Open SubDiv - Pixar


Sorry, read wrong ;)


But sounds nice, that even big studios like opensource. Or maybe they expect a community to improve it?
 
Last edited by a moderator:
What they expect IMHO is more people using their proprietary softs instead of trainning in themselves, which is expensive and time consuming.


Lightwave 3d has been doing that ´subdivisions´ for ages now and the past few versions currently can display FULL final render is real time, with everything in. Can also animate such.


From the point of view of 3d movie production, I see nothing new here, just revamping the old things as new, nothing ever close to VPR or Fprime, in which I can´t work without anymore for the last 4 years.


From the game point of view, it could be something cool to have going.
 
What they expect IMHO is more people using their proprietary softs instead of trainning in themselves, which is expensive and time consuming.
Lightwave 3d has been doing that ´subdivisions´ for ages now and the past few versions currently can display FULL final render is real time, with everything in. Can also animate such.[…]
Except that open subdiv is completely open source and free from license.


True, Lightwave 3D has been doing that for quite a while, but the code used to do it isn't open sourced.


And, yep a 3D game engine with open subdiv implemented (without as many subdiv of course) that would mean with pretty low poly models, you would still have awesome smooth renders if your hardware is powerful enough. And since the source code is available it's not a complete fantasy.
 
Last edited by a moderator:
Maybe this could be also improve the subdiv routines in Blender? Hey are not bad but also not perfect. At least they not as bad as Blenders "bevel" routiness. :D
 
Lightwave 3d is one of the first 3d apps to use and or develope upon sub-d's noted.. it even includes Catmull-Clarks subd calculations, which offers far smoother results than the standard sub-d, but it does comes with the cost of longer render times and reduced refresh rates. Regardless sub-d has proven the method of choice over nurbs for film work (but certainly not CAD). PIXARs rewriting and optimization of Ed catmull's original code is simply amazing and an industry changer for sure. Its possible, much of opensubdiv's speed is because it routed thru the GPU and not the CPU. completely freeing up the CPU and thus the software itself. nevertheless, this is ground breaking in many ways. To freely work with high resolution mesh in modeling texturing and animation is outstanding, and has to be adopted by all 3d platforms including lightwave3d. Mudbox had already broke ground utilizing the gpu that enabled near realtime functionality with high resolution models. And still this is groundbreaking. Ide love to see this tech implemented into LW including PTEX. I rarely ever use UV's and PTEX has been developed to finally put an end to UV texturing. I think No one could develope a solid argument for keeping the UV method of texturing around, when Ptex gets the job done outright error free with no pre work neccessary.
 
Last edited by a moderator:
Back
Top