Recent content by jlippo

Really depends on scene. If camera cannot go near enough to objects or near plane cuts polygons before highres mipmap levels are visible, sure. Just having a texture with high dimensions on a game with low resolution output, doesn't mean every texel isn't contributing to the image. We had a...

Yup, if MSAA/VRS is used the TAAU should be made aware of it. Would be awesome to teach neural to use 4xMSAA buffer as an input, should give amazing results, even if it's quite low result for upscaling output. For VRS, 2 bits per subsample quad to indicate coarse shading rate should be enough...

It only affects edges, but it works on whole framebuffer without knowledge on edges.

It seems to be the case in at least games like rogue galaxy. Not sure if one can adjust mipbias etc. Aliasing it caused is quite horrible.

During early times, boards only did rasterization and texturing, so CPU did the vertex related workloads.

Idea is basically rendering wu antialiased lines at edges of polygons. This most likely increases polygon size with additional pixel at edges. Sorting polygons were needed as always with transparency. During times of software rasterization most games already sorted them, so it was not such a...

Nice. Really liked DC2 back in the days and would love a proper remake. Might have to try the first one, really do not like shortcomings of Ps1 rasterization, so this should be nice.

Surface-Stable Fractal Dithering (source etc.)

I wonder if old accumulation buffer methods would be good for rendering ground truth images. (Render scene n times and average result.) This allows custom sample patterns and if needed a motion blur and DoF. (Quite sure it's used in Gran Turismo camera mode.) AA is also seperate from post...

Has there been any talks on how games handle shading in RT? RT/PT has huge strain due shading all the hits, I would quess all sorts of caching and deferred methods could be quite usable in realtime RT. I know Manuka does shade non directional parts of shaders to vertexes of a tesselated...

I wonder how much of this is just PSSR being more reactive TAA solution to whatever developers have used before. We have seen problems mostly in games which have noisy input for TAA, perhaps PSSR was trained to handle changes as actual valid data, instead of smoothing over for several frames...

Outside shutter and lense stack, these should be almost identical, right? We trace paths that leaves outmost lense. Lense moves through space and we jitter outgoing paths in during shutter being open and within circle of confusion. For game purposes open shutter time shouldn't usually be more...

Yeah. I wonder when we will see first game with primary rays traced in a way to take account the DoF and motion blur. (Each ray with own time, location and direction when it leaves moving camera lense.) Things should become very interesting when ray reconstruction can take all this into account.

Looks like UE5 is finally getting decent transparency.

Will be interesting to see what kind of RTGI. Perhaps it does something similar to Metro, first bounce traced in simplified world with no shading and probes for rest. (Or just probes all atound.)