Search found 24 matches

by Vilem Otte
Tue Apr 17, 2012 9:35 am
Forum: General Development
Topic: Csillámtrace - my new graphics engine
Replies: 153
Views: 100312

Re: Csillámtrace - my new graphics engine

The sky support is handled through sky-dome (testing ray against sphere), sky box (testing ray agains aabb), or just through "hack"?
by Vilem Otte
Sun Apr 15, 2012 2:33 am
Forum: Visuals, Tools, Demos & Sources
Topic: Brigade2 r2017
Replies: 2
Views: 4194

Re: Brigade2 r2017

:shock: That looks really awesome.

It's a bit shame that it doesn't work with AMD gpus (is there "just CPU" still avaible?).

And if you find some little time, could you please share some stats, so I'd see how much I'm slower (-> and grow my TODO list in optimization of my own code).
by Vilem Otte
Sun Apr 15, 2012 2:03 am
Forum: Visuals, Tools, Demos & Sources
Topic: Moderu
Replies: 0
Views: 2934

Moderu

Let me introduce my small "semestral" project here. It took roughly around 20 hours to complete it to current state and I still have about the same time before me. So what it is? I've been ordered to make a project in C#, so I took ray tracing (this time nothing high-performance like my OpenCL stuff...
by Vilem Otte
Sat Apr 14, 2012 3:44 pm
Forum: General Development
Topic: Csillámtrace - my new graphics engine
Replies: 153
Views: 100312

Re: Csillámtrace - my new graphics engine

and i also dont want to use/rely on any thirdparty code/codesnippets/solutions/principles I understand this, I also have my code all written from scratch ;) It gives better feeling about it, not mentioning when a bug is occured (then it can be a bit nightmarish when heavily using third party code)....
by Vilem Otte
Fri Apr 13, 2012 11:59 pm
Forum: General Development
Topic: Csillámtrace - my new graphics engine
Replies: 153
Views: 100312

Re: Csillámtrace - my new graphics engine

Just a little note... Arauna is still open-source and you can learn a lot from going through it ;) (I did, thanks jbikker - but still Arauna is faster (in Sponza a lot faster) ... "What kind of sorcery is this" :mrgreen: ). The image quality is heavily based on models/textures quality ;) You can try...
by Vilem Otte
Fri Mar 30, 2012 1:13 am
Forum: General Development
Topic: Nvidia limiting gtx680 compute speed?
Replies: 20
Views: 17452

Re: Nvidia limiting gtx680 compute speed?

Wonder how they actually get that high numbers (noticed they've got code, yummy ... gotta walk through it during the weekend). On GPU I still get "just" some 10 MRays/s (for primary) in naive (meaning like no packets or stuff) implementation of KDtree ray tracer on Radeon HD 6770 (not the best card,...
by Vilem Otte
Thu Mar 15, 2012 11:29 am
Forum: Visuals, Tools, Demos & Sources
Topic: Glitch Pictures anyone?
Replies: 81
Views: 111832

Re: Glitch Pictures anyone?

http://oi39.tinypic.com/20kxe2a.jpg Can't you see torus knot in a room colorized with barycentric coordinates? ... In fact, my attemp to quickly implement Wald triangle test on GPU :mrgreen: (needs some debugging to work though....) EDIT: Still not there, but getting closer (ended in another nice g...
by Vilem Otte
Sat Mar 10, 2012 3:26 am
Forum: General Development
Topic: Speeding up OpenCL BVH
Replies: 6
Views: 7316

Re: Speeding up OpenCL BVH

As I continue with this, I also moved a bit my KDTree implementation (meaning moved to GPU)... first results are (on HD6770) ... 3 - 4 MRays/s for Sponza (mono-tracing) ... Hooray (okay, now let's make this 30 - 40 MRays/s and I'll be more happy) Some stats - it took me 4 hours to code (with debuggi...
by Vilem Otte
Tue Mar 06, 2012 3:52 pm
Forum: General Development
Topic: Speeding up OpenCL BVH
Replies: 6
Views: 7316

Re: Speeding up OpenCL BVH

Thanks for the paper, this makes me reconsider making a new n-ary BVH builder ... I think this could actually also shine with "KD-trees" :P (making somehow effective N-trees). EDIT: Also a question, as I met functional programming more often in recent time - it shines in writing small code for trees...
by Vilem Otte
Thu Feb 23, 2012 11:44 pm
Forum: General Development
Topic: Speeding up OpenCL BVH
Replies: 6
Views: 7316

Re: Speeding up OpenCL BVH

I think I have QBVH paper lying here around in Documents, but I actually never got to actually implement it. From what I remember the building "how-to" in paper is: 1.) Build standard BVH tree with 4 triangles in node 2.) Flatten it, that every second row will be removed and thus parent will have 4 ...