Re: align particles on deformed surface by tangent?
hi Kris, If its on an animated surface, you'll need to rotate the vectors comparing point reference frame on the static to the post-deformed mesh, so that the vector flow is relative to the mesh and not world space. If that isn't clear, I'll do a video tomorrow. Paul On 19 Sep 2013, at 02:16, Kris Rivel krisri...@gmail.com wrote: Unfortunately, Paul's method doesn't seem to work. I followed each step but it seems somewhat dependent on the geo's global coordinates. Moving the mesh and curve around causes them to move/spin around. Where the mesh bends, it changes direction as well. Not sure why...its an easy setup to repo but my instances do not flow like his does in the video. Kris On Wed, Sep 18, 2013 at 6:46 PM, Kris Rivel krisri...@gmail.com wrote: Ah yes...thanks..I'll check that out for sure. Kris On Wed, Sep 18, 2013 at 6:29 PM, Vincent Ullmann vincent.ullm...@googlemail.com wrote: Another thing you could try: Vector Flow (by Paul Smith) Part 1: https://vimeo.com/36709750 Part 2: https://vimeo.com/36710386 Am 19.09.2013 00:27, schrieb Kris Rivel: That kind of works...but some are still pointed in random directions. But good to know how it should work! Kris On Wed, Sep 18, 2013 at 6:19 PM, Kris Rivel krisri...@gmail.com wrote: oh so you basically lay out the UV like a grid...so everything is lined up? This surface is too high res and detailed to do that. But perhaps I can use a neighboring surface that is a simple tube to extract the value from. Kris On Wed, Sep 18, 2013 at 5:54 PM, Vincent Ullmann vincent.ullm...@googlemail.com wrote: How is your UV-Map set up? You should make sure the UV-Shells are all aligned along the same Axis (U or V). As i build this compound, i made 2 UV-Sets. One for Texturing and one for Particle-Alignment. Vincent Am 18.09.2013 23:40, schrieb Kris Rivel: Oh and I took a look at this but it doesn't work well for me for some reason. They're just pointing in various directions. And I have a good uv map: https://vimeo.com/46081934 On Wed, Sep 18, 2013 at 5:39 PM, Kris Rivel krisri...@gmail.com wrote: Ugh, sounds like its going to be a head trip for me :-/ The deformed mesh is a polymesh tube. I can't change that. But maybe I can have a nurbs tube deformed along the curve too and extract its values? I'm fine with getting point normals but not sure the best way to translate and apply it. Kris On Wed, Sep 18, 2013 at 5:33 PM, Alok Gandhi alok.gandhi2...@gmail.com wrote: They wont because in effect what you are doing is just taking the normal, World Yup and crossing them which will result in a vector perpendicular to both the normal and the Yup. So during a bend, this result vector is not tangential to the surface. You definitely need a vector which is tangential to the surface at the point where the normal is coming out of the surface. I would suggest looking in the approach of getting neighbour points closest to the normal location and generating a vector from that instead of using a cross product at all. Though you have to take care of the logic around edge vertex. But it can be a good start. On Wed, Sep 18, 2013 at 5:09 PM, Vincent Fortin vfor...@gmail.com wrote: Do you want the axis perpendicular to the normal? If so use a cross product, plug the normal in the first input and try something like 0,1,0 for the second input. On Wed, Sep 18, 2013 at 4:05 PM, Kris Rivel krisri...@gmail.com wrote: I'm trying to make a simple setup where I can align particles along a deformed/animated surface based on another vector. No problem aligning them on their normal but I want them to all flow in the same direction along their other axis...all point forward. Is there anyway to do this with a polymesh? Maybe through UV data? Kris -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com
Re: Navigating relationships in Softimage (new to the software)
nice trick! On 9/18/2013 5:43 PM, David Barosin wrote: You can sort of get that with the XSI Explorer - (alt + 8) and set the viewer menu's display to schematic. On Wed, Sep 18, 2013 at 11:21 AM, Sergio Mucino sergio.muc...@modusfx.com mailto:sergio.muc...@modusfx.com wrote: Thanks Eric! Yes, the Hypergraph can get complicated, but I'm used to finding my way through it rather reasonably. I was actually thinking about the scripting side of things... I'm not sure how scriptable the behavior of the Schematic View is, but if I could get it to only display the selected node and any other connected nodes and hide the rest, that would already get me a long way towards a happier healthier life :-) . I will do the search you suggested. Thanks again! -- *Sergio Mucino* Lead Rigger Modus FX On 18/09/2013 11:12 AM, Eric Thivierge wrote: Schematic view is probably as close to the Hypergraph workflow you're going to get honestly. If your rig structure is clean and organized it shouldn't be too bad to deal with. I will say that even the Hypergraph can get a bit soupie. You may need to write yourself some scripts to present the data to you in a way you understand. If I recall there was a command that you can get the connection info in xml form that shows the 2 way connections and parse it from there. Check the google group archive and search for Alan Fregtman's post about it. I think it's been discussed a few times before. Eric T. On September-18-13 10:47:28 AM, Sergio Mucino wrote: Hello. I have a very basic question for anyone who might be able to help. I just got started using Softimage, and I'm having the usual problems anyone faces when starting to use new software they're not familiar with (just for background information purposes, I'm coming from Maya and Max. I state this just so that it gives you an idea of the language I might be using). One of the things that I'm having a hard time right now is navigating scene relationships. I'm trying to find a tool that will help me to efficiently find scene relationships, and so far, I haven't found an easy way to navigate these relationships. I'm trying to find my way through someone else's files, and finding what's related to each object is a bit of a PITA at this point. I'm using the Explorer as a starting point, but it doesn't get me there the entire way. Let me give you an example. I've got a mesh selected. In the Explorer, I can see it's being deformed by a Lattice, and an Envelope Operator. I need to find the actual scene objects that are driving these deformations. For the Envelope Op, I know I can go to Deform/Envelope/Select Deformer from Envelope, and it will select in the scene the objects that are driving the envelope's deformations (I'd rather not lose my selection, just get a list of the objects used as deformers, but I'm ok with Softimage working this way). However, the Lattice is an entirely different story. If I select the Lattice operator and inspect its properties, the only reference I can find in the Properties dialog is to its cluster. If I then find that cluster in the Cluster folder for my object, and check its properties, I get nothing. There seems to be no way I can get to the actual Lattice object in the scene deforming this cluster easily. The only way I've found so far has been to use the Schematic View. If I turn on all the Links (for the selected object only, of course. Otherwise, I just get a noodle soup), with my geo selected I can see links going to different objects in the scene. I had to track down each link (which in a scene this size took a lot of scrolling, especially since zooming out gets rid of the names on the links) until I found the link that had Lattice written on it. Then I could follow it and find the scene object that the Lattice operator refers to. I would like to know if there's a more efficient way of navigating scene relationships than this. I'm sure than being completely new to the software, I'm missing a lot. I've tried googling around, and asking people at work, but haven't been able to uncover much. If someone has some pointers towards getting stuff done faster, I'll be very glad to hear them. Just for information purposes only, I'm trying to get the same information I'd get my graphing an object in the Maya Hypergraph... a view or list of all relationships for the selected object(s). If this is not possible (I understand Softimage might have a totally different way of doing quite a few things), if someone else could describe how they tend to work to find their way in the scene,
