http://popo894.friendlinkup.com/ Umut ERTURK en http://popo894.friendlinkup.com/2008/09/09/cuda-and-ray-tracing-are-they-really-a-good-combination.html Previous week I mailed to people who worked on ray tracing with CUDA architecture as well as I received some disheartening mails. I would like to share these mails since you may require to understand reasons on behalf of the conflict between NVidia’s CUDA 1.1 architecture as well as ray tracing. I sent this mail to “the secret matador” whom I found in NVidia’s CUDA forums; > umut has sent you this email from > http://forums.nvidia.com/index.php. > > > Hello > I am studying my master degree on computer games > tech as well as I’m planning to opt on behalf of ray tracing on cuda > processors in dynamic scenes on behalf of my master thesis, > however, as i read from your post itz really not a > good idea. Is it? I am also concerned about the > number of replies to your post (no reply as i see). > I would like to request you if it is a good idea to do > my master thesis on this topic or would you like to > suggest some different topics about CUDA? > I would really appreciate your answer. > Thanks > > UMUT > umuter { @t } gm@il.com > Well, I was trying to perform raytracing with CUDA also some time ago. My scenes contained triangles. If you desire to utilize CUDA on behalf of just spheres I think can be as fast as Cell ( see http://eric_rollins.home.mindspring.com/ray/cuda.html ) Well, on behalf of xNormal I tryed CUDA with triangle scenes(1-10M triangles) as well as I wasn’t able to get it working at decent speed. Perhaps if you utilize smaller scenes could be ok… but was terrible slow on behalf of me… I think my scenes were too big to fit well in the shared/const memory. I used BVHs, grids as well as stackless kdtrees without any good result. A dual-core was always faster. Btw, the CUDA stackless approach sux a lot! CUDA is very good processing linear data(like an image kernel) but terrible processing random and recursive structures like the ones that requires raytracing. Another problem I found was that there's no CUDA x64 version, neither is Vista compatible as well as the VRAM on desktop G80 cards was too small on behalf of me(256-512Mb is the typical as well as my scenes were occuping an estimated 1Gb). The CUDA 1.1 does not support 3D textures so is a pain to get the uniform grids working… Well, I hope you could improve my results… CUDA was not very good on behalf of me. I hope they add a function stack, better syncronization mechanism as well as improved registers/instruction count on behalf of CUDA 2. After getting this mail I requested a similar question to Eric Rollins who is an estimated all probably one of the first people trying to implement ray tracing algorithm on CUDA. And I also requested if PS3 or CUDA? which one should I opt on behalf of as well as the answer is pretty clear; PS3. Eric Rollin’s complete answer: Umut; I think you are correct about the difficulties with ray tracing on CUDA. It is easier on the Cell/PS3, though still challenging. See papers linked on my PS3 ray tracing page, as well as MIT blue steel: http://cag.csail.mit.edu/ps3/blue-steel.shtml In my code on behalf of CUDA as well as Cell/PS3 the only primitive I have implemented is the sphere. I assume you have already seen this paper on alternative approaches: http://graphics.stanford.edu/papers/i3dkdtree/gpu-kd-i3d.pdf . Also some of the latency avoidance techniques discussed on behalf of Cell/PS3 might apply to CUDA. I do recommend reading the papers on behalf of Cell/PS3 even if you desire to endeavour CUDA. If you have a big insight on how to do CUDA, go with it. Otherwise I recommend Cell/PS3. Good luck. Tue, 09 Sep 2008 23:50:16 -0400 popo894