your production scenario and wheight it against VRay and Arnold, which are also both very nice plattforms with enough momentum to also be around for quite a while. I am sure Redshift is a valuable addition to that arsenal. Cheers, tim On 08.01.2014 22:44, Emilio Hernandez wrote: Backing up Tim, in the forums there is actually a hair test in Softimage with a simple phong shader. And IMHO it looks nice. 2014/1/8 Tim Crowson tim.crow...@magneticdreams.com mailto: tim.crow...@magneticdreams.com They've stated pretty clearly that Hair and Strand support is the next big thing to come... shouldn't be too long now... -Tim On 1/8/2014 3:36 PM, Daniel Kim wrote: I found some weird result of displacement map with Redshift. Elevation is okay, but sometimes I could see weird connection of UVM. All UVM boders wasn't smooth and I had no idea how to fix it. Arnold has that option though : / But more option I need is... hair @__@ and ICE strand --- Daniel Kim Animation Director Professional 3D Generalist http://www.danielkim3d.com --- On Thu, Jan 9, 2014 at 10:31 AM, Emilio Hernandez emi...@e-roja.com mailto:emi...@e-roja.com wrote: Displacement and bump map are there and they work beautiful. They event implement a scalar change range into the displacement node. 2014/1/8 Tim Crowson tim.crow...@magneticdreams.com mailto: tim.crow...@magneticdreams.com I'll repost what I said in the other thread We started using Redshift back in March and pretty much use it exclusively now. Of course it all depends on the needs of the project (and there are still some real limitations). The RS dev team is top notch though. I'm really excited to see how things will be at the end of this calendar year. Redshift development is progressing at a fantastic rate, and the pricing is very competitive. For facilities, even small ones, it does require that you spend some time considering your hardware and infrastructure, especially if you want to start converting CPU farms for GPU rendering, or augmenting them. Fortunately, Redshift isn't licensed per GPU, but per machine, and that should provide some breathing room. To be honest (and I realize we have many Arnold folks here), here at Magnetic we evaluated our rendering options (MR, vRay, 3Delight, Arnold). I even started working on a Soft-to-Modo pipeline. Among these Arnold was
Re: BUG: referenced models with gear rig swapping animation between characters.
sorry my bad, still working in autopilot mode. we reverted back to some old models before the shaders where attached and the problem went awey so i cant catch the error now but it might seem to be something linked to the fact that the rigger has the same names for certain geometry in different models? not quite sure yet but im sure it will sneak up again during the day. On Thu, Jan 9, 2014 at 12:15 PM, Vladimir Jankijevic vladi...@elefantstudios.ch wrote: That's just an information that the emdl was created with version 11.0.525.0 - 2013 On Thu, Jan 9, 2014 at 12:11 PM, Ognjen Vukovic ognj...@gmail.com wrote: It seems to be throwing a delta invalid error and shouting at me that the emdl file was created in another version, but we are running the same version of 2013 on every comp INFO : 4152 - Data loaded from file Z:\\Rusco_Rig.emdl was created with build number: 11.0.525.0 - compatibility version: 1100 ... On Thu, Jan 9, 2014 at 11:49 AM, Ognjen Vukovic ognj...@gmail.comwrote: Hi guys, we have a bit of a strange situation here, as you can see by the title... A couple of referenced models with characters are swapping the animation in random scenes, the bug seems to appear when the models are referenced but the moment you localize everything it seems to revert to the original state, so im presuming it has something to do with the deltas going haywire but i have no clue as to how to attack this problem since my knowledge of animation and anything tied to animation is very limited. I was hoping someone might have run into this before and might have a quick fix. Cheers, Ogi.
exporting camera roll - FBX?
I have an animated camera where I've also animated the roll on it. When I export the scene as FBX, the camera imports without the roll. Is there a way to include it? Thanks, Paul ᐧ
Re: Redshift3D Render
how diminishing are we talking? some enthusiast motherboards support upto 6 cards, also does pci-e slot speed make a difference? On Thu, Jan 9, 2014 at 10:40 PM, Paul Griswold pgrisw...@fusiondigitalproductions.com wrote: I just added 2 780ti's to my machine use my 680 just to drive my displays. One of the negatives right now with Redshift (AFAIK) is that it's not entirely optimized for more than 2 cards. Once you get 3-4 cards in place you start seeing diminishing returns. But that might be an nVidia problem, not a Redshift problem. I think the main thing you have to consider is - do you already have a renderfarm that is CPU based and are you willing to migrate to GPU-based rendering? I have rack-mounted machines that just have Intel motheboard graphics no space for a GPU, so for now Redshift is limited to just 2 machines. For larger companies there's a space heat issue as well. Running just the 2 780ti's all night long kept my office nice and toasty warm. You need to have a very large power supply as well as a case motherboard that can fit the cards. It'd be very interesting to see a comparison of space, heat and watts consumed for CPU's vs GPU's for rendering... -Paul ᐧ On Thu, Jan 9, 2014 at 3:39 AM, Mirko Jankovic mirkoj.anima...@gmail.comwrote: Well I finally got nice reason to get 4 Titans.. mmm 4way SLI and 3 27 mon stereo gaming. Soo nice to have same comp good for both job and gaming. On Thu, Jan 9, 2014 at 3:49 AM, Sebastien Sterling sebastien.sterl...@gmail.com wrote: Call me immature, but i kind of love the idea of my gaming pc also being my render farm :P Sod of Nvidia and your flaby Quadro cards there expensive as fuck and you can't play Crysis on them :) On 8 January 2014 23:13, Tim Leydecker bauero...@gmx.de wrote: For a current list of features available as well as a roadmap, I would like to suggest to just go and give it a free try: https://www.redshift3d.com/get-redshift Yes, actually you don´t even have to commit to spending $100 directly, the Free Beta Trial gives you 30 days of full access to Redshift. A special benefit of this free trial option is that you could actually try out how a bunch of machines would run using redshift in a farm or knot. Reading the docs doesn´t require a login: http://docs.redshift3d.com/Default.html Redshift is a really well balanced renderer and I wholehartedly trust in it´s success. With the above opportunity available it is a good time to test it in your production scenario and wheight it against VRay and Arnold, which are also both very nice plattforms with enough momentum to also be around for quite a while. I am sure Redshift is a valuable addition to that arsenal. Cheers, tim On 08.01.2014 22:44, Emilio Hernandez wrote: Backing up Tim, in the forums there is actually a hair test in Softimage with a simple phong shader. And IMHO it looks nice. 2014/1/8 Tim Crowson tim.crow...@magneticdreams.com mailto: tim.crow...@magneticdreams.com They've stated pretty clearly that Hair and Strand support is the next big thing to come... shouldn't be too long now... -Tim On 1/8/2014 3:36 PM, Daniel Kim wrote: I found some weird result of displacement map with Redshift. Elevation is okay, but sometimes I could see weird connection of UVM. All UVM boders wasn't smooth and I had no idea how to fix it. Arnold has that option though : / But more option I need is... hair @__@ and ICE strand --- Daniel Kim Animation Director Professional 3D Generalist http://www.danielkim3d.com --- On Thu, Jan 9, 2014 at 10:31 AM, Emilio Hernandez emi...@e-roja.com mailto:emi...@e-roja.com wrote: Displacement and bump map are there and they work beautiful. They event implement a scalar change range into the displacement node. 2014/1/8 Tim Crowson tim.crow...@magneticdreams.com mailto: tim.crow...@magneticdreams.com I'll repost what I said in the other thread We started using Redshift back in March and pretty much use it exclusively now. Of course it all depends on the needs of the project (and there are still some real limitations). The RS dev team is top notch though. I'm really excited to see how things will be at the end of this calendar year. Redshift development is progressing at a fantastic rate, and the pricing is very competitive. For facilities, even small ones, it does require that you spend some time considering your hardware and infrastructure, especially if you want to start converting CPU farms for GPU rendering, or augmenting them. Fortunately, Redshift isn't licensed per GPU, but per machine, and that should provide some breathing room.
Re: Linking a light (Sun) to a Physical sky ?
Bump... So nobody has a method ? Le 08/01/2014 16:32, olivier jeannel a écrit : Very dumb question, I did it before, I'm sure... How do I link a light (infinite - sun) rotation to a vector (Sun direction) in a Physicla sky property page ? (So that my light is a sun...) I'm in redshift...
Re: rigging in xsi vs maya
I didn't read every posts so maybe my understanding is wrong but based in last replies from Luc-Eric and Tim Leydecker, it sounds like point cloud scanning is a rigging feature. It is not, so lets return to the subject please :). That illustrate well that it is much more easy to put money on new techs (like point cloud scanning, web based applications, etc...) than to think about how to improve/re-design an existing workflow like character rigging ! We saw some new systems in modeling (ZBrush etc...) and rendering (Katana) some years ago, but still nothing in the rigging area. It make sense as rigging is really a different culture. You need to be a good character rigger to understand and build a good rigging system. But being a good character rigger means spend a lot of time on existing tools like Maya or XSI. At the end you think only through the proposed tools of your app. If you are a developer interested in designing a rigging system, it is the opposite problem, you can have a fresh new vision but you can miss important concepts of character rigging in your tool. Interesting subject, if you forget about Maya and XSI :) Cheers, Guillaume Laforge On Thu, Jan 9, 2014 at 4:18 AM, Tim Leydecker bauero...@gmx.de wrote: Autodesk is doing a lot of development in the area of 3D scan data handling. If you look into what is going on in the area of topology data aquisition for architecture, engineering and the military, there is a shift towards 3D pointcloud data which imho is compareable to what 2D tracking as a concept brought us in the 90s. (Facial recognition and finally image based modeling and camera positional data) It is at hand that the more complex, raw 3D point cloud data will need new and abstracted ways of handling and manipulation, filtering options and adaptive control layers for approximated data. The implication such data for 3D animation brings is that the concept of wheighting a fixed number of vertices to a bone may have to be extended beyond a fixed number of polygons. Unfortunately, taking fall-off based volume wheighting as in it´s current level of finesse may give worse results than before, especially if your shape options for the influence volume are limited to capsules, boxes or spheres. I am a bit worried that the process of riggingwheighting an organic character will become even more frustrating and stiff or at least will need even more steps, like creating an extra controlsurface with a fixed number of points and wrapping it around the high-density data. Such a wrap-deformer takes away control. It´s always the rims and little caveats that need extra care. Cheers, tim On 09.01.2014 02:13, Guillaume Laforge wrote: On Wed, Jan 8, 2014 at 7:55 PM, Luc-Eric Rousseau luceri...@gmail.commailto: luceri...@gmail.com wrote: In the new future ( not talking about autodesk here) I think workflows will standards will be Gator-like tools to deal with topo changes (point clouds tools as necessary also ptex-based workflows) and katana-like proceduralism for render passes-like workflow. I'm still wondering if a company ( not talking about Autodesk here ) will do anything new like that for our little world. Money for such large dev projects is just not in the animation/vfx world anymore. I'm not sarcastic, just realist. So lets embrace old techs like Maya or XSI. They won't evolve too much but won't disappear before many (many) years. Btw, Katana is not the futur, it is now :).
Re: Redshift3D Render
There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again, it's a beta and they're continuing to improve things like the proxy system. Once I'm caught up I'm hoping to try rendering the classroom scene and see how it does. -Paul ᐧ
Re: Linking a light (Sun) to a Physical sky ?
rendereditinit physical sky ?this one constrains the light direction into the shader for you (mental Ray) maybe you could copy the expression from here? On 9 January 2014 11:54, olivier jeannel olivier.jean...@noos.fr wrote: Bump... So nobody has a method ? Le 08/01/2014 16:32, olivier jeannel a écrit : Very dumb question, I did it before, I'm sure... How do I link a light (infinite - sun) rotation to a vector (Sun direction) in a Physicla sky property page ? (So that my light is a sun...) I'm in redshift...
Re: rigging in xsi vs maya
Hey Guillaume, go and skin/rig/wheight a raw 3D scan mesh directly to bones. Look at what comes in terms of animation and skeleton recognition in the xbox kinect sdk and the xbox one. Cheers, tim On 09.01.2014 13:09, Guillaume Laforge wrote: I didn't read every posts so maybe my understanding is wrong but based in last replies from Luc-Eric and Tim Leydecker, it sounds like point cloud scanning is a rigging feature. It is not, so lets return to the subject please :). That illustrate well that it is much more easy to put money on new techs (like point cloud scanning, web based applications, etc...) than to think about how to improve/re-design an existing workflow like character rigging ! We saw some new systems in modeling (ZBrush etc...) and rendering (Katana) some years ago, but still nothing in the rigging area. It make sense as rigging is really a different culture. You need to be a good character rigger to understand and build a good rigging system. But being a good character rigger means spend a lot of time on existing tools like Maya or XSI. At the end you think only through the proposed tools of your app. If you are a developer interested in designing a rigging system, it is the opposite problem, you can have a fresh new vision but you can miss important concepts of character rigging in your tool. Interesting subject, if you forget about Maya and XSI :) Cheers, Guillaume Laforge On Thu, Jan 9, 2014 at 4:18 AM, Tim Leydecker bauero...@gmx.de mailto:bauero...@gmx.de wrote: Autodesk is doing a lot of development in the area of 3D scan data handling. If you look into what is going on in the area of topology data aquisition for architecture, engineering and the military, there is a shift towards 3D pointcloud data which imho is compareable to what 2D tracking as a concept brought us in the 90s. (Facial recognition and finally image based modeling and camera positional data) It is at hand that the more complex, raw 3D point cloud data will need new and abstracted ways of handling and manipulation, filtering options and adaptive control layers for approximated data. The implication such data for 3D animation brings is that the concept of wheighting a fixed number of vertices to a bone may have to be extended beyond a fixed number of polygons. Unfortunately, taking fall-off based volume wheighting as in it´s current level of finesse may give worse results than before, especially if your shape options for the influence volume are limited to capsules, boxes or spheres. I am a bit worried that the process of riggingwheighting an organic character will become even more frustrating and stiff or at least will need even more steps, like creating an extra controlsurface with a fixed number of points and wrapping it around the high-density data. Such a wrap-deformer takes away control. It´s always the rims and little caveats that need extra care. Cheers, tim On 09.01.2014 02:13, Guillaume Laforge wrote: On Wed, Jan 8, 2014 at 7:55 PM, Luc-Eric Rousseau luceri...@gmail.com mailto:luceri...@gmail.com mailto:luceri...@gmail.com mailto:luceri...@gmail.com wrote: In the new future ( not talking about autodesk here) I think workflows will standards will be Gator-like tools to deal with topo changes (point clouds tools as necessary also ptex-based workflows) and katana-like proceduralism for render passes-like workflow. I'm still wondering if a company ( not talking about Autodesk here ) will do anything new like that for our little world. Money for such large dev projects is just not in the animation/vfx world anymore. I'm not sarcastic, just realist. So lets embrace old techs like Maya or XSI. They won't evolve too much but won't disappear before many (many) years. Btw, Katana is not the futur, it is now :).