SDK: Iterate open PPGs
Hey guys, is there any way to iterate all opened (floating + docked) Property Editor type views and figure out what they are showing? I don't seem to be able to retrieve the floating views out of the Desktop.ActiveLayout.Views, and also I don't know what to pass to GetAttributeValue on the view to get the inspected object(s). Thanks! Helge -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: SIC London next week
I wouldn't mind popping along though though it might be tricky timing for me. Maybe I can drag along a couple of other guys from work (with or without me) Simon Reeves London, UK *si...@simonreeves.com* *www.simonreeves.com* * * On 19 September 2013 11:09, Matt Morris matt...@gmail.com wrote: Be great to see you again Ola! On 17 September 2013 17:19, ola.mad...@digitalcontext.se ola.mad...@digitalcontext.se wrote: I'm in London next week so there's at least one person attending =) - Ola 17 sep 2013 kl. 17:23 skrev Matt Morris matt...@gmail.com: C'mon dan, imagine how many names to virtual faces you'll be able to put together! Register anyway and if you find time pop along for an hour or two ;) On 17 September 2013 16:20, Dan Yargici danyarg...@gmail.com wrote: I am actually in the country this time for once, but it's two days before a big delivery here so I can't imagine I'd be able to unfortunately... :( DAN On Tue, Sep 17, 2013 at 4:10 PM, adrian wyer adrian.w...@fluid-pictures.com wrote: ** ** ** ** ** Come on you lot, get yourselves registered for the Londonevent next week, it's for YOUR benefit!!! ** ** ** ** http://www.softimagecreatives.com/siclondon/?page_id=1288 ** ** ** ** a Adrian Wyer Fluid Pictures 75-77 Margaret St. London W1W 8SY ++44(0) 207 580 0829 adrian.w...@fluid-pictures.com www.fluid-pictures.com ** ** Fluid Pictures Limited is registered in **England** and *** *Wales. Company number:5657815 VAT number: 872 6893 71 ** ** -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- www.matinai.com -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- www.matinai.com -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: SDK: Iterate open PPGs
Hi Helge A ppg is not a view, so it won't show up in a Views collection. Unless something changed since 2007 when I answered this question for someone else... On 19/09/2013 5:11 AM, Helge Mathee wrote: Hey guys, is there any way to iterate all opened (floating + docked) Property Editor type views and figure out what they are showing? I don't seem to be able to retrieve the floating views out of the Desktop.ActiveLayout.Views, and also I don't know what to pass to GetAttributeValue on the view to get the inspected object(s). Thanks! Helge -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: SDK: Iterate open PPGs
That bit me more than once in the past too, it's quite a bummer there is no SDK-paved way of doing it (afaik).I ended up using Pythons the win32 extensions to get the floating window under the mouse and from the window name deduct whether it is a Property Editor or something else.But in order to get all floating windows (as opposed to all floating views!) I think you might have to resort to some low-level win32 hacking.As for getting the inspected objects shown in a Property Editor:http://download.autodesk.com/global/docs/softimage2013/en_us/sdkguide/index.html?url=""> Hey guys, is there any way to iterate all opened (floating + docked) Property Editor type views and figure out what they are showing? I don't seem to be able to retrieve the floating views out of the Desktop.ActiveLayout.Views, and also I don't know what to pass to GetAttributeValue on the view to get the inspected object(s). Thanks! Helge -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject "unsubscribe" and reply to the confirmation email.-- --- 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 attachments are confidential and for the recipient only To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Python Debugging Interfaces message
Every time I run a python script in softimage I get the following message: # INFO : Debugging interfaces not available - debugging is disabled.. Started happening randomly a week or so ago. Anyone know the cause? Eric T. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: Python Debugging Interfaces message
We had this last year as well, in 2012SAP, when we switched Soft to use the Windows Python/PyWin instead of the built-in Python. The odd thing was that in fact, debugging seemed to continue just fine, despite this message. -Tim On 9/19/2013 8:55 AM, Eric Thivierge wrote: Every time I run a python script in softimage I get the following message: # INFO : Debugging interfaces not available - debugging is disabled.. Started happening randomly a week or so ago. Anyone know the cause? Eric T. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- Signature -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: Python Debugging Interfaces message
Perhaps this? C:\Program Files\Autodesk\Softimage 2014\Application\python\Lib\site-packages\win32comext\axscript\client\framework.py def SetScriptSite(self, site): # We should still work with an existing site (or so MSXML believes :) self.scriptSite = site if self.debugManager is not None: self.debugManager.Close() import traceback try: import win32com.axdebug.axdebug # see if the core exists. import debug self.debugManager = debug.DebugManager(self) except pythoncom.com_error: # COM errors will occur if the debugger interface has never been # seen on the target system trace(Debugging interfaces not available - debugging is disabled..) self.debugManager = None Debugging interfaces not available - debugging is disabled has been around, on and off, for almost a decade now. It was in the xsibase faq: http://web.archive.org/web/20120715063059/http://www.xsibase.com/forum/index.php?board=14;action=display;threadid=9773 On 19/09/2013 9:55 AM, Eric Thivierge wrote: Every time I run a python script in softimage I get the following message: # INFO : Debugging interfaces not available - debugging is disabled.. Started happening randomly a week or so ago. Anyone know the cause? Eric T. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Question about using ICE to move points
