I've spent the night figuring out how to set per-component ICE attributes
from a custom operator, which most of the threads on the subject had me
beginning to believe was impossible.
I was working with SI2013 and compiling against the SDK that ships with
2014sp2, and as soon as I declared a
maybe it's too early in the morning for me, but care to share how to achieve
the modulo/gradient effect with example images? or a step by step...
always hated the fact that we cant have more than 8 markers in a gradient!
_
From: softimage-boun...@listproc.autodesk.com
A couple of years ago i was doing some volumic lights in arnold and noticed
some artifacts in the render region when using colour management, which turned
out to be as a result of the region being 8 bit. I notice now with 2014sp2 that
i'm getting the same thing with mr and a standard volumic
From the Softimage 2014 SP2 Readme:
SOFT-9089
Very slow access to the ICE attribute arrays
*Note:
You may need to recompile your plugins, if they were compiled using
Softimage versions prior to 2014 SP2.
On Wed, Oct 23, 2013 at 4:33 AM, Mathias N mdawn...@gmail.com wrote:
I've spent the
http://www.deskriptiv.de/form
Cool stuff, I believe there was an article about them in one of the latest
Digital Production issues.
http://www.deskriptiv.de/form
--
---
Stefan Kubicek
---
keyvis digital imagery
I love this kind of stuff and Tim Borgmann is the king of kewl motion
design in my book...along with polynoid.
*Greg Punchatz*
*Sr. Creative Director*
Janimation
214.823.7760
www.janimation.com http://www.janimation.com
On
Any idea what software ?
Houdini maybe ?
I find myself always fighting to get a beautifull noise / Turbulence...
Le 23/10/2013 15:51, Stefan Kubicek a écrit :
Cool stuff, I believe there was an article about them in one of the
latest Digital Production issues.
http://www.deskriptiv.de/form
Wow, very cool work.. Can someone translate the titles?
Eric
Freelance 3D and VFX animator
http://vimeopro.com/user7979713/3d-work
On Wed, Oct 23, 2013 at 9:51 AM, Stefan Kubicek s...@tidbit-images.comwrote:
Cool stuff, I believe there was an article about them in one of the latest
Digital
lovely reference, thanks for sharing!
On 23 October 2013 15:09, Eric Lampi ericla...@gmail.com wrote:
Wow, very cool work.. Can someone translate the titles?
Eric
Freelance 3D and VFX animator
http://vimeopro.com/user7979713/3d-work
On Wed, Oct 23, 2013 at 9:51 AM, Stefan Kubicek
i did this for a lucozade job where a dancer was emitting stuff based on his
motion. I used a duplicate frame with a one frame offset to get a basic
velocity and just set the weightmap value with that, using a rescale node and a
scalar curve node to give a bit more control.
A
I'm not sure if all the images are the work of Christopher Bader but I know
he uses Houdini.
Lovely indeed! Also check out www.field.io
On Wed, Oct 23, 2013 at 10:23 AM, Cristobal Infante cgc...@gmail.comwrote:
lovely reference, thanks for sharing!
On 23 October 2013 15:09, Eric Lampi
I just tested with one of my gear chars.
Global-Scale does break the spine but only when pushed past 3.
Easy to overlook while testing because the range goes from 0 to 3 on that
slider.
Manny Papamanos
Product Support Specialist
Softimage and Motionbuilder
From:
For me it was scaling down which caused the problems with the fingers -
take the default man in the example scenes, scale him down to 0.1 or lower,
and you will see that the fingers are mis-aligned. The same effect happened
to a character that I scaled up too, though I can't replicate that here.
No need to hang yourself Olivier, this is for sure really interesting
stuff ;) Thanks for the link.
Cheers
Tim
http://www.deskriptiv.de/form
I confirm there is a scaling issue in both the Meta and the Spine
components.
I never had the time to fix it. Sorry about that. :(
If anyone is fixing it (some people emailled me a while ago with ideas on
how to fix the meta). I'm sure Miquel will be happy to add it to the next
release of GearMc.