Re: exporting camera roll - FBX?
Plot it before you export? The roll I believe is from a direction constraint. On Thu, Jan 9, 2014 at 6:46 AM, Paul Griswold pgrisw...@fusiondigitalproductions.com wrote: I have an animated camera where I've also animated the roll on it. When I export the scene as FBX, the camera imports without the roll. Is there a way to include it? Thanks, Paul ᐧ
Re: exporting camera roll - FBX?
I've tried both plotted and unplotted. It seems to be a Fusion issue, not Softimage. For some reason Fusion can't see animated roll on the camera, even when it's plotted. -Paul ᐧ On Thu, Jan 9, 2014 at 7:34 AM, David Barosin dbaro...@gmail.com wrote: Plot it before you export? The roll I believe is from a direction constraint. On Thu, Jan 9, 2014 at 6:46 AM, Paul Griswold pgrisw...@fusiondigitalproductions.com wrote: I have an animated camera where I've also animated the roll on it. When I export the scene as FBX, the camera imports without the roll. Is there a way to include it? Thanks, Paul ᐧ
Re: Linking a light (Sun) to a Physical sky ?
N There is a Render /Edit/Create RedShift SkyShader I NEVER go through these menus.. Bangging my head on the desk Pulverized by the shame... Thank you oh grandmaster of the infinite knowledge ! Le 09/01/2014 13:14, Rob Chapman a écrit : rendereditinit physical sky ?this one constrains the light direction into the shader for you (mental Ray) maybe you could copy the expression from here? On 9 January 2014 11:54, olivier jeannel olivier.jean...@noos.fr mailto:olivier.jean...@noos.fr wrote: Bump... So nobody has a method ? Le 08/01/2014 16:32, olivier jeannel a écrit : Very dumb question, I did it before, I'm sure... How do I link a light (infinite - sun) rotation to a vector (Sun direction) in a Physicla sky property page ? (So that my light is a sun...) I'm in redshift...
Re: exporting camera roll - FBX?
Roll is really just rotation. Is any rotation making it over? To be thorough, constrain a null to the soft camera - plot it and copy the animation back to the camera. Export the camera with the interest. On Thu, Jan 9, 2014 at 7:35 AM, Paul Griswold pgrisw...@fusiondigitalproductions.com wrote: I've tried both plotted and unplotted. It seems to be a Fusion issue, not Softimage. For some reason Fusion can't see animated roll on the camera, even when it's plotted. -Paul ᐧ On Thu, Jan 9, 2014 at 7:34 AM, David Barosin dbaro...@gmail.com wrote: Plot it before you export? The roll I believe is from a direction constraint. On Thu, Jan 9, 2014 at 6:46 AM, Paul Griswold pgrisw...@fusiondigitalproductions.com wrote: I have an animated camera where I've also animated the roll on it. When I export the scene as FBX, the camera imports without the roll. Is there a way to include it? Thanks, Paul ᐧ
Re: rigging in xsi vs maya
Why not a node based rigging system ? (not necessarily an ice node system) but its own thing, you arrange your nulls, you add rig trees to them in a small interface graph where you have nodes for different behaviours like ik, fk, hik, twist, strech, you plug the nulls according to the hierarchy you want, each node has its own params so you can expose or lock or modify them in the rig or synoptic. i'm sure such a system wouldn't cover everything, its often what i get told, that rigging is so complex a proses that in the end the longest traditional method is the only one that allows for the flexibility and reactivity necessary for a pipe. in spite of this i think such a system has merrit and deserves to go past prototype, if only to offer another perspective. its quite probable that neither xsi or mayas architecture is able to accommodate such a system natively, but plug-ins like yeti are basically like their own independent little engines running within the shell of a dcc, the same is true for fabric i assume. On 9 January 2014 13:09, Guillaume Laforge guillaume.laforge...@gmail.comwrote: I didn't read every posts so maybe my understanding is wrong but based in last replies from Luc-Eric and Tim Leydecker, it sounds like point cloud scanning is a rigging feature. It is not, so lets return to the subject please :). That illustrate well that it is much more easy to put money on new techs (like point cloud scanning, web based applications, etc...) than to think about how to improve/re-design an existing workflow like character rigging ! We saw some new systems in modeling (ZBrush etc...) and rendering (Katana) some years ago, but still nothing in the rigging area. It make sense as rigging is really a different culture. You need to be a good character rigger to understand and build a good rigging system. But being a good character rigger means spend a lot of time on existing tools like Maya or XSI. At the end you think only through the proposed tools of your app. If you are a developer interested in designing a rigging system, it is the opposite problem, you can have a fresh new vision but you can miss important concepts of character rigging in your tool. Interesting subject, if you forget about Maya and XSI :) Cheers, Guillaume Laforge On Thu, Jan 9, 2014 at 4:18 AM, Tim Leydecker bauero...@gmx.de wrote: Autodesk is doing a lot of development in the area of 3D scan data handling. If you look into what is going on in the area of topology data aquisition for architecture, engineering and the military, there is a shift towards 3D pointcloud data which imho is compareable to what 2D tracking as a concept brought us in the 90s. (Facial recognition and finally image based modeling and camera positional data) It is at hand that the more complex, raw 3D point cloud data will need new and abstracted ways of handling and manipulation, filtering options and adaptive control layers for approximated data. The implication such data for 3D animation brings is that the concept of wheighting a fixed number of vertices to a bone may have to be extended beyond a fixed number of polygons. Unfortunately, taking fall-off based volume wheighting as in it´s current level of finesse may give worse results than before, especially if your shape options for the influence volume are limited to capsules, boxes or spheres. I am a bit worried that the process of riggingwheighting an organic character will become even more frustrating and stiff or at least will need even more steps, like creating an extra controlsurface with a fixed number of points and wrapping it around the high-density data. Such a wrap-deformer takes away control. It´s always the rims and little caveats that need extra care. Cheers, tim On 09.01.2014 02:13, Guillaume Laforge wrote: On Wed, Jan 8, 2014 at 7:55 PM, Luc-Eric Rousseau luceri...@gmail.commailto: luceri...@gmail.com wrote: In the new future ( not talking about autodesk here) I think workflows will standards will be Gator-like tools to deal with topo changes (point clouds tools as necessary also ptex-based workflows) and katana-like proceduralism for render passes-like workflow. I'm still wondering if a company ( not talking about Autodesk here ) will do anything new like that for our little world. Money for such large dev projects is just not in the animation/vfx world anymore. I'm not sarcastic, just realist. So lets embrace old techs like Maya or XSI. They won't evolve too much but won't disappear before many (many) years. Btw, Katana is not the futur, it is now :).
Aw: Re: exporting camera roll - FBX?
Fusion is completely unable to properly import XSI cameras - I already tried pretty much everything. We ended up exporting to maya via alembic (cause fbx cameras produce a mismatch) and export a .ma file from there -_- Gesendet:Donnerstag, 09. Januar 2014 um 13:51 Uhr Von:Paul Griswold pgrisw...@fusiondigitalproductions.com An:softimage@listproc.autodesk.com softimage@listproc.autodesk.com Betreff:Re: exporting camera roll - FBX? Haha - you must be reading my mind. Thats the next thing I tried. Oddly enough, it still doesnt work. I even tried plotting the null exporting it, but the null came in with no animation at all. I think Fusion can read dotXSI cameras, so thats next on my list to test... -Paul On Thu, Jan 9, 2014 at 7:46 AM, David Barosin dbaro...@gmail.com wrote: Roll is really just rotation. Is any rotation making it over? To be thorough, constrain a null to the soft camera - plot it and copy the animation back to the camera. Export the camera with the interest. On Thu, Jan 9, 2014 at 7:35 AM, Paul Griswold pgrisw...@fusiondigitalproductions.com wrote: Ive tried both plotted and unplotted. It seems to be a Fusion issue, not Softimage. For some reason Fusion cant see animated roll on the camera, even when its plotted. -Paul On Thu, Jan 9, 2014 at 7:34 AM, David Barosin dbaro...@gmail.com wrote: Plot it before you export? The roll I believe is from a direction constraint. On Thu, Jan 9, 2014 at 6:46 AM, Paul Griswold pgrisw...@fusiondigitalproductions.com wrote: I have an animated camera where Ive also animated the roll on it. When I export the scene as FBX, the camera imports without the roll. Is there a way to include it? Thanks, Paul
Re: exporting camera roll - FBX?
Here is my tip for exporting cameras to other packages and never had any
trouble. Animate your camera and when ready first plot transformations of
the camera itself and then the interest. Then i plot the constrained
transforms from the camera (for the roll) and then delete the interest form
the scene, leaving just the camera object.
That worked perfect for me, even in maya recognizes de image plane property
after i had a rotoscope image on my camera viewport.
Hope it helps.
Francisco.
On Thursday, January 9, 2014, Leo Quensel wrote:
Fusion is completely unable to properly import XSI cameras - I already
tried pretty much everything. We ended up exporting to maya via alembic
(cause fbx cameras produce a mismatch) and export a .ma file from there -_-
*Gesendet:* Donnerstag, 09. Januar 2014 um 13:51 Uhr
*Von:* Paul Griswold
pgrisw...@fusiondigitalproductions.comjavascript:_e({}, 'cvml',
'pgrisw...@fusiondigitalproductions.com');
*An:* softimage@listproc.autodesk.com javascript:_e({}, 'cvml',
'softimage@listproc.autodesk.com');
softimage@listproc.autodesk.comjavascript:_e({}, 'cvml',
'softimage@listproc.autodesk.com');
*Betreff:* Re: exporting camera roll - FBX?
Haha - you must be reading my mind. That's the next thing I tried.
Oddly enough, it still doesn't work. I even tried plotting the null
exporting it, but the null came in with no animation at all.
I think Fusion can read dotXSI cameras, so that's next on my list to
test...
-Paul
ᐧ
On Thu, Jan 9, 2014 at 7:46 AM, David Barosin dbaro...@gmail.com wrote:
Roll is really just rotation. Is any rotation making it over?
To be thorough, constrain a null to the soft camera - plot it and copy the
animation back to the camera.
Export the camera with the interest.
On Thu, Jan 9, 2014 at 7:35 AM, Paul Griswold
pgrisw...@fusiondigitalproductions.com wrote:
I've tried both plotted and unplotted.
It seems to be a Fusion issue, not Softimage. For some reason Fusion
can't see animated roll on the camera, even when it's plotted.
-Paul
ᐧ
On Thu, Jan 9, 2014 at 7:34 AM, David Barosin dbaro...@gmail.com wrote:
Plot it before you export? The roll I believe is from a direction
constraint.
On Thu, Jan 9, 2014 at 6:46 AM, Paul Griswold
pgrisw...@fusiondigitalproductions.com wrote:
I have an animated camera where I've also animated the roll on it.
Re: Redshift3D Render
Here are some details from testing I did a while ago: On Thu, Jan 9, 2014 at 1:11 PM, Paul Griswold pgrisw...@fusiondigitalproductions.com wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again, it's a beta and they're continuing to improve things like the proxy system. Once I'm caught up I'm hoping to try rendering the classroom scene and see how it does. -Paul ᐧ
Re: rigging in xsi vs maya
go and skin/rig/wheight a raw 3D scan mesh directly to bones. What would be a typical scenario for this? The point count is adjustable (at the expense of detail), but the topology will aways be a mess unless properly retopo-ed, wouldn't it? I agree to the rigging paradigm needing some rethinking. I grew up with black box systems like Character Studio and CAT. Creating a rig based on those takes only minutes to hours, not days, but they lack customizability. Yet, results were good enough that I was constantly asking myself as to why anyone could possible want to use anything else for 90% of the work you see being produced anywhere. It's such a huge cost factor, both in terms of time it takes to create the rig and time it takes to trouble shoot and maintain it if it breaks (which the black box systems next to never do) or needs extensions. Autoriggers (Gear etc) reduce the creation time factor at the expense of flexibility, yet the maintenance aspect stays to a certain degree. What I also miss in them is the ability to have a mesh enveloped to joints and just put the rig on top, allowing to test deformations directly by posing the envelope rig without having to create a control rig - a given with the black box systems because the control rig _is_ the envelope rig. The only thing I know that works in a similar way is Motion Builder, in that you import your enveloped mesh and joints and apply rigging solvers to it, again at the expense of flexibility - it only supports humanoid and 4-legged creatures. Fabric/Osirirs looks like it could deliver such a paradigm change - a modular rigging system where the building blocks are encapsulated and the asset that the user interacts with in the scene is light-weight, fast and easy to manipulate, and hard to break. I'm really looking forward to that, even though flexibility beyond a certain point will probably need to be paid for with programming knowledge and -time again. Look at what comes in terms of animation and skeleton recognition in the xbox kinect sdk and the xbox one. Cheers, tim On 09.01.2014 13:09, Guillaume Laforge wrote: I didn't read every posts so maybe my understanding is wrong but based in last replies from Luc-Eric and Tim Leydecker, it sounds like point cloud scanning is a rigging feature. It is not, so lets return to the subject please :). That illustrate well that it is much more easy to put money on new techs (like point cloud scanning, web based applications, etc...) than to think about how to improve/re-design an existing workflow like character rigging ! We saw some new systems in modeling (ZBrush etc...) and rendering (Katana) some years ago, but still nothing in the rigging area. It make sense as rigging is really a different culture. You need to be a good character rigger to understand and build a good rigging system. But being a good character rigger means spend a lot of time on existing tools like Maya or XSI. At the end you think only through the proposed tools of your app. If you are a developer interested in designing a rigging system, it is the opposite problem, you can have a fresh new vision but you can miss important concepts of character rigging in your tool. Interesting subject, if you forget about Maya and XSI :) Cheers, Guillaume Laforge On Thu, Jan 9, 2014 at 4:18 AM, Tim Leydecker bauero...@gmx.de mailto:bauero...@gmx.de wrote: Autodesk is doing a lot of development in the area of 3D scan data handling. If you look into what is going on in the area of topology data aquisition for architecture, engineering and the military, there is a shift towards 3D pointcloud data which imho is compareable to what 2D tracking as a concept brought us in the 90s. (Facial recognition and finally image based modeling and camera positional data) It is at hand that the more complex, raw 3D point cloud data will need new and abstracted ways of handling and manipulation, filtering options and adaptive control layers for approximated data. The implication such data for 3D animation brings is that the concept of wheighting a fixed number of vertices to a bone may have to be extended beyond a fixed number of polygons. Unfortunately, taking fall-off based volume wheighting as in it´s current level of finesse may give worse results than before, especially if your shape options for the influence volume are limited to capsules, boxes or spheres. I am a bit worried that the process of riggingwheighting an organic character will become even more frustrating and stiff or at least will need even more steps, like creating an extra controlsurface with a fixed number of points and wrapping it around the high-density data. Such a wrap-deformer takes away control. It´s always the rims and little caveats that need extra care. Cheers, tim
Re: Stopping simulation!?