Okay. I'm trying to figure out how I'd need to go about getting this to work. I have a character with two different models for its head. One has a nose, one doesn't. At one point, the animators will need to switch to one or the other, so both models are needed. The head with the nose (lets call it HeadNose) is already rigged and deforming. The rig and the deformations stack is rather complex, with this head making references to another head model via an ICE tree, plus an Envelope that references other objects. My entire point with this, is that I cannot just transfer the deformations from one model to another via GATOR, since I cannot get the deformations from just one place. So, my idea was to create an ICE tree on the head without nose (lets call it HeadNoseless), and make it reference the points on HeadNose to use the same positions. This effectively 'snaps' the verts on this head to the other one. Both models are exactly the same topology (the nose was removed by sculpting it flat, basically), so in essence, this works. The problem, of course, is that by doing this, the nose comes back, since the points for it that were flattened are repositioned to match the existing nose. To avoid this, I thought of using a Linear Interpolation between both heads, using a Weight Map I painted just on the nose area to determine which points will move based on whom (I of course used the HeadNoseless PointPos attribute as the source of the second position for the points). This worked, leaving the nose flattened, but it creates the problem that HeadNoseless follows the deformations of HeadNose correctly, except in the nose area (since it was painted out), leaving the flattened nose floating in space. So, I decided that what I need to do is to keep the nose flattened like this, but get it to read the positions of its matching points on HeadNose, and move the same amount but keeping the existing delta between them. That way, I could get to keep the nose's current flattened shape, but it'd move matching the movements of the HeadNose object. And that's where I'm stuck. I've tried to implement this in different ways, but I can't seem to be able to get the proper nodes and operations to get this to work. I know I need to add movement to the existing point positions, but it has to be relative to these positions. If anyone has any pointers around how to achieve this, I'm all ears. Thanks! -- *Sergio Mucino* Lead Rigger Modus FX -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: Question about using ICE to move points
Okay, I think I misunderstood what you suggested. I was referring to Point Clusters, whereas you mentioned Shape Cluster. I will look into that right now. *Sergio Mucino* Lead Rigger Modus FX On 19/09/2013 4:18 PM, Sergio Mucino wrote: Thanks Christian. Couple of questions... 1. Which property in a cluster can I save vector data onto? I can't seem to find any. I also looked at adding my own custom parameter, but there seems to be no support for vector data types. Am I missing something? (Forgive me, this is my 4th day using SI) (And it seems that Set Data doesn't like Clusters in general... I can't get it to select anything). 2. Freezing the stack... I'm not sure about the implications of this. AFAIK, freezing the stack will delete all deformations happening in it, right? This particular model uses two ICE trees at different parts in the stack. Under the Modeling section, I have an ICE Tree that just gets and sets point position data. I use it to reference into the model some shape changes that come from revisions. It's basically a simple Shape 'morpher'. That way, if another revision comes in, I'll just reference it, and my head updates. The second ICE tree is the one than I'm trying to build, which is under the Animation section. I'm not sure how I'd go about freezing the stack and keeping these trees alive. Can you elaborate if you have some time about what you meant? Thanks again! *Sergio Mucino* Lead Rigger Modus FX On 19/09/2013 3:54 PM, Christian Freisleder wrote: hey, you could save a offset map in a shape cluster property. you subtract the point pos of both heads and save the value in that cluster. then you freeze the stack and and add the value of that property to the point positions of the other head. freezing the modeling stack is enough. hope that gets you going. christian On 19.09.2013 21:30, Sergio Mucino wrote: Okay. I'm trying to figure out how I'd need to go about getting this to work. I have a character with two different models for its head. One has a nose, one doesn't. At one point, the animators will need to switch to one or the other, so both models are needed. The head with the nose (lets call it HeadNose) is already rigged and deforming. The rig and the deformations stack is rather complex, with this head making references to another head model via an ICE tree, plus an Envelope that references other objects. My entire point with this, is that I cannot just transfer the deformations from one model to another via GATOR, since I cannot get the deformations from just one place. So, my idea was to create an ICE tree on the head without nose (lets call it HeadNoseless), and make it reference the points on HeadNose to use the same positions. This effectively 'snaps' the verts on this head to the other one. Both models are exactly the same topology (the nose was removed by sculpting it flat, basically), so in essence, this works. The problem, of course, is that by doing this, the nose comes back, since the points for it that were flattened are repositioned to match the existing nose. To avoid this, I thought of using a Linear Interpolation between both heads, using a Weight Map I painted just on the nose area to determine which points will move based on whom (I of course used the HeadNoseless PointPos attribute as the source of the second position for the points). This worked, leaving the nose flattened, but it creates the problem that HeadNoseless follows the deformations of HeadNose correctly, except in the nose area (since it was painted out), leaving the flattened nose floating in space. So, I decided that what I need to do is to keep the nose flattened like this, but get it to read the positions of its matching points on HeadNose, and move the same amount but keeping the existing delta between them. That way, I could get to keep the nose's current flattened shape, but it'd move matching the movements of the HeadNose object. And that's where I'm stuck. I've tried to implement this in different ways, but I can't seem to be able to get the proper nodes and operations to get this to work. I know I need to add movement to the existing point positions, but it has to be relative to these positions. If anyone has any pointers around how to achieve this, I'm all ears. Thanks! -- *Sergio Mucino* Lead Rigger Modus FX -- To unsubscribe: mailsoftimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mailsoftimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
RE: Navigating relationships in Softimage (new to the software)
It will also select above the scene root - which is not a problem for lattices as lattices cannot exist above the scene root, but other types like custom properties can. Matt From: softimage-boun...@listproc.autodesk.com [mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Christian Gotzinger Sent: Thursday, September 19, 2013 10:02 AM To: softimage@listproc.autodesk.com Subject: Re: Navigating relationships in Softimage (new to the software) Just a quick tip: you can also press Ctrl+A, which always respects your selection filter. If the filter is set to lattice, Ctrl+A selects all lattices in the scene. For example, if you want to find all lattices in the scene, you can set the selection filter to lattice then do a rectangular selection around the entire scene (or in schematic view). -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
ApplyICEOp disable pick session?
Is there a way to disable the pick session when using the ApplyICEOp command? It seems that some compounds are insisting on entering a pick session and some aren't. I'd like to not have the pick session but have the PPG for the main compound to open when clicking on the ICETree operator. Thoughts? Thanks, Eric T. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: ApplyICEOp disable pick session?
