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
-
-
Except they didn't. Anything to do with indirect lighting and shadows in them is all hacks that glitch hideously.
2 replies 0 retweets 0 likes -
I mean, sure, but Monte Carlo path tracing as used in film is pretty much saying “we give up on trying to do anything smart, let’s just trace a shit ton of random light rays and throw giant render farms at the problem” :)
1 reply 0 retweets 0 likes -
Yeah. Well that's the only way to do it right.
2 replies 0 retweets 0 likes -
I dunno, I’m not so willing to give up. :) Reflective shadow maps with virtual point lights have been an interest of mine lately, for example…
1 reply 0 retweets 0 likes -
Replying to @pcwalton @RichFelker and
What bugs me about Monte Carlo simulations is that they aren’t how an artist would think about lighting. Artists think more in terms of VPLs (i.e. large light areas, not individual rays). The human brain has worked out a way to do global illumination more cheaply :)
1 reply 0 retweets 0 likes -
Oh, absolutely. I actually hate how everything in modern gaming graphics and animation-style CGI movies is so divorced from "artist" approach, and would love to see innovation in those areas.
1 reply 0 retweets 1 like -
Replying to @RichFelker @pcwalton and
But for VR/AR, and for not sticking out as ugly in live action movies, and for cool stuff like reproducing real optical illusions, lens optics, special materials, etc. you really need raytracing.
1 reply 0 retweets 2 likes
I think we’re basically in agreement; I just want to see more work on making raytracing approaches less brute force.
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.