Thanks Oscar. I used to do that, but Stephen Blair or Luc Eric informed at one point that it had been fixed from 2012 or 2013 and it was working fine in 2013 SP1 here in the fall. I will try the kb arrow. At leats Mootz tools react to 2 x CTRL. Morten Den 8. januar 2014 kl. 18:51 skrev Oscar Juarez tridi.animei...@gmail.com: You can try this http://xsisupport.com/2011/03/11/softimage-blog-%C2%BB-work-around-to-problems-with-tablets-and-softimage/ http://xsisupport.com/2011/03/11/softimage-blog-%C2%BB-work-around-to-problems-with-tablets-and-softimage/ If it still not working, try to stop playback with keyboard, down arrow. On Wed, Jan 8, 2014 at 4:28 PM, Emilio Hernandez emi...@e-roja.com mailto:emi...@e-roja.com wrote: Now that you mention I am having the same issue. Any ideas? Thx 2014/1/8 Morten Bartholdy x...@colorshopvfx.dk mailto:x...@colorshopvfx.dk I have had the doubtful priviledge of having to do a variety of simulation lately in Soft, and if I remember correctly, in the fall I was able to stop simulations by mouse and pen input, but now I can't do that any longer - actually no matter how simple the sim is. I am on SI 2013 SP1 Win7 and the only thing I have installed in the meantime is Momentum. I have not tried removing it, and before doing that I just want to hear if anyone else has experienced something like this and know of a fix? Thanks Morten
Re: Redshift3D Render
Daniel, You mentioned using it in conjunction with Arnoldwhat way would you use both together...as in render some elements in Arnold and Redshift? On Thu, Jan 9, 2014 at 8:26 AM, Mirko Jankovic mirkoj.anima...@gmail.comwrote: Here are some details from testing I did a while ago: On Thu, Jan 9, 2014 at 1:11 PM, Paul Griswold pgrisw...@fusiondigitalproductions.com wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again, it's a beta and they're continuing to improve things like the proxy system. Once I'm caught up I'm hoping to try rendering the classroom scene and see how it does. -Paul ᐧ
Re: rumor, Soft dead within the next year
Yes please! all mornings i wake up, and the first thing i do is check mails (bad habit) and find myself with the rumor that soft... and its not very possitive checking it every single day for the last 20 days. F. On Wednesday, January 8, 2014, Rob Wuijster wrote: Hi chaps, Now we're moving on to a render discussion about Redshift and such, can we stop using this thread? Oterwise this 'doom thread' will live on for a while ;-) Rob \/-\/\/ On 8-1-2014 20:04, Tim Leydecker wrote: It´s worth using the redshift3d shaders, the new blend material is really nice, normal map blending works nice and the conductor/dielectric option to drive reflection gives believable metal reflection behaviour results easily. You´ll also get better (lights/shadow) sampling compared to using default shaders. Imho, if you spent time with mR or VRay or Arnold shaders, you will have no problem transfering your knowledge to Redshift3D. In terms of benefiting from speed while tweaking, go and set the renderers threshold to 0.2 or even higher, I find that is good enough for judging light/color intensities and gives fast turnaround. Personally, I tend to push per light lightsamples higher than default, even if that is not neccessary in Redshift3D´s unified sampling aproach, to me it feels I have influence on the wheight of samples anyway. Enjoy. It´s really, really cool. Cheers, tim On 08.01.2014 19:08, Byron Nash wrote: When switching over to Redshift, are you all typically redoing the shaders using the Redshift ones or trying to rely on the compatibility with standard ones? I'm interested to check it out but would like to approach it correctly. Thanks, Byron On Wed, Jan 8, 2014 at 11:41 AM, Emilio Hernandez emi...@e-roja.com mailto:emi...@e-roja.com wrote: It sounds promising. I don't know. The funny thing is that Quadros actually render slower than GTX in my experience. As they have lower CUDA cores. My GTX470 alone rendered faster than a Quadro 3000. As the GTX is more focused to games and Quadros to faster video display processing, the Quadros have a lower memory bandwith and less CUDA cores. At least from the last comparisions I have doing in the Nvidia site. Actually I was planning to upgrade my GTX470 to a GTX 780Ti instead of the Titan. A few bucks off the price and it has excellent specs. GTX 780 Ti GPU Engine Specs: 2880CUDA Cores 875Base Clock (MHz) 928Boost Clock (MHz) 210Texture Fill Rate (GigaTexels/sec) GTX 780 Ti Memory Specs: 7.0 GbpsMemory Clock 3072 MBStandard Memory Config GDDR5Memory Interface 384-bitMemory Interface Width 336Memory Bandwidth (GB/sec) From this numbers what you are looking for, is to see which GPU will perform faster are the number of CUDA Cores and the memory bandwith. The higher the better. As the memory bandwith is how fast the data can be transfered to memory to be processed by the CUDA cores. Some guys are already using Redshift with RoyalRender. I don't how fast they are rendering, but now you can have a render farm with cheap processors and a couple of this GPU inside. A quick example. The same scene in round numbers per frame in my machine. Arnold: 15 min Redhsfit: 4 min So you can expect at least a reduction of 73% in your render times. 2014/1/8 Dan Yargici danyarg...@gmail.com mailto:danyarg...@gmail.com Anyone tried using gpubox with Redshift? http://renegatt.com/ - No virus found in this message. Checked by AVG - www.avg.com Version: 2014.0.4259 / Virus Database: 3658/6986 - Release Date: 01/08/14
Re: [ICE] B-Spline/NURBS curve compound?
I did some experiment in ICE with Cubic B Spline. You can see the result on si-community, there even some compound you can download and experiment with. http://www.si-community.com/community/viewtopic.php?f=15t=1802 Hope that help. Daniel On Thu, Jan 9, 2014 at 5:38 AM, Martin Chatterjee martin.chatterjee.li...@googlemail.com wrote: Cheers Vladimir, that definitely helps! Can't believe I actually overlooked this compound in my search... :) Martin -- Martin Chatterjee [ Freelance Technical Director ] [ http://www.chatterjee.de ] [ https://vimeo.com/chatterjee ] On Thu, Jan 9, 2014 at 11:23 AM, Vladimir Jankijevic vladi...@elefantstudios.ch wrote: there is the 'Piecewise Cubic B Spline' Compound you could use as a starting point. Cheers, Vladimir On Thu, Jan 9, 2014 at 11:21 AM, Martin Chatterjee martin.chatterjee.li...@googlemail.com wrote: Hey there, is anybody aware of an ICE compound solving B-Spline/NURBS curves? You know, like the existing 'Bezier 4' and 'Bezier 5' compounds but for a curve with an arbitrary number of control points? Thanks in advance, cheers, Martin -- Martin Chatterjee [ Freelance Technical Director ] [ http://www.chatterjee.de ] [ https://vimeo.com/chatterjee ]
Re: rigging in xsi vs maya
What would be a typical scenario for this? The point count is adjustable (at the expense of detail), but the topology will aways be a mess unless properly retopo-ed, wouldn't it? That is what I meant to suggest when I wrote: It is at hand that the more complex, raw 3D point cloud data will need new and abstracted ways of handling and manipulation, filtering options and adaptive control layers for approximated data Basically, a rigged lower density mesh and a displacement trying to capture small detail while at the same time loosing control over how that detail will actually react, that´s sort of the established standard for working with high detail. Reduce detail until you can handle it and hope nobody will notice. Depending on personality, wave away concerns. The reason why I suggest to go and skin/rig/wheight a raw 3D scan mesh directly to bones: That is the data you want to animate, everything else is already yet another degraded derrivate. Going and trying it will show the limitations of current toolsets. Then do some cloth sim ontop and fix interpenetration issues. Or first of all wait for collision simulation to finish. Like using *.jpg´s with lossy compression as the input for color grading, then re-compressing again as a lossy *.jpg and wondering why there´s block artifacts. Cheers, tim On 09.01.2014 14:34, Stefan Kubicek wrote: go and skin/rig/wheight a raw 3D scan mesh directly to bones. What would be a typical scenario for this? The point count is adjustable (at the expense of detail), but the topology will aways be a mess unless properly retopo-ed, wouldn't it? I agree to the rigging paradigm needing some rethinking. I grew up with black box systems like Character Studio and CAT. Creating a rig based on those takes only minutes to hours, not days, but they lack customizability. Yet, results were good enough that I was constantly asking myself as to why anyone could possible want to use anything else for 90% of the work you see being produced anywhere. It's such a huge cost factor, both in terms of time it takes to create the rig and time it takes to trouble shoot and maintain it if it breaks (which the black box systems next to never do) or needs extensions. Autoriggers (Gear etc) reduce the creation time factor at the expense of flexibility, yet the maintenance aspect stays to a certain degree. What I also miss in them is the ability to have a mesh enveloped to joints and just put the rig on top, allowing to test deformations directly by posing the envelope rig without having to create a control rig - a given with the black box systems because the control rig _is_ the envelope rig. The only thing I know that works in a similar way is Motion Builder, in that you import your enveloped mesh and joints and apply rigging solvers to it, again at the expense of flexibility - it only supports humanoid and 4-legged creatures. Fabric/Osirirs looks like it could deliver such a paradigm change - a modular rigging system where the building blocks are encapsulated and the asset that the user interacts with in the scene is light-weight, fast and easy to manipulate, and hard to break. I'm really looking forward to that, even though flexibility beyond a certain point will probably need to be paid for with programming knowledge and -time again. Look at what comes in terms of animation and skeleton recognition in the xbox kinect sdk and the xbox one. Cheers, tim On 09.01.2014 13:09, Guillaume Laforge wrote: I didn't read every posts so maybe my understanding is wrong but based in last replies from Luc-Eric and Tim Leydecker, it sounds like point cloud scanning is a rigging feature. It is not, so lets return to the subject please :). That illustrate well that it is much more easy to put money on new techs (like point cloud scanning, web based applications, etc...) than to think about how to improve/re-design an existing workflow like character rigging ! We saw some new systems in modeling (ZBrush etc...) and rendering (Katana) some years ago, but still nothing in the rigging area. It make sense as rigging is really a different culture. You need to be a good character rigger to understand and build a good rigging system. But being a good character rigger means spend a lot of time on existing tools like Maya or XSI. At the end you think only through the proposed tools of your app. If you are a developer interested in designing a rigging system, it is the opposite problem, you can have a fresh new vision but you can miss important concepts of character rigging in your tool. Interesting subject, if you forget about Maya and XSI :) Cheers, Guillaume Laforge On Thu, Jan 9, 2014 at 4:18 AM, Tim Leydecker bauero...@gmx.de mailto:bauero...@gmx.de wrote: Autodesk is doing a lot of development in the area of 3D scan data handling. If you look into what is going on in the area of topology data aquisition for architecture,
Re: rumor, Soft dead within the next year
Am 1/9/2014 2:45 PM, schrieb Francisco Criado: Yes please! all mornings i wake up, and the first thing i do is check mails (bad habit) and find myself with the rumor that soft... and its not very possitive checking it every single day for the last 20 days. F. yep, but the good thing is, now that we have 2014, we got another Year for SI! (Now it will (maybe) die at the end of 2015!) Walter -- *Walter Volbers* Senior Animator *FIFTYEIGHT*3D Animation Digital Effects GmbH Kontorhaus Osthafen Lindleystraße 12 60314 Frankfurt am Main Germany Telefon +49 (0) 69.48 000 55.50 Telefax +49 (0) 69.48 000 55.15 _mailto:w...@fiftyeight.com http://www.fiftyeight.com _ ESC*58* Eine Kooperation der escape GmbH und der FIFTYEIGHT3D GmbH _http://www.ESC58.de _
Re: BUG: referenced models with gear rig swapping animation between characters.
I would double check that the deltas have the correct model target they are trying to be applied to in their PPG. There shouldn't be a problem with meshes named the same in both models. The whole point of the model is to give things a unique namespace and encapsulate the asset. Eric T. On Thursday, January 09, 2014 6:44:45 AM, Ognjen Vukovic wrote: sorry my bad, still working in autopilot mode. we reverted back to some old models before the shaders where attached and the problem went awey so i cant catch the error now but it might seem to be something linked to the fact that the rigger has the same names for certain geometry in different models? not quite sure yet but im sure it will sneak up again during the day. On Thu, Jan 9, 2014 at 12:15 PM, Vladimir Jankijevic vladi...@elefantstudios.ch mailto:vladi...@elefantstudios.ch wrote: That's just an information that the emdl was created with version 11.0.525.0 - 2013 On Thu, Jan 9, 2014 at 12:11 PM, Ognjen Vukovic ognj...@gmail.com mailto:ognj...@gmail.com wrote: It seems to be throwing a delta invalid error and shouting at me that the emdl file was created in another version, but we are running the same version of 2013 on every comp INFO : 4152 - Data loaded from file Z:\\Rusco_Rig.emdl was created with build number: 11.0.525.0 - compatibility version: 1100 ... On Thu, Jan 9, 2014 at 11:49 AM, Ognjen Vukovic ognj...@gmail.com mailto:ognj...@gmail.com wrote: Hi guys, we have a bit of a strange situation here, as you can see by the title... A couple of referenced models with characters are swapping the animation in random scenes, the bug seems to appear when the models are referenced but the moment you localize everything it seems to revert to the original state, so im presuming it has something to do with the deltas going haywire but i have no clue as to how to attack this problem since my knowledge of animation and anything tied to animation is very limited. I was hoping someone might have run into this before and might have a quick fix. Cheers, Ogi.
Re: rigging in xsi vs maya
Sebastian, look at ILM's Block Party 2 rigging system. On 1/9/2014 7:53 AM, Sebastien Sterling wrote: Why not a node based rigging system ? (not necessarily an ice node system) but its own thing, you arrange your nulls, you add rig trees to them in a small interface graph where you have nodes for different behaviours like ik, fk, hik, twist, strech, you plug the nulls according to the hierarchy you want, each node has its own params so you can expose or lock or modify them in the rig or synoptic. i'm sure such a system wouldn't cover everything, its often what i get told, that rigging is so complex a proses that in the end the longest traditional method is the only one that allows for the flexibility and reactivity necessary for a pipe. in spite of this i think such a system has merrit and deserves to go past prototype, if only to offer another perspective. its quite probable that neither xsi or mayas architecture is able to accommodate such a system natively, but plug-ins like yeti are basically like their own independent little engines running within the shell of a dcc, the same is true for fabric i assume.