Depends on the compound. The docs say * Input ports withPicking string for operatordefined in their port properties launch a picking session using the string as the prompt. * IfPicking string for operatorhas not been defined for any input port, then a picking session is launched for every scene reference input. On 19/09/2013 3:00 PM, Eric Thivierge wrote: While that is probably so, it wasn't the case as I am applying 3 compounds one after another and the 1st one asked for no pick session, second asked for it, then third did not. On September-19-13 1:55:05 PM, Matt Lind wrote: Pick session is an option the user can assign to the ICE compound at time of creation for the purpose of forcing the user to pick an input when it's missing. Somebody must've activated that feature. Matt -Original Message- From: softimage-boun...@listproc.autodesk.com [mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Eric Thivierge Sent: Thursday, September 19, 2013 10:21 AM To: softimage@listproc.autodesk.com Subject: Re: ApplyICEOp disable pick session? Not sure what the issue was, but exploding the compound and repackaging fixed it. Also I added pass through nodes to the get data nodes that may also be fixing it. Carry on, Eric T. On September-19-13 12:59:22 PM, Eric Thivierge wrote: Is there a way to disable the pick session when using the ApplyICEOp command? It seems that some compounds are insisting on entering a pick session and some aren't. I'd like to not have the pick session but have the PPG for the main compound to open when clicking on the ICETree operator. Thoughts? Thanks, Eric T. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: Question about using ICE to move points (NOT SOLVED... yet)
Okay. I discovered another funky issue. The nose points are double-transforming. I'll have to check why they are doing this. *Sergio Mucino* Lead Rigger Modus FX On 19/09/2013 5:10 PM, Sergio Mucino wrote: SOLVED! There was a 'broken' reference in a different ICE Tree on another object (I have no idea how it broke. I'll double check before calling it a day). But this solution works. Thanks a lot Christian! *Sergio Mucino* Lead Rigger Modus FX On 19/09/2013 4:28 PM, Sergio Mucino wrote: Okay, I think I misunderstood what you suggested. I was referring to Point Clusters, whereas you mentioned Shape Cluster. I will look into that right now. *Sergio Mucino* Lead Rigger Modus FX On 19/09/2013 4:18 PM, Sergio Mucino wrote: Thanks Christian. Couple of questions... 1. Which property in a cluster can I save vector data onto? I can't seem to find any. I also looked at adding my own custom parameter, but there seems to be no support for vector data types. Am I missing something? (Forgive me, this is my 4th day using SI) (And it seems that Set Data doesn't like Clusters in general... I can't get it to select anything). 2. Freezing the stack... I'm not sure about the implications of this. AFAIK, freezing the stack will delete all deformations happening in it, right? This particular model uses two ICE trees at different parts in the stack. Under the Modeling section, I have an ICE Tree that just gets and sets point position data. I use it to reference into the model some shape changes that come from revisions. It's basically a simple Shape 'morpher'. That way, if another revision comes in, I'll just reference it, and my head updates. The second ICE tree is the one than I'm trying to build, which is under the Animation section. I'm not sure how I'd go about freezing the stack and keeping these trees alive. Can you elaborate if you have some time about what you meant? Thanks again! *Sergio Mucino* Lead Rigger Modus FX On 19/09/2013 3:54 PM, Christian Freisleder wrote: hey, you could save a offset map in a shape cluster property. you subtract the point pos of both heads and save the value in that cluster. then you freeze the stack and and add the value of that property to the point positions of the other head. freezing the modeling stack is enough. hope that gets you going. christian On 19.09.2013 21:30, Sergio Mucino wrote: Okay. I'm trying to figure out how I'd need to go about getting this to work. I have a character with two different models for its head. One has a nose, one doesn't. At one point, the animators will need to switch to one or the other, so both models are needed. The head with the nose (lets call it HeadNose) is already rigged and deforming. The rig and the deformations stack is rather complex, with this head making references to another head model via an ICE tree, plus an Envelope that references other objects. My entire point with this, is that I cannot just transfer the deformations from one model to another via GATOR, since I cannot get the deformations from just one place. So, my idea was to create an ICE tree on the head without nose (lets call it HeadNoseless), and make it reference the points on HeadNose to use the same positions. This effectively 'snaps' the verts on this head to the other one. Both models are exactly the same topology (the nose was removed by sculpting it flat, basically), so in essence, this works. The problem, of course, is that by doing this, the nose comes back, since the points for it that were flattened are repositioned to match the existing nose. To avoid this, I thought of using a Linear Interpolation between both heads, using a Weight Map I painted just on the nose area to determine which points will move based on whom (I of course used the HeadNoseless PointPos attribute as the source of the second position for the points). This worked, leaving the nose flattened, but it creates the problem that HeadNoseless follows the deformations of HeadNose correctly, except in the nose area (since it was painted out), leaving the flattened nose floating in space. So, I decided that what I need to do is to keep the nose flattened like this, but get it to read the positions of its matching points on HeadNose, and move the same amount but keeping the existing delta between them. That way, I could get to keep the nose's current flattened shape, but it'd move matching the movements of the HeadNose object. And that's where I'm stuck. I've tried to implement this in different ways, but I can't seem to be able to get the proper nodes and operations to get this to work. I know I need to add movement to the existing point positions, but it has to be relative to these positions. If anyone has any pointers around how to achieve this, I'm all ears. Thanks! -- *Sergio Mucino* Lead Rigger Modus FX -- To unsubscribe: mailsoftimage-requ...@listproc.autodesk.com with subject
Re: Defining GL textures
Thanks a ton Steve, I'd never seen that little tab before... just when you think you've dug around everywhere there's always a usefull little something that stays hidden. On Thu, Sep 19, 2013 at 3:16 PM, Steven Caron car...@gmail.com wrote: inspect the root part of the material, opengl display tab... various options. On Thu, Sep 19, 2013 at 3:03 PM, Simon van de Lagemaat si...@theembassyvfx.com wrote: How can I specify which clipss on a surface appear in GL viewports? Currently it seems like Soft is picking up whatever clip I have loaded last. If I change the clip in the UV editor it changes but the moment I close it it resets back. Changing the clips to not use auto also resets after closing the window. There must be a way to define what clips are used for an object GL display right? -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
RE: Defining GL textures
It's based on the 'current' texture displayed in the shader. You can find the setting by inspecting the material which the shader is connected on the OpenGL tab. However, if you're in realtime shader OpenGL shade mode or High quality viewport, the shader can override that parameter and display something else - assuming the shader is a realtime shader such as GLSL, CgFX, etc... Matt From: softimage-boun...@listproc.autodesk.com [mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Simon van de Lagemaat Sent: Thursday, September 19, 2013 3:04 PM To: softimage@listproc.autodesk.com Subject: Defining GL textures How can I specify which clipss on a surface appear in GL viewports? Currently it seems like Soft is picking up whatever clip I have loaded last. If I change the clip in the UV editor it changes but the moment I close it it resets back. Changing the clips to not use auto also resets after closing the window. There must be a way to define what clips are used for an object GL display right? -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: Navigating relationships in Softimage (new to the software)
Just a quick tip: you can also press Ctrl+A, which always respects your selection filter. If the filter is set to lattice, Ctrl+A selects all lattices in the scene. For example, if you want to find all lattices in the scene, you can set the selection filter to lattice then do a rectangular selection around the entire scene (or in schematic view). -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
RE: ApplyICEOp disable pick session?
Pick session is an option the user can assign to the ICE compound at time of creation for the purpose of forcing the user to pick an input when it's missing. Somebody must've activated that feature. Matt -Original Message- From: softimage-boun...@listproc.autodesk.com [mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Eric Thivierge Sent: Thursday, September 19, 2013 10:21 AM To: softimage@listproc.autodesk.com Subject: Re: ApplyICEOp disable pick session? Not sure what the issue was, but exploding the compound and repackaging fixed it. Also I added pass through nodes to the get data nodes that may also be fixing it. Carry on, Eric T. On September-19-13 12:59:22 PM, Eric Thivierge wrote: Is there a way to disable the pick session when using the ApplyICEOp command? It seems that some compounds are insisting on entering a pick session and some aren't. I'd like to not have the pick session but have the PPG for the main compound to open when clicking on the ICETree operator. Thoughts? Thanks, Eric T. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: ApplyICEOp disable pick session?
