Search found 138 matches

by shiqiu1105
Thu Nov 14, 2013 3:24 pm
Forum: General Development
Topic: About volumetric caustics.
Replies: 23
Views: 19052

Re: About volumetric caustics.

Guys, thanks for the help. I also found a project online here: http://www-graphics.stanford.edu/courses/cs348b-competition/cs348b-04/underwater/index.html which uses volume photon mapping and the result seems pretty good. I have never written a MLT before and I heard it's hard. I guess I will go wit...
by shiqiu1105
Thu Nov 14, 2013 3:19 pm
Forum: General Development
Topic: C++ AMP
Replies: 4
Views: 7215

Re: C++ AMP

Also curious~ Thanks for asking!
by shiqiu1105
Wed Nov 13, 2013 5:08 pm
Forum: General Development
Topic: About volumetric caustics.
Replies: 23
Views: 19052

About volumetric caustics.

Hi, I am trying to render an underwater scene, where volumetric caustics due to the water surface dominates. What is the best approach for this kinda of effect? I already have a Bidiretional path tracing running in my ray tracer, but some related projects that I found online uses volume photon map. ...
by shiqiu1105
Thu Aug 01, 2013 11:48 pm
Forum: General Development
Topic: How to flatten a binary bvh to 4ary BVH?
Replies: 4
Views: 3761

Re: How to flatten a binary bvh to 4ary BVH?

I suspect that it means you should discard non-leaf nodes in even layers. The idea is that in a binary tree each node (say, at level 1) has up to two children (level 2), and thus up to four grandchildren (level 3). If you can connect the level 1 node to its level 3 grandchildren you don't need the ...
by shiqiu1105
Thu Aug 01, 2013 10:41 pm
Forum: General Development
Topic: How to flatten a binary bvh to 4ary BVH?
Replies: 4
Views: 3761

How to flatten a binary bvh to 4ary BVH?

I have been trying to implement this ftp://134.169.37.4/pub/public/people/eisemann/Students/JakobGarbe/RayTracing/Dammertz08ShallowBVH.pdf It is not clear on how to flatten a normal bvh into a qBVH. It says discard nodes in the even layers, but what if the even layer is leaf node? We simply create 2...
by shiqiu1105
Tue Jul 02, 2013 7:39 pm
Forum: General Development
Topic: About Timo's speculative traversal
Replies: 0
Views: 2644

About Timo's speculative traversal

I have been reading Timo's paper: understanding-the-efficiency-of-ray-traversal-on-gpus In which it mentioned that speculative traversal will help the performance. I read that paragraph several times and still don't know what's doing. Is it keeping all the rays in wavefront? or simply keeping tracin...
by shiqiu1105
Tue Jul 02, 2013 2:47 pm
Forum: General Development
Topic: OpenCL traversal kernel
Replies: 8
Views: 7189

Re: OpenCL traversal kernel

Where can I find the paper on stack less bvh?
by shiqiu1105
Fri Jun 21, 2013 10:31 pm
Forum: General Development
Topic: Is it still worth it to do CPU SIMD ray tracing?
Replies: 12
Views: 11646

Re: Is it still worth it to do CPU SIMD ray tracing?

Yes, it still makes sense. Have a look at Intel Embree for example. It uses SSE to speed up bvh traversal although it does not do packet tracing. I don't have recent cpu hardware to test on, but I suspect it could outperform a single gpu for medium to complex scenes on a Ivy Bridge or Haswell cpu. ...
by shiqiu1105
Fri Jun 21, 2013 8:41 pm
Forum: General Development
Topic: Is it still worth it to do CPU SIMD ray tracing?
Replies: 12
Views: 11646

Is it still worth it to do CPU SIMD ray tracing?

Recently I have been reading about doing ray tracing with SSE intrinsics. And I felt that it's really troublesome to pack data into ray packets and use SSE. Now that we have GPU, is it still worth doing it?? Timo Alia has their fastest ray tracer written in CUDA, right? Do they also pack their ray t...
by shiqiu1105
Mon Apr 15, 2013 9:18 pm
Forum: General Development
Topic: Why is tokaspt so freaking fast!?
Replies: 12
Views: 8274

Re: Why is tokaspt so freaking fast!?

Or did you use some trick on what size of your kernel is? Mine is simply a 16x16 = 256 threads per block, and 640 / 16 = 40 blocks in the whole launch. I use the workgroup size suggested by the driver after the kernel compilation (I have to check but it is likely to be 256) and <image width>x<image...