Re: Rendering ZBrush displacement in Soft
yeah, about fine detail and skin shading, subsurface scattering does tend to cover up fine surface detail – recesses and wrinkles in the skin in particular. just think of it: a wrinkle on top of the skin is like a fine ridge. without scattering, if the light comes from the left, the left of that ridge would be bright and the right in shadow. but the scattering makes the shadow part light up – and if your SSS shader compensates the diffuse (which it does by default) the left part will be a bit less bright than it would be without scattering. so as a net result this counters the diffuse light, effectively washing out the ridge. so I’d first check with diffuse(spec) only to see if the displacement and bump detail looks right to you - conform to what you had in the sculpt. exaggerating the bump/displace might help a bit but probably not enough. You’ll really distort the object a lot while the fine detail will still be washed out by scattering. mixing the bump and displacement maps with the scattering color and AO or cavity map can help to bring back some of that surface detail. don’t forget a spec/reflectivity map as well but that’s regardless of scattering. if you don’t have one, my standin solution is use the diffuse inverted to drive spec/reflectivity as well as shinyness/gloss (with proper change range nodes) - so dark diffuse results in a stronger but concentrated highlight and bright diffuse in a less intense but wider highlight. this works rather well and more often than not, it’s like this in reality. it certainly beats not texturing the spec/ref. ah, it’s the classic “it looks great in zbrush, why don’t I get the same in the render?” and there’s no easy solution to that. Zbrush viewport makes detail in the sculpt stand out very well and it takes work to bring that back in the render. From: Szabolcs Matefy Sent: Thursday, January 09, 2014 12:12 PM To: softimage@listproc.autodesk.com Subject: RE: Rendering ZBrush displacement in Soft Actually finally I managed to work with displacement. However, displacement is not as detailed as I’d like, but boosting with a bump map, it looks fine. Unfortunately, the midpoint seems to be off, I have to somehow tune it. In ZB I set midpoint to 0.5, because if I set it to 0, It looked as if I have no recesses on the skin. Now what I think is that I have to exaggerate the details to make it work properly with skin shading, but that’s another story. It looks like that the details are in the texture, but somehow the model doesn’t want to reflect it, maybe I should pump up the subdivision in the displacement tab of geoapprox PPG. Cheers Szabolcs PS. For me, after GoZ is used then I have two issues: 1) I can’t Alt-Tab to change between tasks, I have to minimize ZBrush, 2) All playback function, simulation is not available anymore in Softimage Cheers From: softimage-boun...@listproc.autodesk.com [mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Emilio Hernandez Sent: Wednesday, January 08, 2014 4:58 PM To: softimage@listproc.autodesk.com Subject: Re: Rendering ZBrush displacement in Soft I also use GoZ a lot, the only bug that happens for me is that if you send from Softimage to Zbrush, make an UV check before start sculpting. Sometimes they got screwed. So export an OBJ and import in Zbrush do your sculpt and sendback to SI. Make sure your UV's are consistent and flip the U before exporting from Zbrush or your map is going to come up flipped. upside down. Regularly I don't subd the mesh in Zbrush more than 4 subd. As you need to subd your mesh also inside SI the same amount to properly displace the geo when rendering. If you find that you need more subd in your sculpt. I willo go back, subd the mesh to get more polys and send it back to Zbrush, so you can go as high as 8 subd from the original mesh, that is a lot. As far as I remember you can use a 32bit depth image to plug that into the scalar change range node. But this depth in the bitmap has nothing to do with the linear workflow, as the linear workflow is related to gamma correct display and rendering, and this maps are to drive values of units to displace the geometry inwards and outwards and to be interpretated by the render engine Here is video that will help to understand the displacement in SI and MR. https://vimeo.com/29898426 . 2014/1/8 Cristobal Infante cgc...@gmail.com You read it linear for sure. What exactly is your problem, are you not getting enough detail?. If this is the problem it can also be the UVs are not big enough for each poly. To be honest the final shader doesn't really matter for the disp, in fact GoZ export with a phong. I personally only use Architectural materials when in MR though. On 8 January 2014 12:45, Szabolcs Matefy szabol...@crytek.com wrote: OK, next question. If you are working in linear workflow, how would you set up the
Re: rigging in xsi vs maya
Maybe we could rename constraints with ICE? Eat it Maya! On Wed, Jan 8, 2014 at 9:57 PM, Matt Lind ml...@carbinestudios.com wrote: Butbut.buteverybody said ICE can do oh so much more. Say it ain't so.
Re: Redshift3D Render
We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again, it's a beta and they're continuing to improve things like the proxy system. Once I'm caught up I'm hoping to try rendering the classroom scene and see how it does. -Paul ᐧ -- Signature
Re: [ICE] B-Spline/NURBS curve compound?
Thanks Daniel, I'll have a look. -- Martin Chatterjee [ Freelance Technical Director ] [ http://www.chatterjee.de ] [ https://vimeo.com/chatterjee ] On Thu, Jan 9, 2014 at 2:48 PM, Daniel Brassard dbrassar...@gmail.comwrote: I did some experiment in ICE with Cubic B Spline. You can see the result on si-community, there even some compound you can download and experiment with. http://www.si-community.com/community/viewtopic.php?f=15t=1802 Hope that help. Daniel On Thu, Jan 9, 2014 at 5:38 AM, Martin Chatterjee martin.chatterjee.li...@googlemail.com wrote: Cheers Vladimir, that definitely helps! Can't believe I actually overlooked this compound in my search... :) Martin -- Martin Chatterjee [ Freelance Technical Director ] [ http://www.chatterjee.de ] [ https://vimeo.com/chatterjee ] On Thu, Jan 9, 2014 at 11:23 AM, Vladimir Jankijevic vladi...@elefantstudios.ch wrote: there is the 'Piecewise Cubic B Spline' Compound you could use as a starting point. Cheers, Vladimir On Thu, Jan 9, 2014 at 11:21 AM, Martin Chatterjee martin.chatterjee.li...@googlemail.com wrote: Hey there, is anybody aware of an ICE compound solving B-Spline/NURBS curves? You know, like the existing 'Bezier 4' and 'Bezier 5' compounds but for a curve with an arbitrary number of control points? Thanks in advance, cheers, Martin -- Martin Chatterjee [ Freelance Technical Director ] [ http://www.chatterjee.de ] [ https://vimeo.com/chatterjee ]
Re: rigging in xsi vs maya
I've been doing quite a bit of rigging in Modo lately, and I have been very surprised by its capabilities. One thing they do support is heat mapping. It's quite nice to use, but there are several requirement that need to be met for a mesh to be acceptable for heat binding. I don't know if all heat mapping implementations are based on the same algo(s), and therefore, inherit the same requirements, but here they go (copying/pasting from the docs): --Mesh must form a volume, though holes are supported (such as eye sockets). --Target mesh must be only polygonal, no single vertices, floating edges or curves can be present. --No shared vertices, edges or polygons (non-manifold surfaces) allowed between multiple components. --All joints must be contained within the volume of the mesh. Otherwise, you can still use the available smooth or rigid binding methods. I don't know if any problems you ran into could be due to some of these conditions, but there... just in case. On 08/01/2014 8:31 AM, Sebastien Sterling wrote: One feature i would have loved to see implemented across the board of autodesk products (apart from Alembic which should really just be a new standard by now...) is the heat map algorithm. in theory, is this that difficult to implement in Soft and Max ? apparently it was made by a bunch of students checking up on heat distribution algorithm papers for designing old radiators. http://www.youtube.com/watch?v=aCBx8MjEvvo On paper it looks like the best shit ever, so we of the CHR Dep wanted to use it to test characters for deformation in maya pre rigging. trouble was, apparently its extremely susceptible, and i'm not quite sure to what, topology, mesh density... but in any case a Lead at rigging scripted a small ui allowing us to just bypass most of the checks, making the tech actually usable, and it worked great... until we realised that it actually pops vertices slightly away from their initial position... in fairness we used a script to access these capabilities so maybe that caused the problem, i doubt it but there was tampering, maybe someone else has had more controled experiences with Heat mapping, like i said before it still seems like a really useful addition, On 8 January 2014 10:52, Tim Leydecker bauero...@gmx.de wrote: Using a 3DSMax rigged sample character scene from the UDK docs, I made a roundtrip through Maya and Softimage using the *.fbx format. I didnt try to export any rig controls, just a "human" rig. Its worth checking to have the latest *.fbx version installed and using an export preset that seems applicable, I think I resorted to "Autodesk Media Entertainment 2012 bla" (im on 2012s). I cant say if that was the best way but that roundtrip worked. I ended up with Maya/3DSMax/Softimage each having the rigged, animated character in a scene. In my case, there was some nuisance with the BIPED rig getting interpreted as a second rig the character is rigged to in Softimage, I had to delete that biped in XSI to get back to similar results as in 3DSMax, leaving only the rig meant for export - it is likely that was my export settings or selection settings. I had straight results going from Maya to Softimage. Cheers, tim On 07.01.2014 23:58, Steven Caron wrote: this thread is some what well timed... i am in maya right now. i need to get a mesh and its skin/envelope into softimage. i did not rig this object and i don't know enough about maya to try and understand it through inspection. in softimage i would select the mesh, then select the deformers from envelope, then key frame those objects and remove the constraints on them in mass with 'remove all constraints' is NONE of that doable in maya? cause i am having a hell of a time figuring it out. s
Re: rigging in xsi vs maya
I posted this on the Softimage User Voice but I really really want to try this Geodesic Voxel Binding: https://vimeo.com/69268846 On Thursday, January 09, 2014 10:34:36 AM, Sergio Mucino wrote: I've been doing quite a bit of rigging in Modo lately, and I have been very surprised by its capabilities. One thing they do support is heat mapping. It's quite nice to use, but there are several requirement that need to be met for a mesh to be acceptable for heat binding. I don't know if all heat mapping implementations are based on the same algo(s), and therefore, inherit the same requirements, but here they go (copying/pasting from the docs): /--//Mesh must form a volume, though holes are supported (such as eye sockets).// --//Target mesh must be //only//polygonal, no single vertices, floating edges or curves can be present.// --//No shared vertices, edges or polygons (non-manifold surfaces) allowed between multiple components. // --//All joints must be contained within the volume of the mesh. / Otherwise, you can still use the available smooth or rigid binding methods. I don't know if any problems you ran into could be due to some of these conditions, but there... just in case. On 08/01/2014 8:31 AM, Sebastien Sterling wrote: One feature i would have loved to see implemented across the board of autodesk products (apart from Alembic which should really just be a new standard by now...) is the heat map algorithm. in theory, is this that difficult to implement in Soft and Max ? apparently it was made by a bunch of students checking up on heat distribution algorithm papers for designing old radiators. http://www.youtube.com/watch?v=aCBx8MjEvvo On paper it looks like the best shit ever, so we of the CHR Dep wanted to use it to test characters for deformation in maya pre rigging. trouble was, apparently its extremely susceptible, and i'm not quite sure to what, topology, mesh density... but in any case a Lead at rigging scripted a small ui allowing us to just bypass most of the checks, making the tech actually usable, and it worked great... until we realised that it actually pops vertices slightly away from their initial position... in fairness we used a script to access these capabilities so maybe that caused the problem, i doubt it but there was tampering, maybe someone else has had more controled experiences with Heat mapping, like i said before it still seems like a really useful addition, On 8 January 2014 10:52, Tim Leydecker bauero...@gmx.de mailto:bauero...@gmx.de wrote: Using a 3DSMax rigged sample character scene from the UDK docs, I made a roundtrip through Maya and Softimage using the *.fbx format. I didn´t try to export any rig controls, just a human rig. It´s worth checking to have the latest *.fbx version installed and using an export preset that seems applicable, I think I resorted to Autodesk Media Entertainment 2012 bla (im on 2012´s). I can´t say if that was the best way but that roundtrip worked. I ended up with Maya/3DSMax/Softimage each having the rigged, animated character in a scene. In my case, there was some nuisance with the BIPED rig getting interpreted as a second rig the character is rigged to in Softimage, I had to delete that biped in XSI to get back to similar results as in 3DSMax, leaving only the rig meant for export - it is likely that was my export settings or selection settings. I had straight results going from Maya to Softimage. Cheers, tim On 07.01.2014 23:58, Steven Caron wrote: this thread is some what well timed... i am in maya right now. i need to get a mesh and its skin/envelope into softimage. i did not rig this object and i don't know enough about maya to try and understand it through inspection. in softimage i would select the mesh, then select the deformers from envelope, then key frame those objects and remove the constraints on them in mass with 'remove all constraints' is NONE of that doable in maya? cause i am having a hell of a time figuring it out. s --
Re: Redshift3D Render
Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.comwrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again, it's a beta and they're continuing to improve things like the proxy system. Once I'm caught up I'm hoping to try rendering the classroom scene and see how it does. -Paul ᐧ --
Re: rigging in xsi vs maya
I saw that video a while ago. I would expect to see this show up in Maya sometime 'soon' (hopefully).' On 09/01/2014 10:38 AM, Eric Thivierge wrote: I posted this on the Softimage User Voice but I really really want to try this Geodesic Voxel Binding: https://vimeo.com/69268846 On Thursday, January 09, 2014 10:34:36 AM, Sergio Mucino wrote: I've been doing quite a bit of rigging in Modo lately, and I have been very surprised by its capabilities. One thing they do support is heat mapping. It's quite nice to use, but there are several requirement that need to be met for a mesh to be acceptable for heat binding. I don't know if all heat mapping implementations are based on the same algo(s), and therefore, inherit the same requirements, but here they go (copying/pasting from the docs): /--//Mesh must form a volume, though holes are supported (such as eye sockets).// --//Target mesh must be //only//polygonal, no single vertices, floating edges or curves can be present.// --//No shared vertices, edges or polygons (non-manifold surfaces) allowed between multiple components. // --//All joints must be contained within the volume of the mesh. / Otherwise, you can still use the available smooth or rigid binding methods. I don't know if any problems you ran into could be due to some of these conditions, but there... just in case. On 08/01/2014 8:31 AM, Sebastien Sterling wrote: One feature i would have loved to see implemented across the board of autodesk products (apart from Alembic which should really just be a new standard by now...) is the heat map algorithm. in theory, is this that difficult to implement in Soft and Max ? apparently it was made by a bunch of students checking up on heat distribution algorithm papers for designing old radiators. http://www.youtube.com/watch?v=aCBx8MjEvvo On paper it looks like the best shit ever, so we of the CHR Dep wanted to use it to test characters for deformation in maya pre rigging. trouble was, apparently its extremely susceptible, and i'm not quite sure to what, topology, mesh density... but in any case a Lead at rigging scripted a small ui allowing us to just bypass most of the checks, making the tech actually usable, and it worked great... until we realised that it actually pops vertices slightly away from their initial position... in fairness we used a script to access these capabilities so maybe that caused the problem, i doubt it but there was tampering, maybe someone else has had more controled experiences with Heat mapping, like i said before it still seems like a really useful addition, On 8 January 2014 10:52, Tim Leydecker bauero...@gmx.de mailto:bauero...@gmx.de wrote: Using a 3DSMax rigged sample character scene from the UDK docs, I made a roundtrip through Maya and Softimage using the *.fbx format. I didn´t try to export any rig controls, just a "human" rig. It´s worth checking to have the latest *.fbx version installed and using an export preset that seems applicable, I think I resorted to "Autodesk Media Entertainment 2012 bla" (im on 2012´s). I can´t say if that was the best way but that roundtrip worked. I ended up with Maya/3DSMax/Softimage each having the rigged, animated character in a scene. In my case, there was some nuisance with the BIPED rig getting interpreted as a second rig the character is rigged to in Softimage, I had to delete that