I think they are houdini? I've been making something similar recently
rotating strands around and what not, hopefully I will show soon!
Simon Reeves
London, UK
*si...@simonreeves.com*
*www.simonreeves.com*
*
*
On 23 October 2013 16:39, Tim Borgmann i...@bt-3d.de wrote:
No need to hang
I didn't have time to check it :P but I will check ASAP. But if some one
have the fix before me, I will be more than happy to add it to GearMc repo
:D Thanks!
Miquel Campos
www.miquelTD.com
On Wed, Oct 23, 2013 at 11:43 AM, Jeremie
if you are emitting particles on the mesh, getting point velocity of the
emit location is easy. no need to duplicate any geo or do frame offsets.
On Wed, Oct 23, 2013 at 7:54 AM, Andi Farhall hack...@outlook.com wrote:
i did this for a lucozade job where a dancer was emitting stuff based on
would that be the case with a basic cached character mesh? It was a few years
ago but i'm sure i ran into trouble. Or maybe i just assumed i would and did it
that way.
A
...
unsubscibe
Regards,
Robert Lansdale, CTO Product Manager
Okino Computer Graphics.
---
1515 Britannia Road, East. Mississauga, Ontario. L4W 4K1.
Tel: 888-3D-OKINO, (905) 672-9328. http://www.okino.com
if the particles are simulated, it should still work.
On Wed, Oct 23, 2013 at 10:23 AM, Andi Farhall hack...@outlook.com wrote:
would that be the case with a basic cached character mesh? It was a few
years ago but i'm sure i ran into trouble. Or maybe i just assumed i would
and did it that
to a degree, you can get around that, by plugging a gradient into a gradient
(ad infinitum).
so if you have a simple B/W gradient at first, plug it into one which goes from
BWBWBWBW again into one which goes from BWBWBWBW you will end up
with some very fine detail high contrast lines. Chain as
If you emit from geometry, then stick the particles to the emit
location, you do get a pointvelocity from the emitlocation. This works
for enveloped or cached or just transforming.
S.
On 2013/10/23 7:35 PM, Steven Caron wrote:
if the particles are simulated, it should still work.
On Wed,
The plugin loads fine in 2014 SP2 even when compiled against the 2013SP1
SDK; it is not forward but backward compatibility that appears to be broken.
Although the only proper rundown of SDK backwards compatibility I can find
is from the 2011 SDK wiki, it states that The Softimage SDK is backwards
exactly...
On Wed, Oct 23, 2013 at 10:40 AM, Sandy Sutherland
sandy.mailli...@gmail.com wrote:
If you emit from geometry, then stick the particles to the emit
location, you do get a pointvelocity from the emitlocation. This works for
enveloped or cached or just transforming.
S.
On
The Softimage SDK is backwards compatible: a newer version of Softimage
can load plugins that were compiled with older versions of the SDK.
In general, that backwards compatiblity is rarely broken. 2014 SP2 was an
exception (SItoA, for example, would not load into 2014 SP2 until it was
recompiled
well not if they changed its behavior... and you ran into the a combination
which isn't supported.
i usually go the other way around, compile with the minimum SDK version you
can. is there any reason you can't compile with the 2013 SDK for use in
2013?
again, in this case if you want to use your
Huh, so I guess I should stop using the 2014SP2 SDK for my 2011 plugins.
Thanks for clearing that up.
On Wed, Oct 23, 2013 at 7:50 PM, Alok Gandhi alok.gan...@modusfx.comwrote:
On 10/23/2013 1:40 PM, Mathias N wrote:
The plugin loads fine in 2014 SP2 even when compiled against the 2013SP1
ya, what advantage where you hoping to attain by using the newer SDK?
On Wed, Oct 23, 2013 at 10:54 AM, Mathias N mdawn...@gmail.com wrote:
Huh, so I guess I should stop using the 2014SP2 SDK for my 2011 plugins.
