Search found 6 matches

by jbarcz1
Tue Oct 23, 2012 5:37 am
Forum: General Development
Topic: AABB for clipped triangle
Replies: 8
Views: 5607

Re: AABB for clipped triangle

For the TinyRT KDTree builder, I did it by clipping only to one split plane at a time, then intersecting each half with the clipped box from the previous split. If you incrementally compute set intersections on AABBs, this keeps a tight bound all the way down the tree. If you just want a generic fun...
by jbarcz1
Wed Feb 22, 2012 3:31 pm
Forum: General Development
Topic: Total SAH of a BVH
Replies: 11
Views: 11196

Re: Total SAH of a BVH

For a BVH, you can actually expand the recurrance into something like: sum( Ci * Ai/Aroot ) Where Ci is the cost of node i (Ct for an inner node, sum of primtiives for a leaf), and Ai is its area. (I may have it wrong above, I worked it out and wrote it down, but I dont have my notebook handy). The ...
by jbarcz1
Wed Feb 01, 2012 6:50 pm
Forum: General Development
Topic: Path tracing in games
Replies: 20
Views: 16260

Re: Path tracing in games

Just read that paper too. Its pretty neat stuff. The thing is though, alias free hard shadows aren't really what you want. What you want, visually, is a convincing soft shadow (not a physically correct one, per se, just a fuzzy one that isn't visually jarring). One of the strengths that shadowmaps h...
by jbarcz1
Thu Jan 05, 2012 3:43 pm
Forum: General Development
Topic: Tangents computation problem
Replies: 11
Views: 9665

Re: Tangents computation problem

Thanks, In fact, you can see the tangent distribution on the image (the car). In the reality, the tangents are corrects, but suddenly switch of direction and so, by example when I use a diffuse shader (or any other tangent based shader) the same effect appear ! Can you post an image of a diffuse sh...
by jbarcz1
Tue Dec 27, 2011 5:35 pm
Forum: General Development
Topic: Sampling lights
Replies: 22
Views: 15207

Re: Sampling lights

spectral wrote:And how can you sample N lights with an array of N elements if you want to sample proportional to the PDF ?


Perhaps by allocating N*M spots, adding p*M copies of each light, then indexing with rand() % M ?
Neat trick, assuming I've guessed correctly :)
by jbarcz1
Sat Dec 24, 2011 5:44 pm
Forum: Considered Harmfull
Topic: Mem / core clock mystery
Replies: 4
Views: 9514

Re: Mem / core clock mystery

Ow sorry, completely forgot to mention. It's a GPU, Fermi, so it has cache. That could indeed be the problem. But I don't see how a cache could hide the mem underclock almost completely, in the context of path tracing? So, assuming your tools aren't lying about the memory clock, you must be getting...

Go to advanced search