Re: rigging in xsi vs maya
This has been pretty much my only "um..." regarding ICE. It seems to be like a (powerful) local black box that is related to one object. I know that an ICE graph can actually get and set data to multiple locations, but in some cases, one needs to jump through hoops (for example, it's difficult to read-write data from other ICE graphs... or at least, not straight-forward). In Maya, everything is part of the scene graph, so its a lot easier to read/write data, and find all related operations to a certain node. However, Maya has to have the worst node editor I've ever had to touch. I would definitely not want to see something like that in Softimage (or anywhere else for that matter). Every time I try to use it, it makes me want to kick puppies, and come back flying to the Hypergraph. I much prefer the ICE UI/workflow (I'd just like it more if it was "global") and Modo's Schematic View (by orders of magnitude). On 08/01/2014 5:00 PM, Eric Thivierge wrote: Yeah, ICE could do that if they keep pushing it... maybe? Though I think it's pretty black boxed in terms of just having the high level access to objects, not the underlying nodes. A Node Editor like Maya plus exposing more of the internals in the Scene Explorer would be something to look at if this ever gets any attention. @Emilio, we need this in Softimage as well! On Wednesday, January 08, 2014 4:58:03 PM, Emilio Hernandez wrote: Haha. Maybe because Maya needs it, so you can dig in there and get it working properly. While in Softimage not ;) Just fueling the fire! 2014/1/8 Eric Thivierge ethivie...@hybride.com mailto:ethivie...@hybride.com Just because I want to fuel the fire, I'll toss in that while the workflow in Maya is quite flawed out of the box, you can get to more internals of the scene graph and manipulate it than we have in Softimage. On Wednesday, January 08, 2014 4:15:04 PM, Alan Fregtman wrote: Bravo! Bravo!! :) I echo your exact sentiments, David (though my own credentials are puny by comparison.) The operator stack should be permanently on the box as a "hot feature". We all take it for granted all the time, but seriously it's one of the best features in Soft. On Wed, Jan 8, 2014 at 3:10 PM, Steven Caron car...@gmail.com mailto:car...@gmail.com mailto:car...@gmail.com mailto:car...@gmail.com wrote: thank you! thank you! thank you!... i knew i wasn't crazy thinking rigging in maya is a PITA! On Wed, Jan 8, 2014 at 11:45 AM, David Gallagher davegsoftimagel...@gmail.com mailto:davegsoftimagel...@gmail.com mailto:davegsoftimagelist@__gmail.com mailto:davegsoftimagel...@gmail.com wrote: I rigged on quite a few characters in Maya at Blue Sky Studios and now (Softimage) AnimSchool. We offer the well-known "Malcolm" rig for free. There is no comparison to rigging in Softimage and Maya--not the kind of rigging I do. I often assume by now they have better workflows in Maya, but I'm often surprised to find how convoluted and limiting the workflows are to this day. Most Maya people must not know there are better ways of working or aren't doing the kinds of things I am, because the difference is profound. -At any point in the rigging process, you can make edits in the model stack to change the shape and topology of
Re: Redshift3D Render
Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again, it's a beta and they're continuing to improve things like the proxy system. Once I'm caught up I'm hoping to try rendering the classroom scene and see how it does. -Paul ᐧ --
Re: Rendering ZBrush displacement in Soft
Something that helps a lot is using a normal map. To get the fine details without pumping up additional geometry. Get a normal map out of Zbrush the same way you do the displacemnt. Add a Binormal map to the object in softimage and use it to drive bump. This will add the extra fine details in your render. Cheers. 2014/1/9 pete...@skynet.be yeah, about fine detail and skin shading, subsurface scattering does tend to cover up fine surface detail – recesses and wrinkles in the skin in particular. just think of it: a wrinkle on top of the skin is like a fine ridge. without scattering, if the light comes from the left, the left of that ridge would be bright and the right in shadow. but the scattering makes the shadow part light up – and if your SSS shader compensates the diffuse (which it does by default) the left part will be a bit less bright than it would be without scattering. so as a net result this counters the diffuse light, effectively washing out the ridge. so I’d first check with diffuse(spec) only to see if the displacement and bump detail looks right to you - conform to what you had in the sculpt. exaggerating the bump/displace might help a bit but probably not enough. You’ll really distort the object a lot while the fine detail will still be washed out by scattering. mixing the bump and displacement maps with the scattering color and AO or cavity map can help to bring back some of that surface detail. don’t forget a spec/reflectivity map as well but that’s regardless of scattering. if you don’t have one, my standin solution is use the diffuse inverted to drive spec/reflectivity as well as shinyness/gloss (with proper change range nodes) - so dark diffuse results in a stronger but concentrated highlight and bright diffuse in a less intense but wider highlight. this works rather well and more often than not, it’s like this in reality. it certainly beats not texturing the spec/ref. ah, it’s the classic “it looks great in zbrush, why don’t I get the same in the render?” and there’s no easy solution to that. Zbrush viewport makes detail in the sculpt stand out very well and it takes work to bring that back in the render. *From:* Szabolcs Matefy szabol...@crytek.com *Sent:* Thursday, January 09, 2014 12:12 PM *To:* softimage@listproc.autodesk.com *Subject:* RE: Rendering ZBrush displacement in Soft Actually finally I managed to work with displacement. However, displacement is not as detailed as I’d like, but boosting with a bump map, it looks fine. Unfortunately, the midpoint seems to be off, I have to somehow tune it. In ZB I set midpoint to 0.5, because if I set it to 0, It looked as if I have no recesses on the skin. Now what I think is that I have to exaggerate the details to make it work properly with skin shading, but that’s another story. It looks like that the details are in the texture, but somehow the model doesn’t want to reflect it, maybe I should pump up the subdivision in the displacement tab of geoapprox PPG. Cheers Szabolcs PS. For me, after GoZ is used then I have two issues: 1) I can’t Alt-Tab to change between tasks, I have to minimize ZBrush, 2) All playback function, simulation is not available anymore in Softimage Cheers *From:* softimage-boun...@listproc.autodesk.com [mailto: softimage-boun...@listproc.autodesk.com] *On Behalf Of *Emilio Hernandez *Sent:* Wednesday, January 08, 2014 4:58 PM *To:* softimage@listproc.autodesk.com *Subject:* Re: Rendering ZBrush displacement in Soft I also use GoZ a lot, the only bug that happens for me is that if you send from Softimage to Zbrush, make an UV check before start sculpting. Sometimes they got screwed. So export an OBJ and import in Zbrush do your sculpt and sendback to SI. Make sure your UV's are consistent and flip the U before exporting from Zbrush or your map is going to come up flipped. upside down. Regularly I don't subd the mesh in Zbrush more than 4 subd. As you need to subd your mesh also inside SI the same amount to properly displace the geo when rendering. If you find that you need more subd in your sculpt. I willo go back, subd the mesh to get more polys and send it back to Zbrush, so you can go as high as 8 subd from the original mesh, that is a lot. As far as I remember you can use a 32bit depth image to plug that into the scalar change range node. But this depth in the bitmap has nothing to do with the linear workflow, as the linear workflow is related to gamma correct display and rendering, and this maps are to drive values of units to displace the geometry inwards and outwards and to be interpretated by the render engine Here is video that will help to understand the displacement in SI and MR. https://vimeo.com/29898426 . 2014/1/8 Cristobal Infante cgc...@gmail.com You read it linear for sure. What exactly is your problem, are you not getting enough detail?. If this is
Re: rigging in xsi vs maya
I absolutely hate this behavior in Maya. It's, frankly, ridiculous. Maya's weighting tools are totally sub-par compared to any other 3d application I've used (including Max). Why it is this way, I don't know, but as a user, it's incredibly frustrating to have to focus on not shooting yourself in the foot (as daring to perform a smooth weights operation with all bones unlocked) more than getting actual work done. Maya has great things for it, but binding and weighting is definitely not one of them. It's pretty bad, actually. Ok, rant off. :-) On 07/01/2014 9:57 PM, Sebastien Sterling wrote: I was quite shocked to learn from riggers in my last job, that in maya you have to "lock all bones but the ones you want to weight to via small tick boxes" failure to do so aparently causing maya to through random influences around... On 8 January 2014 02:22, Alan Fregtman alan.fregt...@gmail.com wrote: Last time I had to use Maya I would use Crosswalk to transfer the skinned mesh from Maya to Soft, do my weighting in home sweet home, then I wrote an exporter that saved out my weights in the "cometSaveWeights" format. Life saver! On Tue, Jan 7, 2014 at 6:15 PM, Steven Caron car...@gmail.com wrote: arg, figured it out. import pymel.core as pm pm.select(pm.skinCluster(pm.selected()[0], query=True, influence=True)) best UI ever! On Tue, Jan 7, 2014 at 2:58 PM, Steven Caron car...@gmail.com wrote: this thread is some what well timed... i am in maya right now. i need to get a mesh and its skin/envelope into softimage. i did not rig this object and i don't know enough about maya to try and understand it through inspection. in softimage i would select the mesh, then select the deformers from envelope, then key frame those objects and remove the constraints on them in mass with 'remove all constraints' is NONE of that doable in maya? cause i am having a hell of a time figuring it out. s --
Re: Redshift3D Render
I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.com mailto:emi...@e-roja.com wrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com mailto:mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure :) But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com mailto:tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were
Re: Redshift3D Render
As far as I remember from watching the specs in order to high a better rate the magic number is the memory speed transfer. I was going to buy a 680 but then realized that my 470 has more speed in the memory transfer. Don't know why but Nvidia choked the speed in most of the 600 series except for the Titan. And open again the hoose in the 690. In my experience and from what we have discussing in the Redshift forums the two magic numbers to see which GPU will perform faster is not only the Fill triangles per sec number but also the memory transfer speed. Below are the specs of the Titan, the 680 and the 470 I have. *GTX TITAN GPU Engine Specs:* 2688CUDA Cores 837Base Clock (MHz) 876Boost Clock (MHz) 187.5Texture Fill Rate (billion/sec) *GTX TITAN Memory Specs:* 6.0 GbpsMemory Clock 6144 MBStandard Memory Config GDDR5Memory Interface 384-bit GDDR5Memory Interface Width 288.4Memory Bandwidth (GB/sec) *GTX 680 GPU Engine Specs:* CUDA Cores 1006Base Clock (MHz) 1058Boost Clock (MHz) 128.8Texture Fill Rate (billion/sec) *GTX 680 Memory Specs:* 6.0 GbpsMemory Speed 2048MBStandard Memory Config 256-bit GDDR5Memory Interface Width 192.2Memory Bandwidth (GB/sec) GTX 470 GPU Engine Specs CUDA Cores 448 Graphics Clock (MHz) 607 MHz Processor Clock (MHz) 1215 MHz Texture Fill Rate (billion/sec) 34.0 Memory Specs Memory Clock 1674 MHz (3348 data rate) Standard Memory Config 1280 MB Memory Interface GDDR5 Memory Interface Width 320-bit Memory Bandwidth (GB/sec) 133.9 GB/sec So as you can see my 470 MB (Memory Bandwith) is not that far from the 680 but the Memory Clock speed, and the Processor Clock speed, of the 470 is higher than the 680 Not a guru here but that can be an explanation of why the 470 has faster fps than the 680... For rendering you make sure you compare the Memory Bandwith of the GPU. Beside the CUDA. Double the CUDAs but half the Memory Bandwith, and I cannot assure but it will render almost the same. 2014/1/9 Leonard Koch leonardkoch...@gmail.com I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it
Re: Redshift3D Render
Yes Mirko tell the secret. I don't want to break my mind thinking about memory clocks and bandwiths 2014/1/9 Tim Crowson tim.crow...@magneticdreams.com I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again, it's a beta and they're continuing to improve things like the proxy system. Once I'm caught up I'm hoping to try rendering the classroom scene and see how it does. -Paul ᐧ -- --
Re: Redshift3D Render
Ah interesting. That begins to explain it. Thanks Emilio. On Thu, Jan 9, 2014 at 5:36 PM, Emilio Hernandez emi...@e-roja.com wrote: Yes Mirko tell the secret. I don't want to break my mind thinking about memory clocks and bandwiths 2014/1/9 Tim Crowson tim.crow...@magneticdreams.com I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again, it's a beta and they're continuing to improve things like the proxy system. Once I'm caught up I'm hoping to try rendering the classroom scene and see how it does. -Paul ᐧ -- --
Re: Redshift3D Render
For GPU speeds, you always need to consult this list, it is pretty representative of what to expect from things like Redshift3D: http://www.videocardbenchmark.net/high_end_gpus.html -ben On Thu, Jan 9, 2014 at 11:36 AM, Emilio Hernandez emi...@e-roja.com wrote: Yes Mirko tell the secret. I don't want to break my mind thinking about memory clocks and bandwiths 2014/1/9 Tim Crowson tim.crow...@magneticdreams.com I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again, it's a beta and they're continuing to improve things like the proxy system. Once I'm caught up I'm hoping to try rendering the classroom scene and see how it does. -Paul ᐧ -- -- -- Best regards, Ben Houston Voice: 613-762-4113 Skype: ben.exocortex Twitter: @exocortexcom
Re: Redshift3D Render
Just a funny fact that I found... After opening several tabs for looking for the Nvidia specs, I left them open. I am using FireFox. I hit play again and the fps dropped by half 24 fps. Start scratching my head. Closed the addtional tabs of Firefox, restarted Softimage and the speed was back to 50.7 So my guess is that this modern browser suck a lot from the GPU... pfff. 2014/1/9 Leonard Koch leonardkoch...@gmail.com Ah interesting. That begins to explain it. Thanks Emilio. On Thu, Jan 9, 2014 at 5:36 PM, Emilio Hernandez emi...@e-roja.comwrote: Yes Mirko tell the secret. I don't want to break my mind thinking about memory clocks and bandwiths 2014/1/9 Tim Crowson tim.crow...@magneticdreams.com I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again,
Re: Redshift3D Render
Title: Signature
I guess 60fps is the refresh rate of your display, right?
Have you disabled VSync in the driver settings?
I just get "60.0 fps +"
How are you getting it display a value higher than 60? I'm pretty
sure it the actual fps is higher, but the value in the viewport is
capped at 60
-Tim
On 1/9/2014 10:12 AM, Leonard Koch
wrote:
I get about 28-31 out of my 680. Does anyone have a
common explanation for that?
On Thu, Jan 9, 2014 at 5:10 PM, Emilio
Hernandez emi...@e-roja.com
wrote:
Hey Mirko I ran your script and I got 50.7 fps...