Not sure what the issue was, but exploding the compound and repackaging fixed it. Also I added pass through nodes to the get data nodes that may also be fixing it. Carry on, Eric T. On September-19-13 12:59:22 PM, Eric Thivierge wrote: Is there a way to disable the pick session when using the ApplyICEOp command? It seems that some compounds are insisting on entering a pick session and some aren't. I'd like to not have the pick session but have the PPG for the main compound to open when clicking on the ICETree operator. Thoughts? Thanks, Eric T. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: Question about using ICE to move points
Okay. I think this sorta worked, but I cannot be too sure for one reason... only the hair on the head is moving. The reason for this might be that the head has a Weight Map that has the hair weighted to 1 on it. This map is then referenced by the Modeling ICE tree (which I moved over to Shapes before freezing the Modeling stack, to avoid losing it) so that I can quickly update any modeling revisions done to the hair. So, for some reason, it seems this weight map is preventing the rest of the head from moving. However, I cannot know if it has something to do with the ICE Tree in the Shapes stack (since they cannot be muted, I have no idea), or with the data stored in the Shape Cluster. Is there any way I can visualize the data stored in the attribute holding the deltas? That would allow me to know if the stored deltas are wrong, or if it has something to do with the other operators in the stack. Thanks! *Sergio Mucino* Lead Rigger Modus FX On 19/09/2013 4:28 PM, Sergio Mucino wrote: Okay, I think I misunderstood what you suggested. I was referring to Point Clusters, whereas you mentioned Shape Cluster. I will look into that right now. *Sergio Mucino* Lead Rigger Modus FX On 19/09/2013 4:18 PM, Sergio Mucino wrote: Thanks Christian. Couple of questions... 1. Which property in a cluster can I save vector data onto? I can't seem to find any. I also looked at adding my own custom parameter, but there seems to be no support for vector data types. Am I missing something? (Forgive me, this is my 4th day using SI) (And it seems that Set Data doesn't like Clusters in general... I can't get it to select anything). 2. Freezing the stack... I'm not sure about the implications of this. AFAIK, freezing the stack will delete all deformations happening in it, right? This particular model uses two ICE trees at different parts in the stack. Under the Modeling section, I have an ICE Tree that just gets and sets point position data. I use it to reference into the model some shape changes that come from revisions. It's basically a simple Shape 'morpher'. That way, if another revision comes in, I'll just reference it, and my head updates. The second ICE tree is the one than I'm trying to build, which is under the Animation section. I'm not sure how I'd go about freezing the stack and keeping these trees alive. Can you elaborate if you have some time about what you meant? Thanks again! *Sergio Mucino* Lead Rigger Modus FX On 19/09/2013 3:54 PM, Christian Freisleder wrote: hey, you could save a offset map in a shape cluster property. you subtract the point pos of both heads and save the value in that cluster. then you freeze the stack and and add the value of that property to the point positions of the other head. freezing the modeling stack is enough. hope that gets you going. christian On 19.09.2013 21:30, Sergio Mucino wrote: Okay. I'm trying to figure out how I'd need to go about getting this to work. I have a character with two different models for its head. One has a nose, one doesn't. At one point, the animators will need to switch to one or the other, so both models are needed. The head with the nose (lets call it HeadNose) is already rigged and deforming. The rig and the deformations stack is rather complex, with this head making references to another head model via an ICE tree, plus an Envelope that references other objects. My entire point with this, is that I cannot just transfer the deformations from one model to another via GATOR, since I cannot get the deformations from just one place. So, my idea was to create an ICE tree on the head without nose (lets call it HeadNoseless), and make it reference the points on HeadNose to use the same positions. This effectively 'snaps' the verts on this head to the other one. Both models are exactly the same topology (the nose was removed by sculpting it flat, basically), so in essence, this works. The problem, of course, is that by doing this, the nose comes back, since the points for it that were flattened are repositioned to match the existing nose. To avoid this, I thought of using a Linear Interpolation between both heads, using a Weight Map I painted just on the nose area to determine which points will move based on whom (I of course used the HeadNoseless PointPos attribute as the source of the second position for the points). This worked, leaving the nose flattened, but it creates the problem that HeadNoseless follows the deformations of HeadNose correctly, except in the nose area (since it was painted out), leaving the flattened nose floating in space. So, I decided that what I need to do is to keep the nose flattened like this, but get it to read the positions of its matching points on HeadNose, and move the same amount but keeping the existing delta between them. That way, I could get to keep the nose's current flattened shape, but it'd move matching the movements of the HeadNose object.
Re: Defining GL textures
inspect the root part of the material, opengl display tab... various options. On Thu, Sep 19, 2013 at 3:03 PM, Simon van de Lagemaat si...@theembassyvfx.com wrote: How can I specify which clipss on a surface appear in GL viewports? Currently it seems like Soft is picking up whatever clip I have loaded last. If I change the clip in the UV editor it changes but the moment I close it it resets back. Changing the clips to not use auto also resets after closing the window. There must be a way to define what clips are used for an object GL display right? -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: ApplyICEOp disable pick session?