Thanks for clearing that up.
I made the move to the 2014SP2 SDK a while back hoping that some of the
problems I was encountering were bugs rather than me simply having no idea
what I was doing (Spoiler: it was the latter). Seeing as all my plugins
compiled against it also worked fine in 2011 I figured it would be
preferable
could someone explain the “auto-spreading”, I have noticed this undesired (for
me) effect on the fingers when the arms are in certain positions, however I
don’t know what caused it or why.
From: Matt Morris
Sent: Tuesday, October 22, 2013 9:15 PM
To: softimage@listproc.autodesk.com
Subject:
Usually as Steven said, I would run older plugins in 2014 as is first.
If they seem to fail I would check what is going on.
The good workflow would be:
Always start with the SDK for the Soft version. So if I am writing a
plugin for 2011, I would use 2011 SDK. Then in 2012 and up I would load
the real shame is, that it was deceiving you... you got none of the
advantages of the changes made to the SDK. from what i understand (which i
am still a c++ n00b) it loads the sicppsdk.dll dynamically from the version
of softimage which is loading the plugin and that library doesn't have the
new
I'm stuck back using Mental Ray on a project and on 1 particular frame I'm
getting this:
// ERROR : FATAL: BSP2 0.4 39 MB fatal 551109: runtime exception
0xc0fd: stack overflow
// ERROR : FATAL: BSP2 0.4 39 MB fatal 551109: runtime exception
0xc0fd: stack overflow
// ERROR :
Straight render from Softimage, not a preview/region render. Though I
think I'd get the same if I tried those as well.
I'm searching through all my textures for the scene now - there are a lot
of animated ones, and I'm wondering of one of those are corrupt.
-Paul
On Wed, Oct 23, 2013 at 3:37
Is this coming from render region, preview, or render(interface)?
Or is it from xsibatch render?
Have you tried using the older bsp and changing the bsp settings?
Search for MI_FORCE_OLD_BSP in the documentation.
--
Joey Ponthieux
LaRC Information Technology Enhanced Services (LITES)
Mymic
BSP is an optimization algorithm. It divides up the scene into smaller boxes
in order to optimize the raytracing speed. When BSP2 was introduced it was
apparently further optimized to set automatically to what it thinks is the best
settings internally. I'd try changing the render type to
I think at that particular frame I have an object with it's render
visibility animated from off to on.
I'll give batch a try see what happens. I've tried rendering with all
diagnostics turned on and I can't see anything else listed that it's having
a problem with.
-Paul
On Wed, Oct 23, 2013
You need to tweak the BSP settings as the error indicates the BSP tree is
subdividing too much causing it to exhaust available memory on the stack.
Usually occurs where there is an error of high geometric density (lots of
triangles in a particular area), but there are other causes. One
Animating render vis may be interfering with things like motion-blur in some
negative way...
Just set the visibility to 'on' for that frame (remove the keys) and re-render
it.
-manny
From: softimage-boun...@listproc.autodesk.com
[mailto:softimage-boun...@listproc.autodesk.com] On Behalf Of
A couple of years ago I decided I needed inter-colliding strands and spent
an ungodly amount of time trying to make it in ICE.
As a result of not knowing anything back when I started working on it the
code is a complete mess, and fixing the more glaring issues would be a
significant undertaking.
Outstanding work sir! :)
Can't wait to have time to play with this.
On Wed, Oct 23, 2013 at 10:56 PM, Mathias N mdawn...@gmail.com wrote:
A couple of years ago I decided I needed inter-colliding strands and spent
an ungodly amount of time trying to make it in ICE.
As a result of not
Honestly that's pretty phenomenal. No reason to downplay, it's not a
trivial problem to solve.
On Wed, Oct 23, 2013 at 10:56 PM, Mathias N mdawn...@gmail.com wrote:
A couple of years ago I decided I needed inter-colliding strands and spent
an ungodly amount of time trying to make it in ICE.
44 matches
Mail list logo