But then I remembered I have my displays plugged in to
my 470.. hahaha.
Don't ask why, but when using AE with the displays
plugged into the Ti, AE does not like it and disables
GPU for calculations...
P.
2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com
Hey Tim
Would you be able to take 2 minutes of your
tmie and run this ol python script for SI with
your titan?
I'm getting weird results with an 780 in my
home system outperforming titan a lot... well
here is copy paste from forum if you are able
to check it out as well.. thanks!:
itan:
~170 fps
780:
~245 fps
Go
figure
But
I'm suspecting something weird with my titan
system for some time will have to test
further but would be great if anyone with
titan as well could run it too?
This
old python script:
Application.CreatePrim("Cube",
"MeshSurface", "", "")
Application.SetValue("cube.polymsh.geom.subdivu",
831, "")
Application.SetValue("cube.polymsh.geom.subdivv",
800, "")
Application.SetValue("cube.polymsh.geom.subdivbase",
800, "")
Application.SetValue("Camera.camvis.refreshrate",
True, "")
Application.SetDisplayMode("Camera",
"shaded")
Application.DeselectAll()
Application.SetValue("PlayControl.Out",
5000, "")
Application.DeselectAll()
Application.GetPrim("Null",
"", "", "")
Application.SelectObj("Camera_Root",
"", "")
Application.CopyPaste("Camera_Root",
"", "null", 1)
Application.SelectObj("null",
"", "")
Application.SaveKey("null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz",
1, "", "", "", "", "")
Application.SetValue("PlayControl.Key",
5000, "")
Application.SetValue("PlayControl.Current",
5000, "")
Application.Rotate("",
0, 8000, 0, "siAbsolute", "siPivot",
"siObj", "siY", "", "", "", "", "", "", "",
0, "")
Application.SaveKey("null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz",
5000, "", "", "", "", "")
Application.FirstFrame()
Just paste in python script run and hit
play.
Thakns!
On Thu,
Re: Redshift3D Render
So does Chrome btw. I also notice this wehn running my laptop without power supply. The GPU sucks the battery dry in an hour.With Chrome closed it's 4 hrs. GPU acceleration can be turned off in Chrome though, don't know about Firefox.Just a funny fact that I found...After opening several tabs for looking for the Nvidia specs, I left them open. I am using FireFox.I hit play again and the fps dropped by half 24 fps.
Start scratching my head. Closed the addtional tabs of Firefox, restarted Softimage and the speed was back to 50.7So my guess is that this modern browser suck a lot from the GPU... pfff.
2014/1/9 Leonard Koch leonardkoch...@gmail.com
Ah interesting. That begins to explain it.Thanks Emilio.On Thu, Jan 9, 2014 at 5:36 PM, Emilio Hernandez emi...@e-roja.com wrote:
Yes Mirko tell the secret. I don't want to break my mind thinking about memory clocks and bandwiths
2014/1/9 Tim Crowson tim.crow...@magneticdreams.com
I just get "60.0 fps +"
How are you getting it display a value higher than 60? I'm pretty
sure it the actual fps is higher, but the value in the viewport is
capped at 60
-Tim
On 1/9/2014 10:12 AM, Leonard Koch
wrote:
I get about 28-31 out of my 680. Does anyone have a
common explanation for that?
On Thu, Jan 9, 2014 at 5:10 PM, Emilio
Hernandez emi...@e-roja.com
wrote:
Hey Mirko I ran your script and I got 50.7 fps...
But then I remembered I have my displays plugged in to
my 470.. hahaha.
Don't ask why, but when using AE with the displays
plugged into the Ti, AE does not like it and disables
GPU for calculations...
P.
2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com
Hey Tim
Would you be able to take 2 minutes of your
tmie and run this ol python script for SI with
your titan?
I'm getting weird results with an 780 in my
home system outperforming titan a lot... well
here is copy paste from forum if you are able
to check it out as well.. thanks!:
itan:
~170 fps
780:
~245 fps
Go
figure
But
I'm suspecting something weird with my titan
system for some time will have to test
further but would be great if anyone with
titan as well could run it too?
This
old python script:
Application.CreatePrim("Cube",
"MeshSurface", "", "")
Application.SetValue("cube.polymsh.geom.subdivu",
831, "")
Application.SetValue("cube.polymsh.geom.subdivv",
800, "")
Application.SetValue("cube.polymsh.geom.subdivbase",
800, "")
Application.SetValue("Camera.camvis.refreshrate",
True, "")
Application.SetDisplayMode("Camera",
"shaded")
Application.DeselectAll()
Application.SetValue("PlayControl.Out",
5000, "")
Application.DeselectAll()
Application.GetPrim("Null",
"", "", "")
Application.SelectObj("Camera_Root",
"", "")
Application.CopyPaste("Camera_Root",
"", "null", 1)
Application.SelectObj("null",
"", "")
Application.SaveKey("null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz",
1, "", "", "", "", "")
Application.SetValue("PlayControl.Key",
5000, "")
Application.SetValue("PlayControl.Current",
5000, "")
Re: Redshift3D Render
Emilio, did you try telling the application (firefox, chrome) to use the cpu instead of the gpu? that happened to me too and found it usefu,l through the nvidia control panel. F. 2014/1/9 Emilio Hernandez emi...@e-roja.com Just a funny fact that I found... After opening several tabs for looking for the Nvidia specs, I left them open. I am using FireFox. I hit play again and the fps dropped by half 24 fps. Start scratching my head. Closed the addtional tabs of Firefox, restarted Softimage and the speed was back to 50.7 So my guess is that this modern browser suck a lot from the GPU... pfff. 2014/1/9 Leonard Koch leonardkoch...@gmail.com Ah interesting. That begins to explain it. Thanks Emilio. On Thu, Jan 9, 2014 at 5:36 PM, Emilio Hernandez emi...@e-roja.comwrote: Yes Mirko tell the secret. I don't want to break my mind thinking about memory clocks and bandwiths 2014/1/9 Tim Crowson tim.crow...@magneticdreams.com I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of
Re: Redshift3D Render
I am going to try your suggestions right now. That will explain a lot of things. 2014/1/9 Tim Crowson tim.crow...@magneticdreams.com I wish it was always that simple. For instance, Redshift will perform better with more vram, and the Titan comes standard with 6GB, which is not even an option on the 780 or 780Ti. If you can get by with less vram though, the 780s are pretty sweet. Can't wait to see what they announce next. -Tim On 1/9/2014 10:39 AM, Ben Houston wrote: For GPU speeds, you always need to consult this list, it is pretty representative of what to expect from things like Redshift3D: http://www.videocardbenchmark.net/high_end_gpus.html -ben On Thu, Jan 9, 2014 at 11:36 AM, Emilio Hernandez emi...@e-roja.comwrote: Yes Mirko tell the secret. I don't want to break my mind thinking about memory clocks and bandwiths 2014/1/9 Tim Crowson tim.crow...@magneticdreams.com I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them.
Re: Redshift3D Render
I wish it was always that simple. For instance, Redshift will perform better with more vram, and the Titan comes standard with 6GB, which is not even an option on the 780 or 780Ti. If you can get by with less vram though, the 780s are pretty sweet. Can't wait to see what they announce next. -Tim On 1/9/2014 10:39 AM, Ben Houston wrote: For GPU speeds, you always need to consult this list, it is pretty representative of what to expect from things like Redshift3D: http://www.videocardbenchmark.net/high_end_gpus.html -ben On Thu, Jan 9, 2014 at 11:36 AM, Emilio Hernandez emi...@e-roja.com mailto:emi...@e-roja.com wrote: Yes Mirko tell the secret. I don't want to break my mind thinking about memory clocks and bandwiths 2014/1/9 Tim Crowson tim.crow...@magneticdreams.com mailto:tim.crow...@magneticdreams.com I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.com mailto:emi...@e-roja.com wrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com mailto:mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure :) But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com mailto:tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot,
Re: Redshift3D Render
Well switch displays to the Titan and I am getting with the Vsync option off the same as Tim 63.6fps Maybe this Rubikjancovik cube has a trick :) 2014/1/9 Emilio Hernandez emi...@e-roja.com I am going to try your suggestions right now. That will explain a lot of things. 2014/1/9 Tim Crowson tim.crow...@magneticdreams.com I wish it was always that simple. For instance, Redshift will perform better with more vram, and the Titan comes standard with 6GB, which is not even an option on the 780 or 780Ti. If you can get by with less vram though, the 780s are pretty sweet. Can't wait to see what they announce next. -Tim On 1/9/2014 10:39 AM, Ben Houston wrote: For GPU speeds, you always need to consult this list, it is pretty representative of what to expect from things like Redshift3D: http://www.videocardbenchmark.net/high_end_gpus.html -ben On Thu, Jan 9, 2014 at 11:36 AM, Emilio Hernandez emi...@e-roja.comwrote: Yes Mirko tell the secret. I don't want to break my mind thinking about memory clocks and bandwiths 2014/1/9 Tim Crowson tim.crow...@magneticdreams.com I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not
Re: Redshift3D Render
just got back , yes 60 is vsync on, turn of vsync in nvida control panel On Thu, Jan 9, 2014 at 5:52 PM, Stefan Kubicek s...@tidbit-images.com wrote: I guess 60fps is the refresh rate of your display, right? Have you disabled VSync in the driver settings? I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again, it's a beta and they're continuing to improve things like the proxy system. Once I'm caught up I'm hoping to try rendering the classroom scene and see how it does. -Paul ᐧ -- -- -- --- Stefan Kubicek --- keyvis digital imagery Alfred Feierfeilstraße 3 A-2380 Perchtoldsdorf bei Wien Phone: +43/699/12614231 www.keyvis.at ste...@keyvis.at -- This email and its
Re: Redshift3D Render
I already did that and still getting the 65 fps limit with the Titan. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com just got back , yes 60 is vsync on, turn of vsync in nvida control panel On Thu, Jan 9, 2014 at 5:52 PM, Stefan Kubicek s...@tidbit-images.comwrote: I guess 60fps is the refresh rate of your display, right? Have you disabled VSync in the driver settings? I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again, it's a beta and they're continuing to improve things like the proxy system. Once I'm caught up I'm hoping to try rendering the classroom scene and see how it does. -Paul ᐧ -- -- -- --- Stefan Kubicek --- keyvis digital imagery Alfred
Re: Redshift3D Render
hmm another gues.. is it set to RT or ALL in play? On Thu, Jan 9, 2014 at 6:53 PM, Emilio Hernandez emi...@e-roja.com wrote: I already did that and still getting the 65 fps limit with the Titan. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com just got back , yes 60 is vsync on, turn of vsync in nvida control panel On Thu, Jan 9, 2014 at 5:52 PM, Stefan Kubicek s...@tidbit-images.comwrote: I guess 60fps is the refresh rate of your display, right? Have you disabled VSync in the driver settings? I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again, it's a beta and they're continuing to improve things like the proxy system. Once I'm caught up I'm hoping to try rendering the classroom scene and see how it does. -Paul ᐧ -- -- --
Re: Redshift3D Render
It is set to ALL. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com hmm another gues.. is it set to RT or ALL in play? On Thu, Jan 9, 2014 at 6:53 PM, Emilio Hernandez emi...@e-roja.comwrote: I already did that and still getting the 65 fps limit with the Titan. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com just got back , yes 60 is vsync on, turn of vsync in nvida control panel On Thu, Jan 9, 2014 at 5:52 PM, Stefan Kubicek s...@tidbit-images.comwrote: I guess 60fps is the refresh rate of your display, right? Have you disabled VSync in the driver settings? I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per frame - so on a frame that took 28 seconds to render, 20 seconds was spent exporting the mesh and 8 seconds were spent on rendering. But again, it's a beta and they're continuing to improve things like the proxy system. Once I'm caught up I'm hoping to try
Re: Redshift3D Render
yea I guess that woul dbe the case but just tried.. strange really no idea.. any chance to run cinebench 15 opengl test then? if that gives too low result as well then something is not good On Thu, Jan 9, 2014 at 6:58 PM, Emilio Hernandez emi...@e-roja.com wrote: It is set to ALL. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com hmm another gues.. is it set to RT or ALL in play? On Thu, Jan 9, 2014 at 6:53 PM, Emilio Hernandez emi...@e-roja.comwrote: I already did that and still getting the 65 fps limit with the Titan. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com just got back , yes 60 is vsync on, turn of vsync in nvida control panel On Thu, Jan 9, 2014 at 5:52 PM, Stefan Kubicek s...@tidbit-images.comwrote: I guess 60fps is the refresh rate of your display, right? Have you disabled VSync in the driver settings? I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually create proxies per object my scene had hundreds and hundreds of objects, so I didn't have time to create them. Therefore, it was creating a renderable mesh per
Re: Redshift3D Render
Yes specailly if you are getting more than the double speed. I ran the test and I got 75.78 fps Thx 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com yea I guess that woul dbe the case but just tried.. strange really no idea.. any chance to run cinebench 15 opengl test then? if that gives too low result as well then something is not good On Thu, Jan 9, 2014 at 6:58 PM, Emilio Hernandez emi...@e-roja.comwrote: It is set to ALL. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com hmm another gues.. is it set to RT or ALL in play? On Thu, Jan 9, 2014 at 6:53 PM, Emilio Hernandez emi...@e-roja.comwrote: I already did that and still getting the 65 fps limit with the Titan. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com just got back , yes 60 is vsync on, turn of vsync in nvida control panel On Thu, Jan 9, 2014 at 5:52 PM, Stefan Kubicek s...@tidbit-images.comwrote: I guess 60fps is the refresh rate of your display, right? Have you disabled VSync in the driver settings? I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the speed of the card. Also, although it doesn't load the entire scene into your card's memory, the more memory your card has, the better it is. But overall, for the type of work I'm mainly doing these days, it's extremely fast. In fact, it's so fast that I was finding the bottleneck was the time taken to export the mesh to Redshift, not rendering. Redshift has a proxy system like Vray Arnold, but you have to manually
Re: Linking a light (Sun) to a Physical sky ?
For future reference, to connect an existing sun's light direction with an existing sky shader in Redshift or Mental Ray, you use the command ApplySunDirectionOp (undocumented I believe). E.g. (in python): Application.ApplySunDirectionOp( sun.kine.global, [Passes.Default_Pass.Redshift_PhysicalSky.sun_direction.x, Passes.Default_Pass.Redshift_PhysicalSky.sun_direction.y, Passes.Default_Pass.Redshift_PhysicalSky.sun_direction.z] ) On Thu, Jan 9, 2014 at 4:39 AM, olivier jeannel olivier.jean...@noos.frwrote: N There is a Render /Edit/Create RedShift SkyShader I NEVER go through these menus.. Bangging my head on the desk Pulverized by the shame... Thank you oh grandmaster of the infinite knowledge ! Le 09/01/2014 13:14, Rob Chapman a écrit : rendereditinit physical sky ?this one constrains the light direction into the shader for you (mental Ray) maybe you could copy the expression from here? On 9 January 2014 11:54, olivier jeannel olivier.jean...@noos.fr wrote: Bump... So nobody has a method ? Le 08/01/2014 16:32, olivier jeannel a écrit : Very dumb question, I did it before, I'm sure... How do I link a light (infinite - sun) rotation to a vector (Sun direction) in a Physicla sky property page ? (So that my light is a sun...) I'm in redshift...