That would explain it Stephen and also explain why putting a pass through fixes my issue. Stephen, the best Softimage support technician... working for Solid Angle. :P On September-19-13 3:18:45 PM, Stephen Blair wrote: Depends on the compound. The docs say * Input ports withPicking string for operatordefined in their port properties launch a picking session using the string as the prompt. * IfPicking string for operatorhas not been defined for any input port, then a picking session is launched for every scene reference input. On 19/09/2013 3:00 PM, Eric Thivierge wrote: While that is probably so, it wasn't the case as I am applying 3 compounds one after another and the 1st one asked for no pick session, second asked for it, then third did not. On September-19-13 1:55:05 PM, Matt Lind wrote: Pick session is an option the user can assign to the ICE compound at time of creation for the purpose of forcing the user to pick an input when it's missing. Somebody must've activated that feature. Matt -Original Message- From:softimage-boun...@listproc.autodesk.com [mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Eric Thivierge Sent: Thursday, September 19, 2013 10:21 AM To:softimage@listproc.autodesk.com Subject: Re: ApplyICEOp disable pick session? Not sure what the issue was, but exploding the compound and repackaging fixed it. Also I added pass through nodes to the get data nodes that may also be fixing it. Carry on, Eric T. On September-19-13 12:59:22 PM, Eric Thivierge wrote: Is there a way to disable the pick session when using the ApplyICEOp command? It seems that some compounds are insisting on entering a pick session and some aren't. I'd like to not have the pick session but have the PPG for the main compound to open when clicking on the ICETree operator. Thoughts? Thanks, Eric T. -- To unsubscribe: mailsoftimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mailsoftimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mailsoftimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mailsoftimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: Question about using ICE to move points
Thanks Christian. Couple of questions... 1. Which property in a cluster can I save vector data onto? I can't seem to find any. I also looked at adding my own custom parameter, but there seems to be no support for vector data types. Am I missing something? (Forgive me, this is my 4th day using SI) (And it seems that Set Data doesn't like Clusters in general... I can't get it to select anything). 2. Freezing the stack... I'm not sure about the implications of this. AFAIK, freezing the stack will delete all deformations happening in it, right? This particular model uses two ICE trees at different parts in the stack. Under the Modeling section, I have an ICE Tree that just gets and sets point position data. I use it to reference into the model some shape changes that come from revisions. It's basically a simple Shape 'morpher'. That way, if another revision comes in, I'll just reference it, and my head updates. The second ICE tree is the one than I'm trying to build, which is under the Animation section. I'm not sure how I'd go about freezing the stack and keeping these trees alive. Can you elaborate if you have some time about what you meant? Thanks again! *Sergio Mucino* Lead Rigger Modus FX On 19/09/2013 3:54 PM, Christian Freisleder wrote: hey, you could save a offset map in a shape cluster property. you subtract the point pos of both heads and save the value in that cluster. then you freeze the stack and and add the value of that property to the point positions of the other head. freezing the modeling stack is enough. hope that gets you going. christian On 19.09.2013 21:30, Sergio Mucino wrote: Okay. I'm trying to figure out how I'd need to go about getting this to work. I have a character with two different models for its head. One has a nose, one doesn't. At one point, the animators will need to switch to one or the other, so both models are needed. The head with the nose (lets call it HeadNose) is already rigged and deforming. The rig and the deformations stack is rather complex, with this head making references to another head model via an ICE tree, plus an Envelope that references other objects. My entire point with this, is that I cannot just transfer the deformations from one model to another via GATOR, since I cannot get the deformations from just one place. So, my idea was to create an ICE tree on the head without nose (lets call it HeadNoseless), and make it reference the points on HeadNose to use the same positions. This effectively 'snaps' the verts on this head to the other one. Both models are exactly the same topology (the nose was removed by sculpting it flat, basically), so in essence, this works. The problem, of course, is that by doing this, the nose comes back, since the points for it that were flattened are repositioned to match the existing nose. To avoid this, I thought of using a Linear Interpolation between both heads, using a Weight Map I painted just on the nose area to determine which points will move based on whom (I of course used the HeadNoseless PointPos attribute as the source of the second position for the points). This worked, leaving the nose flattened, but it creates the problem that HeadNoseless follows the deformations of HeadNose correctly, except in the nose area (since it was painted out), leaving the flattened nose floating in space. So, I decided that what I need to do is to keep the nose flattened like this, but get it to read the positions of its matching points on HeadNose, and move the same amount but keeping the existing delta between them. That way, I could get to keep the nose's current flattened shape, but it'd move matching the movements of the HeadNose object. And that's where I'm stuck. I've tried to implement this in different ways, but I can't seem to be able to get the proper nodes and operations to get this to work. I know I need to add movement to the existing point positions, but it has to be relative to these positions. If anyone has any pointers around how to achieve this, I'm all ears. Thanks! -- *Sergio Mucino* Lead Rigger Modus FX -- To unsubscribe: mailsoftimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: Question about using ICE to move points (SOLVED)
SOLVED! There was a 'broken' reference in a different ICE Tree on another object (I have no idea how it broke. I'll double check before calling it a day). But this solution works. Thanks a lot Christian! *Sergio Mucino* Lead Rigger Modus FX On 19/09/2013 4:28 PM, Sergio Mucino wrote: Okay, I think I misunderstood what you suggested. I was referring to Point Clusters, whereas you mentioned Shape Cluster. I will look into that right now. *Sergio Mucino* Lead Rigger Modus FX On 19/09/2013 4:18 PM, Sergio Mucino wrote: Thanks Christian. Couple of questions... 1. Which property in a cluster can I save vector data onto? I can't seem to find any. I also looked at adding my own custom parameter, but there seems to be no support for vector data types. Am I missing something? (Forgive me, this is my 4th day using SI) (And it seems that Set Data doesn't like Clusters in general... I can't get it to select anything). 2. Freezing the stack... I'm not sure about the implications of this. AFAIK, freezing the stack will delete all deformations happening in it, right? This particular model uses two ICE trees at different parts in the stack. Under the Modeling section, I have an ICE Tree that just gets and sets point position data. I use it to reference into the model some shape changes that come from revisions. It's basically a simple Shape 'morpher'. That way, if another revision comes in, I'll just reference it, and my head updates. The second ICE tree is the one than I'm trying to build, which is under the Animation section. I'm not sure how I'd go about freezing the stack and keeping these trees alive. Can you elaborate if you have some time about what you meant? Thanks again! *Sergio Mucino* Lead Rigger Modus FX On 19/09/2013 3:54 PM, Christian Freisleder wrote: hey, you could save a offset map in a shape cluster property. you subtract the point pos of both heads and save the value in that cluster. then you freeze the stack and and add the value of that property to the point positions of the other head. freezing the modeling stack is enough. hope that gets you going. christian On 19.09.2013 21:30, Sergio Mucino wrote: Okay. I'm trying to figure out how I'd need to go about getting this to work. I have a character with two different models for its head. One has a nose, one doesn't. At one point, the animators will need to switch to one or the other, so both models are needed. The head with the nose (lets call it HeadNose) is already rigged and deforming. The rig and the deformations stack is rather complex, with this head making references to another head model via an ICE tree, plus an Envelope that references other objects. My entire point with this, is that I cannot just transfer the deformations from one model to another via GATOR, since I cannot get the deformations from just one place. So, my idea was to create an ICE tree on the head without nose (lets call it HeadNoseless), and