ompf2.com

shocker_0x15: I don't understand, I thought in OpenCL you had to use an image2d_t type in order to get hardware filtering/sampling. How are you getting filtering using a global char*? szellmann: yes, for new hardware that makes the most sense. I have to support old hardware, so I've gone with the te...

How do you guys handle large numbers of textures on the GPU? You can either pass them all in as separate arguments with a huge switch/if statement, or you can pack them into image arrays or 3d textures (which wastes memory). Lastly, you could pack them into a large texture atlas, which may be the be...

New video: https://www.youtube.com/watch?v=zeLJRtsvGAk

You can keep up with development here: https://www.twitter.com/rove3d

Maybe the direct connection to the light requires a subtended solid angle from the connecting eye vertex to the emissive surface area for MIS, something that distinguishes that light vertex from the other light vertices. There was a point when I understood most of it, but it's fuzzy to me right now ...

I haven't read the paper, but I have implemented VCM. For example, SmallVCM creates light vertices at each bounce and does not create a light vertex on the light itself. It then casts eye rays and performs direct light sampling at each eye ray bounce. Maybe this is to reduce the overhead of storing ...

Price point between the devices is also a consideration, I'm not sure this is an apples-to-apples comparison. Power envelopes aside, how many GPUs can you buy for the price of a Knight's Landing?

Getting over 2.5B rays/s on a CPU is exciting though, but I agree we have to see the incoherent numbers.

Just a quick announcement, I'm releasing my interactive pathtracer for Unity on October 11. Unity isn't really known for being an environment for production rendering, but many CG pipelines are moving more towards these sort of engines. More information can be found on the Unity thread: https://foru...

When sampling a surface, every material type (conductor, dielectric, diffuse) has 2 layers. The top layer is a dielectric BSDF, so we can do things like clearcoats, varnish, etc. Light is either reflected, or transmitted towards the bottom layer. The bottom layer is the main material type. Attenuati...

Just posting a few shots of my pathtracer. The material system is a bi-layer, multi-scattering microfacet model, with anistropic roughness for all material types. The renderer is HDR, with the usual DOF, super-sampling, etc. You can keep up with my renderer here: http://www.twitter.com/rove3d http:/...

You mentioned that you use surface-local space to compute your microfacet BRDF. Here's an optimization for converting between the two spaces, this paper describes a faster way to calculate the basis from your surface normal. https://www.semanticscholar.org/paper/Building-an-Orthonormal-Basis-from-a-...