Re: Redshift3D Render
getting around 95 score in cinebench with titan, and again 780 in home comp getting 140 score in cinebench... wondering how can 780 crash titan so much in opengl. titan system beside 4 titans have i7 3930K, asus p9x79-e ws MBO and home comp i7 4770k on asus maximus VI hero Will test some redshift rendering later to compare single GPU 780 vs titan On Thu, Jan 9, 2014 at 7:11 PM, Emilio Hernandez emi...@e-roja.com wrote: Yes specailly if you are getting more than the double speed. I ran the test and I got 75.78 fps Thx 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com yea I guess that woul dbe the case but just tried.. strange really no idea.. any chance to run cinebench 15 opengl test then? if that gives too low result as well then something is not good On Thu, Jan 9, 2014 at 6:58 PM, Emilio Hernandez emi...@e-roja.comwrote: It is set to ALL. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com hmm another gues.. is it set to RT or ALL in play? On Thu, Jan 9, 2014 at 6:53 PM, Emilio Hernandez emi...@e-roja.comwrote: I already did that and still getting the 65 fps limit with the Titan. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com just got back , yes 60 is vsync on, turn of vsync in nvida control panel On Thu, Jan 9, 2014 at 5:52 PM, Stefan Kubicek s...@tidbit-images.comwrote: I guess 60fps is the refresh rate of your display, right? Have you disabled VSync in the driver settings? I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.comwrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11 AM, Paul Griswold wrote: There was a discussion on the RS forums about it. I don't recall the numbers, though. I don't think the speed of the PCIe slot made a huge difference. It's really all about the
Re: Redshift3D Render
Now I am starting to scratch my head to figure out the maze of Nvidia... At least my benchmark said the Titatn beated up a Quadro 4000K ha ha 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com getting around 95 score in cinebench with titan, and again 780 in home comp getting 140 score in cinebench... wondering how can 780 crash titan so much in opengl. titan system beside 4 titans have i7 3930K, asus p9x79-e ws MBO and home comp i7 4770k on asus maximus VI hero Will test some redshift rendering later to compare single GPU 780 vs titan On Thu, Jan 9, 2014 at 7:11 PM, Emilio Hernandez emi...@e-roja.comwrote: Yes specailly if you are getting more than the double speed. I ran the test and I got 75.78 fps Thx 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com yea I guess that woul dbe the case but just tried.. strange really no idea.. any chance to run cinebench 15 opengl test then? if that gives too low result as well then something is not good On Thu, Jan 9, 2014 at 6:58 PM, Emilio Hernandez emi...@e-roja.comwrote: It is set to ALL. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com hmm another gues.. is it set to RT or ALL in play? On Thu, Jan 9, 2014 at 6:53 PM, Emilio Hernandez emi...@e-roja.comwrote: I already did that and still getting the 65 fps limit with the Titan. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com just got back , yes 60 is vsync on, turn of vsync in nvida control panel On Thu, Jan 9, 2014 at 5:52 PM, Stefan Kubicek s...@tidbit-images.com wrote: I guess 60fps is the refresh rate of your display, right? Have you disabled VSync in the driver settings? I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.com wrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day, and this thing was letting me tumble around it at ~15fps in Shaded mode. That ain't shabby! -Tim On 1/9/2014 6:11
Re: Redshift3D Render
cinebench you mean? so what is score? :) On Thu, Jan 9, 2014 at 7:47 PM, Emilio Hernandez emi...@e-roja.com wrote: Now I am starting to scratch my head to figure out the maze of Nvidia... At least my benchmark said the Titatn beated up a Quadro 4000K ha ha 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com getting around 95 score in cinebench with titan, and again 780 in home comp getting 140 score in cinebench... wondering how can 780 crash titan so much in opengl. titan system beside 4 titans have i7 3930K, asus p9x79-e ws MBO and home comp i7 4770k on asus maximus VI hero Will test some redshift rendering later to compare single GPU 780 vs titan On Thu, Jan 9, 2014 at 7:11 PM, Emilio Hernandez emi...@e-roja.comwrote: Yes specailly if you are getting more than the double speed. I ran the test and I got 75.78 fps Thx 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com yea I guess that woul dbe the case but just tried.. strange really no idea.. any chance to run cinebench 15 opengl test then? if that gives too low result as well then something is not good On Thu, Jan 9, 2014 at 6:58 PM, Emilio Hernandez emi...@e-roja.comwrote: It is set to ALL. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com hmm another gues.. is it set to RT or ALL in play? On Thu, Jan 9, 2014 at 6:53 PM, Emilio Hernandez emi...@e-roja.comwrote: I already did that and still getting the 65 fps limit with the Titan. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com just got back , yes 60 is vsync on, turn of vsync in nvida control panel On Thu, Jan 9, 2014 at 5:52 PM, Stefan Kubicek s...@tidbit-images.com wrote: I guess 60fps is the refresh rate of your display, right? Have you disabled VSync in the driver settings? I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.com wrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame, hardware/software costs, rack space used, etc. But hey, Redshift aside, the Titan in my workstation is doing wonders for my viewport performance in Soft. I had a 58M, 2500-item model derived from a CAD file the other day,
Re: Redshift3D Render
Dumb question here. Where do I see the score? I only get a fps result for OpenGl and a 759cb in the CPU results in Cinebench. and I get 75.78fps at the OpenGl test. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com cinebench you mean? so what is score? :) On Thu, Jan 9, 2014 at 7:47 PM, Emilio Hernandez emi...@e-roja.comwrote: Now I am starting to scratch my head to figure out the maze of Nvidia... At least my benchmark said the Titatn beated up a Quadro 4000K ha ha 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com getting around 95 score in cinebench with titan, and again 780 in home comp getting 140 score in cinebench... wondering how can 780 crash titan so much in opengl. titan system beside 4 titans have i7 3930K, asus p9x79-e ws MBO and home comp i7 4770k on asus maximus VI hero Will test some redshift rendering later to compare single GPU 780 vs titan On Thu, Jan 9, 2014 at 7:11 PM, Emilio Hernandez emi...@e-roja.comwrote: Yes specailly if you are getting more than the double speed. I ran the test and I got 75.78 fps Thx 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com yea I guess that woul dbe the case but just tried.. strange really no idea.. any chance to run cinebench 15 opengl test then? if that gives too low result as well then something is not good On Thu, Jan 9, 2014 at 6:58 PM, Emilio Hernandez emi...@e-roja.comwrote: It is set to ALL. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com hmm another gues.. is it set to RT or ALL in play? On Thu, Jan 9, 2014 at 6:53 PM, Emilio Hernandez emi...@e-roja.comwrote: I already did that and still getting the 65 fps limit with the Titan. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com just got back , yes 60 is vsync on, turn of vsync in nvida control panel On Thu, Jan 9, 2014 at 5:52 PM, Stefan Kubicek s...@tidbit-images.com wrote: I guess 60fps is the refresh rate of your display, right? Have you disabled VSync in the driver settings? I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.com wrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they could really go before it started to cost more money than it was worth. In our case, I'm thinking 3 might be our max, based on some nerdy mathematics by one of our IT guys analyzing render times per shot, per frame,
Re: Redshift3D Render
yes fps is score. hmm maybe if I got some time I could get one of titans and test it out on maximus MBO to see if that is the diference... On Thu, Jan 9, 2014 at 8:09 PM, Emilio Hernandez emi...@e-roja.com wrote: Dumb question here. Where do I see the score? I only get a fps result for OpenGl and a 759cb in the CPU results in Cinebench. and I get 75.78fps at the OpenGl test. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com cinebench you mean? so what is score? :) On Thu, Jan 9, 2014 at 7:47 PM, Emilio Hernandez emi...@e-roja.comwrote: Now I am starting to scratch my head to figure out the maze of Nvidia... At least my benchmark said the Titatn beated up a Quadro 4000K ha ha 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com getting around 95 score in cinebench with titan, and again 780 in home comp getting 140 score in cinebench... wondering how can 780 crash titan so much in opengl. titan system beside 4 titans have i7 3930K, asus p9x79-e ws MBO and home comp i7 4770k on asus maximus VI hero Will test some redshift rendering later to compare single GPU 780 vs titan On Thu, Jan 9, 2014 at 7:11 PM, Emilio Hernandez emi...@e-roja.comwrote: Yes specailly if you are getting more than the double speed. I ran the test and I got 75.78 fps Thx 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com yea I guess that woul dbe the case but just tried.. strange really no idea.. any chance to run cinebench 15 opengl test then? if that gives too low result as well then something is not good On Thu, Jan 9, 2014 at 6:58 PM, Emilio Hernandez emi...@e-roja.comwrote: It is set to ALL. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com hmm another gues.. is it set to RT or ALL in play? On Thu, Jan 9, 2014 at 6:53 PM, Emilio Hernandez emi...@e-roja.com wrote: I already did that and still getting the 65 fps limit with the Titan. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com just got back , yes 60 is vsync on, turn of vsync in nvida control panel On Thu, Jan 9, 2014 at 5:52 PM, Stefan Kubicek s...@tidbit-images.com wrote: I guess 60fps is the refresh rate of your display, right? Have you disabled VSync in the driver settings? I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.com wrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim Crowson tim.crow...@magneticdreams.com wrote: We've been testing 1 Titan vs. 3 and so far, the speed increase of the triple-Titan box is holding at about 2.45x. In an email exchange (or maybe it was on the forums, can't recall) it was mentioned that on the topic parallelization, Pixar had determined that even for them, 4 units together (of whatever, not necessarily Titans) was the max they
Re: rigging in xsi vs maya
Maya's component editor really sucks because of that lock thing. It really slow my workflow so I usually deal with the export / import to SI and work there my weighting. But when I can, I skin my model in Maya before sending it to SI, because it has a much better default weighting than SI and Maya bones are easier to deal with when you do game assets (and FBX convert them nicely into nulls when importing). With proper settings you can use the default Maya weights for a test model where precise weights aren't needed. SI weights are pretty much useless, not even for a mob or test character. You need to re-weight almost everything manually, and if you work with nulls, it is even worst. I haven't used yet that heat skin or whatever from newer Maya versions but it looks cool. (I only use old versions for work) Martin Sent from my iPhone Maya has great things for it, but binding and weighting is definitely not one of them. It's pretty bad, actually. Ok, rant off.
RE: Batch function for Ultimapper and Render Map
You'll have to modify all references to rendermap to find/call ultimapper
instead. That includes the parameters which are adjusted dynamically such as
the output filename. You can get that information by setting your scene
explorer to show all parameters ('all nodes' filter), use script names, show
parameter values, then inspect the property in the scene explorer. The names
you see will be the names you use in code. Select/mark the parameters to see
their values. To get the commands to call, you can click the generate maps
button and see what pops out in the script log, then convert that to the
scripting object model if you want to learn.
For example, changing the output filename parameter for rendermap would dump
something like this to the script log:
SetValue( cube.rendermap.imagefilepath, c:\\tmp\\my_image.tga, null );
If you have a reference to the rendermap property in your code, the object
model equivalent would be this:
oRenderMapProperty.Parameters( imagefilepath ).value =
c:\\tmp\\my_image.tga;
Notice the parameter name is the same, but having the reference to the
rendermap property simplifies your code so you don't have to deal with absolute
paths and strings.
The generate maps button calls the RegenerateMaps() command, so no object model
equivalent for that. But since it requires a rendermap property as an input
argument, you can pass the value from the rendermap property reference than
having to deal with strings, as illustrated in my example. Compare that with
what you see in the script log for comparison.
Ultimapper will have a few extra wrinkles than rendermap, such as distance from
surface to probe with rays, and so on. It's not a difficult task, success is
purely based on whether you account for all the details.
Matt
From: softimage-boun...@listproc.autodesk.com
[mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Nicolas Esposito
Sent: Thursday, January 09, 2014 12:29 AM
To: softimage@listproc.autodesk.com
Subject: Re: Batch function for Ultimapper and Render Map
Wow Matt,
Thanks for spending time writing the script, I'll test it out together with the
Mapify plugin to see which one could suit my needs
Sorry to ask the same question as before, but based on your Jscript for
Ultimapper I'll just need to replace the function with the Ultimapper function,
add the maps variables ( normals, albedo, depth, AO ) and change the specified
ultimapper property to be what is called Ultimapper thingy at the end of the
script
I'll start experimenting with the SDK editor to see if there are some usefull
informations ;)
Thanks again Matt for the script, amazing how you guys could come out with a
solution within few minutes just writing it down! :)
2014/1/9 Matt Lind ml...@carbinestudios.commailto:ml...@carbinestudios.com
// Jscript - will need some tweaking to be functional
RenderMapSequence( 1, 100, C:\\tmp\\my_sequence.CURRENTFRAME.tga );
function RenderMapSequence( FrameStart, FrameEnd, FileName )
{
var oPlayControl = Dictionary.GetObject( PlayControl );
// get eligible objects from selection
var aFilterNames = new Array( siPolyMeshFilter, siSurfaceMeshFilter );
var oObjects = SIFilter( null, aFilterNames.join(,), true, siQuickSearch );
If ( !oObjects || oObjects.Count =0 ) {
LogMessage( nothing selected, siError );
}
for ( var i = 0; i oObjects.Count; i++ ) {
var oObject = oObjects(i);
var oProperties = oObject.Properties.Filter( rendermap );
if ( oProperties.Count = 0 ) {
// rendermap property not found
continue;
}
var oRenderMapProperty = oProperties(0);
for ( var j = FrameStart; j = FrameEnd; j++ ) {
var FrameCurrent = j;
// advance timeline
oPlayControl.Parameters( Current ).value = FrameCurrent;
// update the parameter defining output image file name
var ImageFileName = FileName.replace( /CURRENTFRAME/,
FrameCurrent );
oRenderMapProperty.Parameters( imagefilepath ).value =
ImageFileName;
// execute specified rendermap property
RegenerateMaps ( oRenderMapProperty.FullName );
}
}
}
From:
softimage-boun...@listproc.autodesk.commailto:softimage-boun...@listproc.autodesk.com
[mailto:softimage-boun...@listproc.autodesk.commailto:softimage-boun...@listproc.autodesk.com]
On Behalf Of Nicolas Esposito
Sent: Wednesday, January 08, 2014 3:19 PM
To: softimage@listproc.autodesk.commailto:softimage@listproc.autodesk.com
Subject: Re: Batch function for Ultimapper and Render Map
The same script functionality to execute every frame and update the output file
could be applied to Ultimapper as well, am I correct?
