Search found 50 matches

by Dietger
Fri Jun 07, 2013 9:33 pm
Forum: General Development
Topic: Questions about adjoint bsdf
Replies: 3
Views: 3971

Re: Questions about adjoint bsdf

1. Yes, that sounds about right. Using adjoint bsdf's is just a view on bidirectional light transport that cleans up the math. You may freely ignore them and always evaluate the bsdf in the same consistent direction. The catch is in the probability densities; When tracing in the opposite direction, ...
by Dietger
Fri May 24, 2013 5:09 pm
Forum: General Development
Topic: Shadows & shading normals
Replies: 8
Views: 7270

Re: Shadows & shading normals

One moronic idea from my side: For interpolated shading normals and bump maps you could compute a virtual shading position above/below the actual surface and use this for tracing shading rays. It might fix the issue at hand, at the cost of opening Pandora's box of rendering artifacts. Of the top of ...
by Dietger
Wed May 22, 2013 9:04 am
Forum: General Development
Topic: Shadows & shading normals
Replies: 8
Views: 7270

Re: Shadows & shading normals

Ah, sorry, you are right, I skipped over the fact that you where using point lights. Well, you already answered your own question there :) either don't use point lights or tessellate your geometry (or better, both). In general, this is a hard problem to solve. Shading normals are a hack, so expect t...
by Dietger
Tue May 21, 2013 4:56 pm
Forum: General Development
Topic: Shadows & shading normals
Replies: 8
Views: 7270

Re: Shadows & shading normals

It looks like you mixed up shading and geometric normals in your sampling and evaluation. Make sure that all your quantities w.r.t projected solid angle (radiance, sampling densities) are using the geometric normal and not the shading normal for projection. If you mix those up, you get artifacts the...
by Dietger
Tue Nov 06, 2012 9:25 pm
Forum: ompf2
Topic: IP BAN TESSELLATED TITS?
Replies: 5
Views: 9355

Re: IP BAN TESSELLATED TITS?

I trust in the infallible wisdom of our beloved Dr. dictator. The mere act of casting a vote is an act of treason!
by Dietger
Sun Nov 04, 2012 11:35 am
Forum: General Development
Topic: Heightmap Stepping
Replies: 3
Views: 3859

Re: Heightmap Stepping

I haven't read that particular paper, but if you use some form of interval arithmetic on your highfield shader/texture to compute a hierarchy of bounding boxes (either on the fly or as pre-process), then the surface does not need to be planar. You should express the curved surface using interval ari...
by Dietger
Wed Oct 31, 2012 12:23 pm
Forum: General Development
Topic: GPU/CPU floating point accuracy
Replies: 7
Views: 4683

Re: GPU/CPU floating point accuracy

I have seen some precision issues in bdpt before, but none that couldn't be fixed. For example, similar to what ingenious said, when computing MIS weights you should cancel out all common factors in the numerator and denominator (which you should do anyway as it is cheaper to compute). Another commo...
by Dietger
Sat Oct 13, 2012 3:33 pm
Forum: Links & Papers
Topic: Vertex merging paper and SmallVCM
Replies: 33
Views: 29579

Re: Vertex merging paper and SmallVCM

Let me try to clarify what I meant here. Of course you are right. The initial noise will be less, the added bias will go away and the algorithm does inherit BDPT's high asymptotic performance. But in practice nobody waits that long! If you would be willing to wait that long, you might as well have s...
by Dietger
Fri Oct 12, 2012 2:09 pm
Forum: Links & Papers
Topic: Vertex merging paper and SmallVCM
Replies: 33
Views: 29579

Re: Vertex merging paper and SmallVCM

Thanks, very interesting stuff! In terms of PPM, does anyone have an idea how one can estimate the most appropriate kernel radius for area density approximation without manually fine tuning it? :D That is indeed a good question. From my personal experience PPM convergence and quality is mostly deter...
by Dietger
Mon Oct 08, 2012 3:01 pm
Forum: General Development
Topic: PMC-ERPT
Replies: 1
Views: 2764

Re: PMC-ERPT

The paper couldn't convince me that PMC-ERPT is unbiased either (or even consistent for that matter). I recently reread the PMC sources referenced in the paper and I still don't see how the authors got from there to the presented PMC-ERPT algorithm. Aside from the lack of proofs, the authors skip ov...