Search found 86 matches

by friedlinguini
Tue Mar 26, 2019 1:33 pm
Forum: General Development
Topic: Path tracing in games
Replies: 23
Views: 16433

Re: Path tracing in games

I think the hope is that it will follow roughly the same trajectory as programmable shading, which was also once an extravagant feature from the offline world. Unlike with Cg, I think there’s a bit more of a push for multi-vendor adoption of the interface as seen from getting it in DirectX early, al...
by friedlinguini
Mon Mar 04, 2019 1:09 pm
Forum: ompf2
Topic: Moving ompf2 to a new host
Replies: 6
Views: 358

Re: Moving ompf2 to a new host

Thanks for this, Jacco, and for all of your other hard work on the forum!
by friedlinguini
Sun Feb 10, 2019 2:51 pm
Forum: General Development
Topic: Performance issues with Wald&havran's nlogn kdtree construction
Replies: 5
Views: 523

Re: Performance issues with Wald&havran's nlogn kdtree construction

Calculating a median can be O(n) (see std::nth_element, median-of-median algorithm, etc), so the overall algorithm I described should run in O(n log n).

If you’re using a language that doesn’t have a linear median/partition library function, then that can be a bit of a rabbit hole.
by friedlinguini
Fri Feb 08, 2019 2:01 am
Forum: General Development
Topic: Performance issues with Wald&havran's nlogn kdtree construction
Replies: 5
Views: 523

Re: Performance issues with Wald&havran's nlogn kdtree construction

Seems like unnecessary complexity to use a SAH construction algorithm just to organize a point cloud. Especially if the heuristic is just a median split. That means you don't need to do a plane sweep or a full sort. I'd also get rid of mixing component values from different dimensions. As you've not...
by friedlinguini
Thu Sep 20, 2018 3:19 pm
Forum: Links & Papers
Topic: CUDA 10 available
Replies: 6
Views: 9498

Re: CUDA 10 available

I don't think CUDA itself has been advertised with ray-tracing goodness; it's all delegated to OptiX, DirectX, and Vulkan.
by friedlinguini
Wed Jul 25, 2018 3:24 am
Forum: Links & Papers
Topic: Primary Sample Space Path Guiding
Replies: 3
Views: 1738

Re: Primary Sample Space Path Guiding

It seemed to have a lot of the same ideas as multidimensional adaptive sampling (https://www.ci.i.u-tokyo.ac.jp/~hachisuka/mdas.pdf), and also the same weaknesses.
by friedlinguini
Mon Jul 23, 2018 1:44 pm
Forum: General Development
Topic: Emissive surfaces and visibility term
Replies: 6
Views: 2490

Re: Emissive surfaces and visibility term

I'd say no. Without explicit light source sampling, all geometric factors relating to the light source--surface area, distance, and cosine term--are accounted for in the probability of a ray hitting the light source "by accident." For example, if the geometric normal of a triangular light source is ...
by friedlinguini
Thu Jun 15, 2017 1:59 pm
Forum: Visuals
Topic: My First Spectral rendering
Replies: 5
Views: 6981

Re: My First Spectral rendering

Whoah, weird. I had thought you could do something very similar, but by adding a random point on the *interior* of a sphere to the normal, rather than on the surface: http://psgraphics.blogspot.com/2014/09/random-diffuse-rays.html. Obviously it would be better to use exactly two random parameters th...
by friedlinguini
Mon Apr 17, 2017 10:30 pm
Forum: Tools, Demos & Sources
Topic: Implementation for Blue-noise Dithered Sampling
Replies: 26
Views: 19975

Re: Implementation for Blue-noise Dithered Sampling

The idea ideas not adding a deterministic (blue-noise) number to a random number. In fact, the idea is to get rid of the randomness altogether - you use the deterministic number to post your pattern, instead of using a random number. Right, I was using "random" loosely. My understanding of this alg...
by friedlinguini
Sun Apr 16, 2017 8:34 pm
Forum: Tools, Demos & Sources
Topic: Implementation for Blue-noise Dithered Sampling
Replies: 26
Views: 19975

Re: Implementation for Blue-noise Dithered Sampling

sin(2 * pi * (ps - qs)) * 0.5 + 0.5
OK, that makes no sense. I tried to jam a couple of ideas together without making sure that one didn't break the other.

Maybe c * (x - x * x), where x = ps - qs and c is some constant I haven't worked out yet.