Sorry but I'm not too familiar with scripting, but looking at the manual looks
nothing super-complicated ;)
2014/1/9 Nicolas Esposito 3dv...@gmail.commailto:3dv...@gmail.com
This is gold!
Thanks Matt and Alan ;)
2014/1/8 Alan
Re: rigging in xsi vs maya
if you do a lot of bipeds, and you have a base set of deformers with a good naming convention. you can skin a generic biped mesh and use gator to transfer the weights... never use default weighting again. On Thu, Jan 9, 2014 at 11:32 AM, Martin furik...@gmail.com wrote: Maya's component editor really sucks because of that lock thing. It really slow my workflow so I usually deal with the export / import to SI and work there my weighting. But when I can, I skin my model in Maya before sending it to SI, because it has a much better default weighting than SI and Maya bones are easier to deal with when you do game assets (and FBX convert them nicely into nulls when importing). With proper settings you can use the default Maya weights for a test model where precise weights aren't needed. SI weights are pretty much useless, not even for a mob or test character. You need to re-weight almost everything manually, and if you work with nulls, it is even worst. I haven't used yet that heat skin or whatever from newer Maya versions but it looks cool. (I only use old versions for work) Martin Sent from my iPhone Maya has great things for it, but binding and weighting is definitely not one of them. It's pretty bad, actually. Ok, rant off.
Re: rigging in xsi vs maya
I used the default weighting before a lot, but never again, neither Maya nor Softimage. It is much faster to have a proper weighting using the inside out method on both apps. Wich in my case it is faster and more controlable in Softimage than in Maya. Specially by using the weight editor in Softimage, I had some funky experiences using the same in Maya. 2014/1/9 Steven Caron car...@gmail.com if you do a lot of bipeds, and you have a base set of deformers with a good naming convention. you can skin a generic biped mesh and use gator to transfer the weights... never use default weighting again. On Thu, Jan 9, 2014 at 11:32 AM, Martin furik...@gmail.com wrote: Maya's component editor really sucks because of that lock thing. It really slow my workflow so I usually deal with the export / import to SI and work there my weighting. But when I can, I skin my model in Maya before sending it to SI, because it has a much better default weighting than SI and Maya bones are easier to deal with when you do game assets (and FBX convert them nicely into nulls when importing). With proper settings you can use the default Maya weights for a test model where precise weights aren't needed. SI weights are pretty much useless, not even for a mob or test character. You need to re-weight almost everything manually, and if you work with nulls, it is even worst. I haven't used yet that heat skin or whatever from newer Maya versions but it looks cool. (I only use old versions for work) Martin Sent from my iPhone Maya has great things for it, but binding and weighting is definitely not one of them. It's pretty bad, actually. Ok, rant off.
Re: rigging in xsi vs maya
looks interesting eric, but i can only find one page, kind of what i had in mind but a little more abstract, i suppose it all depends how much of this system is basically reusing elements in maya, and how much is its own thing, e.g does it still use maya bones or does it have its own custom primitive, locator null, helper... thanks for sharing i had not seen this before its pretty cool :) On 9 January 2014 15:28, Eric Thivierge ethivie...@hybride.com wrote: Sebastian, look at ILM's Block Party 2 rigging system. On 1/9/2014 7:53 AM, Sebastien Sterling wrote: Why not a node based rigging system ? (not necessarily an ice node system) but its own thing, you arrange your nulls, you add rig trees to them in a small interface graph where you have nodes for different behaviours like ik, fk, hik, twist, strech, you plug the nulls according to the hierarchy you want, each node has its own params so you can expose or lock or modify them in the rig or synoptic. i'm sure such a system wouldn't cover everything, its often what i get told, that rigging is so complex a proses that in the end the longest traditional method is the only one that allows for the flexibility and reactivity necessary for a pipe. in spite of this i think such a system has merrit and deserves to go past prototype, if only to offer another perspective. its quite probable that neither xsi or mayas architecture is able to accommodate such a system natively, but plug-ins like yeti are basically like their own independent little engines running within the shell of a dcc, the same is true for fabric i assume.
Re: rigging in xsi vs maya
We worked out the default weighting for Justin using the lowpoly 'slices' of the mesh (we needed them anyway) and a bit of smoothing. Simple stuff, but with 2 clicks (a simple script) we were able to get a quite decent base to work on.
Re: rigging in xsi vs maya
Hi guys. Now I adapt the Norman's rig for XSI https://www.youtube.com/watch?v=fIUTkJcWPv8 https://www.youtube.com/watch?v=2nUTCBbQaYM https://www.youtube.com/watch?v=pZajrQCVbLU 2014/1/10 Cesar Saez cesa...@gmail.com We worked out the default weighting for Justin using the lowpoly 'slices' of the mesh (we needed them anyway) and a bit of smoothing. Simple stuff, but with 2 clicks (a simple script) we were able to get a quite decent base to work on. -- Евграфов Максим.(Summatr) https://vimeo.com/user3098735/videos --- Хорошего Вам настроения !!! :-)
Re: rigging in xsi vs maya
there are many free rigs for maya and there are little free rigs for xi. justice must prevail! 2014/1/10 Max Evgrafov summ...@gmail.com Hi guys. Now I adapt the Norman's rig for XSI https://www.youtube.com/watch?v=fIUTkJcWPv8 https://www.youtube.com/watch?v=2nUTCBbQaYM https://www.youtube.com/watch?v=pZajrQCVbLU 2014/1/10 Cesar Saez cesa...@gmail.com We worked out the default weighting for Justin using the lowpoly 'slices' of the mesh (we needed them anyway) and a bit of smoothing. Simple stuff, but with 2 clicks (a simple script) we were able to get a quite decent base to work on. -- Евграфов Максим.(Summatr) https://vimeo.com/user3098735/videos --- Хорошего Вам настроения !!! :-) -- Евграфов Максим.(Summatr) https://vimeo.com/user3098735/videos --- Хорошего Вам настроения !!! :-)
Re: rigging in xsi vs maya
Lovely stuff Max ! On 10 January 2014 04:29, Max Evgrafov summ...@gmail.com wrote: Hi guys. Now I adapt the Norman's rig for XSI https://www.youtube.com/watch?v=fIUTkJcWPv8 https://www.youtube.com/watch?v=2nUTCBbQaYM https://www.youtube.com/watch?v=pZajrQCVbLU 2014/1/10 Cesar Saez cesa...@gmail.com We worked out the default weighting for Justin using the lowpoly 'slices' of the mesh (we needed them anyway) and a bit of smoothing. Simple stuff, but with 2 clicks (a simple script) we were able to get a quite decent base to work on. -- Евграфов Максим.(Summatr) https://vimeo.com/user3098735/videos --- Хорошего Вам настроения !!! :-)
Re: rigging in xsi vs maya
Looks like you really brought him through ! Is it identical to the maya version ? at any rate i salute you sir ! On 10 January 2014 04:57, Sebastien Sterling sebastien.sterl...@gmail.comwrote: Lovely stuff Max ! On 10 January 2014 04:29, Max Evgrafov summ...@gmail.com wrote: Hi guys. Now I adapt the Norman's rig for XSI https://www.youtube.com/watch?v=fIUTkJcWPv8 https://www.youtube.com/watch?v=2nUTCBbQaYM https://www.youtube.com/watch?v=pZajrQCVbLU 2014/1/10 Cesar Saez cesa...@gmail.com We worked out the default weighting for Justin using the lowpoly 'slices' of the mesh (we needed them anyway) and a bit of smoothing. Simple stuff, but with 2 clicks (a simple script) we were able to get a quite decent base to work on. -- Евграфов Максим.(Summatr) https://vimeo.com/user3098735/videos --- Хорошего Вам настроения !!! :-)
Re: rigging in xsi vs maya
Great! I love Norman. On 1/9/2014 8:29 PM, Max Evgrafov wrote: Hi guys. Now I adapt the Norman's rig for XSI https://www.youtube.com/watch?v=fIUTkJcWPv8 https://www.youtube.com/watch?v=2nUTCBbQaYM https://www.youtube.com/watch?v=pZajrQCVbLU 2014/1/10 Cesar Saez cesa...@gmail.com mailto:cesa...@gmail.com We worked out the default weighting for Justin using the lowpoly 'slices' of the mesh (we needed them anyway) and a bit of smoothing. Simple stuff, but with 2 clicks (a simple script) we were able to get a quite decent base to work on. -- Евграфов Максим.(Summatr) https://vimeo.com/user3098735/videos --- Хорошего Вам настроения !!! :-)
Re: rigging in xsi vs maya
In games we have different name conventions, bones count and structure, model and unit size, character proportions, etc. so a base mesh is not very useful outside that project. But of course I do use Gator and Maya's copy weights to accelerate my character mass production when it is possible. I also use the SI default weighting with 1 deformer as a base, enforce limit bones if needed, then retouch and smooth where needed. SI's envelope with more than 1 deformer is unusable because you don't have any control in the smooth distance or dropoff rate. In Softimage's Set Envelope you have what, number of skeletons option and a Method option that I haven't figured out yet in what situation I could use the Normal-based one. We use only nulls in games (Softimage), not bones (bones are only for rigging), and therefore the default weights sucks even harder because the nulls position are taken as the bone center, so it is totally unusable. Re-weight from scratch. I could slice the character, envelope, smooth, and gator like Cesar said, I guess. Or I could convert my nulls to bones. I may have to write something to convert nulls to SI bones. Does anyone have a null to bone script ? When I have to do more than 1 similar character, I usually create a very low poly base character and then use Gator / Copy Weights. But if not, I have to set all the weights manually without a decent base, or do it in Maya. I remember that when I was a junior, learning Maya, somehow I messed up the weights with a couple of hours to deadline to deliver a few still renders. The character was a monster with a few tentacles and all that weird stuff so copy weights wasn't an option. The senior came, increased the default dropoff rate, apply, and the character was good enough for posing and deliver. With heat map now, it seems it's got better. And that's the only weight related thing that I like in Maya. Another thing I like in maya is that you can lock everything, even points positions (In SI I had to write an script and an ice compound to lock points), but that's another story. Martin
Re: Redshift3D Render
just played a bit with overclocking... now I pushed PCI GEN in bios back to auto instead of GEN 3 so it is now PCI 2.0 x16 instead of 3.. That alone gave me slightly higher score on cinebench.. and after overclocking CPU to 4.5 now new result in cinebench for titan is 100, and script in softimage gives me ~200 fps. Oh well... On Thu, Jan 9, 2014 at 8:24 PM, Mirko Jankovic mirkoj.anima...@gmail.comwrote: yes fps is score. hmm maybe if I got some time I could get one of titans and test it out on maximus MBO to see if that is the diference... On Thu, Jan 9, 2014 at 8:09 PM, Emilio Hernandez emi...@e-roja.comwrote: Dumb question here. Where do I see the score? I only get a fps result for OpenGl and a 759cb in the CPU results in Cinebench. and I get 75.78fps at the OpenGl test. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com cinebench you mean? so what is score? :) On Thu, Jan 9, 2014 at 7:47 PM, Emilio Hernandez emi...@e-roja.comwrote: Now I am starting to scratch my head to figure out the maze of Nvidia... At least my benchmark said the Titatn beated up a Quadro 4000K ha ha 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com getting around 95 score in cinebench with titan, and again 780 in home comp getting 140 score in cinebench... wondering how can 780 crash titan so much in opengl. titan system beside 4 titans have i7 3930K, asus p9x79-e ws MBO and home comp i7 4770k on asus maximus VI hero Will test some redshift rendering later to compare single GPU 780 vs titan On Thu, Jan 9, 2014 at 7:11 PM, Emilio Hernandez emi...@e-roja.comwrote: Yes specailly if you are getting more than the double speed. I ran the test and I got 75.78 fps Thx 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com yea I guess that woul dbe the case but just tried.. strange really no idea.. any chance to run cinebench 15 opengl test then? if that gives too low result as well then something is not good On Thu, Jan 9, 2014 at 6:58 PM, Emilio Hernandez emi...@e-roja.comwrote: It is set to ALL. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com hmm another gues.. is it set to RT or ALL in play? On Thu, Jan 9, 2014 at 6:53 PM, Emilio Hernandez emi...@e-roja.com wrote: I already did that and still getting the 65 fps limit with the Titan. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com just got back , yes 60 is vsync on, turn of vsync in nvida control panel On Thu, Jan 9, 2014 at 5:52 PM, Stefan Kubicek s...@tidbit-images.com wrote: I guess 60fps is the refresh rate of your display, right? Have you disabled VSync in the driver settings? I just get 60.0 fps + How are you getting it display a value higher than 60? I'm pretty sure it the actual fps is higher, but the value in the viewport is capped at 60 -Tim On 1/9/2014 10:12 AM, Leonard Koch wrote: I get about 28-31 out of my 680. Does anyone have a common explanation for that? On Thu, Jan 9, 2014 at 5:10 PM, Emilio Hernandez emi...@e-roja.com wrote: Hey Mirko I ran your script and I got 50.7 fps... But then I remembered I have my displays plugged in to my 470.. hahaha. Don't ask why, but when using AE with the displays plugged into the Ti, AE does not like it and disables GPU for calculations... P. 2014/1/9 Mirko Jankovic mirkoj.anima...@gmail.com Hey Tim Would you be able to take 2 minutes of your tmie and run this ol python script for SI with your titan? I'm getting weird results with an 780 in my home system outperforming titan a lot... well here is copy paste from forum if you are able to check it out as well.. thanks!: itan: ~170 fps 780: ~245 fps Go figure [image: :)] But I'm suspecting something weird with my titan system for some time will have to test further but would be great if anyone with titan as well could run it too? This old python script: Application.CreatePrim(Cube, MeshSurface, , ) Application.SetValue(cube.polymsh.geom.subdivu, 831, ) Application.SetValue(cube.polymsh.geom.subdivv, 800, ) Application.SetValue(cube.polymsh.geom.subdivbase, 800, ) Application.SetValue(Camera.camvis.refreshrate, True, ) Application.SetDisplayMode(Camera, shaded) Application.DeselectAll() Application.SetValue(PlayControl.Out, 5000, ) Application.DeselectAll() Application.GetPrim(Null, , , ) Application.SelectObj(Camera_Root, , ) Application.CopyPaste(Camera_Root, , null, 1) Application.SelectObj(null, , ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 1, , , , , ) Application.SetValue(PlayControl.Key, 5000, ) Application.SetValue(PlayControl.Current, 5000, ) Application.Rotate(, 0, 8000, 0, siAbsolute, siPivot, siObj, siY, , , , , , , , 0, ) Application.SaveKey(null.kine.local.rotx,null.kine.local.roty,null.kine.local.rotz, 5000, , , , , ) Application.FirstFrame() Just paste in python script run and hit play. Thakns! On Thu, Jan 9, 2014 at 3:34 PM, Tim