make it reference the points on HeadNose to use the same positions. This effectively 'snaps' the verts on this head to the other one. Both models are exactly the same topology (the nose was removed by sculpting it flat, basically), so in essence, this works. The problem, of course, is that by doing this, the nose comes back, since the points for it that were flattened are repositioned to match the existing nose. To avoid this, I thought of using a Linear Interpolation between both heads, using a Weight Map I painted just on the nose area to determine which points will move based on whom (I of course used the HeadNoseless PointPos attribute as the source of the second position for the points). This worked, leaving the nose flattened, but it creates the problem that HeadNoseless follows the deformations of HeadNose correctly, except in the nose area (since it was painted out), leaving the flattened nose floating in space. So, I decided that what I need to do is to keep the nose flattened like this, but get it to read the positions of its matching points on HeadNose, and move the same amount but keeping the existing delta between them. That way, I could get to keep the nose's current flattened shape, but it'd move matching the movements of the HeadNose object. And that's where I'm stuck. I've tried to implement this in different ways, but I can't seem to be able to get the proper nodes and operations to get this to work. I know I need to add movement to the existing point positions, but it has to be relative to these positions. If anyone has any pointers around how to achieve this, I'm all ears. Thanks! -- *Sergio Mucino* Lead Rigger Modus FX -- To unsubscribe: mailsoftimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mailsoftimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: Question about using ICE to move points
hey, you could save a offset map in a shape cluster property. you subtract the point pos of both heads and save the value in that cluster. then you freeze the stack and and add the value of that property to the point positions of the other head. freezing the modeling stack is enough. hope that gets you going. christian On 19.09.2013 21:30, Sergio Mucino wrote: Okay. I'm trying to figure out how I'd need to go about getting this to work. I have a character with two different models for its head. One has a nose, one doesn't. At one point, the animators will need to switch to one or the other, so both models are needed. The head with the nose (lets call it HeadNose) is already rigged and deforming. The rig and the deformations stack is rather complex, with this head making references to another head model via an ICE tree, plus an Envelope that references other objects. My entire point with this, is that I cannot just transfer the deformations from one model to another via GATOR, since I cannot get the deformations from just one place. So, my idea was to create an ICE tree on the head without nose (lets call it HeadNoseless), and make it reference the points on HeadNose to use the same positions. This effectively 'snaps' the verts on this head to the other one. Both models are exactly the same topology (the nose was removed by sculpting it flat, basically), so in essence, this works. The problem, of course, is that by doing this, the nose comes back, since the points for it that were flattened are repositioned to match the existing nose. To avoid this, I thought of using a Linear Interpolation between both heads, using a Weight Map I painted just on the nose area to determine which points will move based on whom (I of course used the HeadNoseless PointPos attribute as the source of the second position for the points). This worked, leaving the nose flattened, but it creates the problem that HeadNoseless follows the deformations of HeadNose correctly, except in the nose area (since it was painted out), leaving the flattened nose floating in space. So, I decided that what I need to do is to keep the nose flattened like this, but get it to read the positions of its matching points on HeadNose, and move the same amount but keeping the existing delta between them. That way, I could get to keep the nose's current flattened shape, but it'd move matching the movements of the HeadNose object. And that's where I'm stuck. I've tried to implement this in different ways, but I can't seem to be able to get the proper nodes and operations to get this to work. I know I need to add movement to the existing point positions, but it has to be relative to these positions. If anyone has any pointers around how to achieve this, I'm all ears. Thanks! -- *Sergio Mucino* Lead Rigger Modus FX -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: ApplyICEOp disable pick session?
While that is probably so, it wasn't the case as I am applying 3 compounds one after another and the 1st one asked for no pick session, second asked for it, then third did not. On September-19-13 1:55:05 PM, Matt Lind wrote: Pick session is an option the user can assign to the ICE compound at time of creation for the purpose of forcing the user to pick an input when it's missing. Somebody must've activated that feature. Matt -Original Message- From: softimage-boun...@listproc.autodesk.com [mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Eric Thivierge Sent: Thursday, September 19, 2013 10:21 AM To: softimage@listproc.autodesk.com Subject: Re: ApplyICEOp disable pick session? Not sure what the issue was, but exploding the compound and repackaging fixed it. Also I added pass through nodes to the get data nodes that may also be fixing it. Carry on, Eric T. On September-19-13 12:59:22 PM, Eric Thivierge wrote: Is there a way to disable the pick session when using the ApplyICEOp command? It seems that some compounds are insisting on entering a pick session and some aren't. I'd like to not have the pick session but have the PPG for the main compound to open when clicking on the ICETree operator. Thoughts? Thanks, Eric T. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email. -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Defining GL textures
How can I specify which clipss on a surface appear in GL viewports? Currently it seems like Soft is picking up whatever clip I have loaded last. If I change the clip in the UV editor it changes but the moment I close it it resets back. Changing the clips to not use auto also resets after closing the window. There must be a way to define what clips are used for an object GL display right? -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
RE: 2014
Matt, it looks like you are right, as usual. The delta positions and constraints are silently failing to be applied on scene load. The deltas are there, just not being applied. Using the “Models to reconnect to” has no apparent effect. Re-adding the constraints just adds another entry in the delta which is ignored just like the first one when the scene is loaded. At least it doesn’t crash which is what I think it used to do.. looks like I’m back to embedding models though, at least on the animation authoring scenes. So that’s a slight improvement. I am really surprised something as potentially useful as referenced models remain so broken after so many years. Is there some trick to making it work? External deltas perhaps? I’m going to give up again on these unless I hear of something that might work. Out of curiosity, does max or maya have truly functioning refmodels? Or any app? From: softimage-boun...@listproc.autodesk.com [mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Matt Lind Sent: Wednesday, September 18, 2013 8:14 PM To: softimage@listproc.autodesk.com Subject: RE: 2014 I do not advise using nested models. When you get to the more complicated relationships such as model deltas, group relations, animation mixer propagation, instances, and so on, you’re just asking for trouble. Glad to see improvements, however, as it has far reaching effects beyond nesting. Matt From: softimage-boun...@listproc.autodesk.com [mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Nick Angus Sent: Wednesday, September 18, 2013 8:10 PM To: softimage@listproc.autodesk.com Subject: RE: 2014 Thanks Eric, we had huge troubles with nested references in 2013, in fact we gave up on it. So that is definitely a tick in the pro box! Cheers, Nick Sent from my Windows Phone _ From: Eric Cosky mailto:e...@cosky.com Sent: 19/09/2013 9:18 AM To: softimage@listproc.autodesk.com Subject: RE: 2014 I’ve been successfully using referenced models that contain other reference models with SI2014. This used to cause stability problems in earlier versions, but it’s been a while since I checked.. it might have been 2012 or earlier that I last checked. I’m pretty happy to see it working these days, regardless. My usage is generally limited to game export related stuff without referenced material libs, so YMMV. There’s been enough problems with refmodels over the years I’m not about to say it’s problem free but being able to nest referenced models is pretty useful for my workflow and I’m glad it is working for me. From: softimage-boun...@listproc.autodesk.com [mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Nick Angus Sent: Tuesday, September 17, 2013 10:12 PM To: softimage@listproc.autodesk.com Subject: 2014 About to take the plunge and update from 2013 (not sp1) to 2014 (latest build). Are there any caveats I should be wary of? Appreciate any advice either way! Cheers, Nick -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: 2014
