The trajectory of real-time raytracing kind of mirrors is a microcosm of CS: by the time we got it, the traditional rasterization approaches got so good that raytracing seems hardly worth the trouble
-
-
I spent a few years of research doing fast hacky approximations to global illumination, and I've been won over by the "get the highest quality samples you can of the actual distribution and do heuristic reconstruction" approach. ML-denoising just enables complex heuristics.
-
Fair enough, I defer to your expertise :)
- 4 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.