Search found 164 matches

by graphicsMan
Tue Nov 05, 2013 12:33 am
Forum: Visuals, Tools, Demos & Sources
Topic: in response to an assertion by mr john carmack who is wrong
Replies: 2
Views: 7382

Re: in response to an assertion by mr john carmack who is wr

Wow, neat, ridiculous, and hilarious all at the same time :) nice!
by graphicsMan
Mon Sep 16, 2013 1:40 am
Forum: Visuals, Tools, Demos & Sources
Topic: giga rays on intel phi
Replies: 24
Views: 27000

Re: giga rays on intel phi

Yeah, raw FLOPS definitely has little to do with how well a processor will compute specific tasks. I'll be interested to see how these two processors compete as you develop similar specialized kernels for the Titan platform. I think a key thing about Phi is that it has more memory than GPUs have, an...
by graphicsMan
Sat Sep 14, 2013 3:53 pm
Forum: Visuals, Tools, Demos & Sources
Topic: giga rays on intel phi
Replies: 24
Views: 27000

Re: giga rays on intel phi

yes titan (serveral) are in place. as i said above, both are more or less equal in performing diffuse bounces (around 165mray/s). for primary rays mic is 2x ahead. Strange, I would have expected exactly the opposite result :?: I mean, MIC should be less sensible to thread divergence. May be cache, ...
by graphicsMan
Fri Sep 13, 2013 5:49 pm
Forum: General Development
Topic: BVH instances support : approaches
Replies: 12
Views: 9968

Re: BVH instances support : approaches

BTW, if there are students interested in publishing a paper about how to solve this problem, I would be interested in helping out. I've been interested in this for a while, but I simply don't have the time to do a full implementation of my ideas. If you're interested in pursuing this, and would like...
by graphicsMan
Fri Sep 13, 2013 5:16 pm
Forum: Visuals, Tools, Demos & Sources
Topic: giga rays on intel phi
Replies: 24
Views: 27000

Re: giga rays on intel phi

Just curious... if you run 8 bounces in Sibenik Cathedral or something with even more occlusion/visibility complexity, what kind of ray throughput do you see? Any luck getting numbers for Titan yet?
by graphicsMan
Mon Sep 09, 2013 6:53 pm
Forum: Visuals, Tools, Demos & Sources
Topic: giga rays on intel phi
Replies: 24
Views: 27000

Re: giga rays on intel phi

256 eye rays or total rays?
by graphicsMan
Thu Sep 05, 2013 1:20 pm
Forum: Visuals, Tools, Demos & Sources
Topic: giga rays on intel phi
Replies: 24
Views: 27000

Re: giga rays on intel phi

Interesting. Do you have numbers for diffuse interreflections?
by graphicsMan
Fri Aug 30, 2013 10:37 pm
Forum: General Development
Topic: BVH instances support : approaches
Replies: 12
Views: 9968

Re: BVH instances support : approaches

I think that's pretty close to the usual approach I mentioned before. I think a lot of people use it, and it probably makes for very fast build times. I'm much more concerned with render time than build time because I may be tracing up to 100s of billions of rays for the lifetime of a single acceler...
by graphicsMan
Fri Aug 30, 2013 4:49 pm
Forum: General Development
Topic: BVH instances support : approaches
Replies: 12
Views: 9968

Re: BVH instances support : approaches

Hey. What I mean is this: Let's say you have a dozen instances of a chair, and they are simply sitting on the ground in a room. Their bounding boxes do not overlap. The traditional way of instancing is great, and there will be practically no slowdown from building the acceleration structure over the...
by graphicsMan
Thu Aug 29, 2013 3:36 pm
Forum: General Development
Topic: BVH instances support : approaches
Replies: 12
Views: 9968

Re: BVH instances support : approaches

I haven't been able to find a good source on this. I looked pretty extensively. Mostly I think people do the dumb obvious thing, which is to build BVHs over the mesh for the instances and then build a BVH of those.

I'd love to see research on good heuristics for instancing...