They aren't broken, referencing references is a million miles away from a trivial problem. The key to it is order of execution, which is neigh impossible to guarantee without responsabilizing users well beyond what your average seat warmer can deal with from a technical stand point. The way Soft deals with name spaces and hierarchies and referencing within models when you have nested models you assume an order of operations that is simply not guaranteed. So it's perfectly possible, in example, that a model depending on its father receives a namespace and before the parent does. When you factor in deltas and cross dependencies the problem becomes even worse. It's simply an incredibly hard problem to solve generically in the context of a pipeline (end user's) which the provider client (Soft) can't have any notion of. We and others spend untold amount of hours to develop tools and procedures to allow cross model connections, and they are a brittle, nasty thing to deal with, and it only works because we're extremely well serviced on the asset dynamicism and management side of the pipe, and everything users do that establishes those cross namespace connections is managed by wrappers that also take care of establishing order of stashes and stacking within the stash. Offering the same thing arbitrarily in a DCC client would be exceedingly hard, or would greatly restrict what you could do with those namespaces and referencing. This isn't to say ref models are perfect, but refs of refs is one of those things people take for granted is just buggy when it's actually an exceedingly hard problem to work out generically if you want the app to remain friendly to all levels of technical competence. On Fri, Sep 20, 2013 at 1:03 PM, Eric Cosky e...@cosky.com wrote: ** ** Matt, it looks like you are right, as usual. The delta positions and constraints are silently failing to be applied on scene load. The deltas are there, just not being applied. Using the “Models to reconnect to” has no apparent effect. Re-adding the constraints just adds another entry in the delta which is ignored just like the first one when the scene is loaded. At least it doesn’t crash which is what I think it used to do.. looks like I’m back to embedding models though, at least on the animation authoring scenes. So that’s a slight improvement. ** ** I am really surprised something as potentially useful as referenced models remain so broken after so many years. Is there some trick to making it work? External deltas perhaps? I’m going to give up again on these unless I hear of something that might work. ** ** Out of curiosity, does max or maya have truly functioning refmodels? Or any app? ** ** ** ** ** -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Toggle viewport zoom lock from script when rotoscope
Hi, Anyone knows if it's possible to set the viewport zoom lock from script. It seems that that flag is per camera becouse when I click that zoom icon from one viewport it also affect other viewport zoom lock. p.s. what I mean by zoom lock it the icon that appears next to the X Y Z on each view when rotoscope is enabled. -- Edy Susanto Lim TD http://sawamura.neorack.com -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
Re: Toggle viewport zoom lock from script when rotoscope
SOLVED bummer found it 2 minutes after posting the question.. so it was the subpixelzoom parameter of the Camera object edy On Fri, Sep 20, 2013 at 12:18 PM, Edy Susanto Lim edysusant...@gmail.comwrote: Hi, Anyone knows if it's possible to set the viewport zoom lock from script. It seems that that flag is per camera becouse when I click that zoom icon from one viewport it also affect other viewport zoom lock. p.s. what I mean by zoom lock it the icon that appears next to the X Y Z on each view when rotoscope is enabled. -- Edy Susanto Lim TD http://sawamura.neorack.com -- Edy Susanto Lim TD http://sawamura.neorack.com -- To unsubscribe: mail softimage-requ...@listproc.autodesk.com with subject unsubscribe and reply to the confirmation email.
RE: 2014
Hi Raffaele, I appreciate your response and since you are saying they work I will take that at face value and I apologize for jumping the gun by saying they are broken. That being said, there are clearly usability issues that make them hard to use effectively and this has left a lot of people thinking they are broken (it wasn't that long ago they were implicated in crashes on a regular basis, too). I'm not about to advocate dumbing down the ref model system; I'd much rather learn the tricks and make it work for me. Unfortunately the docs don't explain much about how to work through problems when ref models don't work as expected. We really don't have many clues to work with when things don't work as we think it should - things just stop working as previously configured without any indication beyond the visual results which can easy to overlook. If I could make one suggestion, some error messages when a scene is loaded and deltas can't be applied for some reason would be really helpful. Here's a practical example of what I'm trying to do, it seems pretty simple and maybe there is a simple explanation of what I did wrong and how to correct it that would make the process more clear. I have a scene two top level models: a local model of a weapon (a rifle), and a referenced model of a biped rig. The biped rig has one nested model inside of it which contains the mesh of the game character and the null hierarchy it is enveloped to. In the biped rig scene that creates the biped rig EMDL, the mesh model deformers are constrained to the biped rig. Back to the weapon scene: the weapon model was created first, and then the biped rig (and in turn it's nested mesh model for the character) was imported as a referenced model. I then constrained the hands of the biped rig to the weapon so that moving the weapon around would cause the biped rig to track it. Looks good so far, saved it, loaded it and tried moving the weapon around. The constraints were no longer active. They are visible in the delta, but not doing anything. Any idea what might make those hand constraints work in this scenario? This seems pretty straightforward, but it's not working and I have no idea why. It seems like the order of model creation is correct, assuming that's relevant here for dependencies which I'm not sure is the case here. One thing I want to point out is I expect that the use of nested models here is a non-issue; I don't actually manipulate the mesh model at all here, I'm only adding constraints and position deltas to the biped rig at the topmost level. The mesh is basically a passive preview of the game character, not being directly manipulated at all although it does have a StoredPositions delta due to the hands of the biped rig moving onto the weapon. Despite having used Softimage for several years now (mostly on the technical/api/custom exporter side of things) I always feel like I have a lot to learn and I appreciate any suggestions you might have. I wouldn't be surprised if I completely overlooked something dumb, but I reproduced the same exact steps except one which was to convert the biped rig to a local model and that scene loaded with the constraints intact and active. Thanks for any insights, Eric From: softimage-boun...@listproc.autodesk.com [mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of Raffaele Fragapane Sent: Thursday, September 19, 2013 8:45 PM To: softimage@listproc.autodesk.com Subject: Re: 2014 They aren't broken, referencing references is a million miles away from a trivial problem. The key to it is order of execution, which is neigh impossible to guarantee without responsabilizing users well beyond what your average seat warmer can deal with from a technical stand point. The way Soft deals with name spaces and hierarchies and referencing within models when you have nested models you assume an order of operations that is simply not guaranteed. So it's perfectly possible, in example, that a model depending on its father receives a namespace and before the parent does. When you factor in deltas and cross dependencies the problem becomes even worse. It's simply an incredibly hard problem to solve generically in the context of a pipeline (end user's) which the provider client (Soft) can't have any notion of. We and others spend untold amount of hours to develop tools and procedures to allow cross model connections, and they are a brittle, nasty thing to deal with, and it only works because we're extremely well serviced on the asset dynamicism and management side of the pipe, and everything users do that establishes those cross namespace connections is managed by wrappers that also take care of establishing order of stashes and stacking within the stash. Offering the same thing arbitrarily in a DCC client would be exceedingly hard, or would greatly restrict what you could do with those namespaces and referencing. This isn't to say
