Sunday, April 22, 2012

Independent Study: Week 7

Quick update to start the week off, I think it's as good a time as any to show that I have in fact brought this back to interactive frame rates with the help of a spatial hashing system:




It's a shame that, aesthetically, the video isn't very pleasing. It comes down to limitations of raw photon mapping really, as well as the hardware I have to work with. This scene has just short of 400,000 photons bouncing around it, but even that is not enough to uniformly light the objects within it. I would need to incorporate a final gathering phase of some sort if I wanted to correct the issue, and that would have a performance impact as well. I've got one or two ideas I'd like to explore but short of optimizing heavily and then using the new cycles to perform final gathering I don't see the quality improving much sadly.

More to come throughout the week.

No comments:

Post a Comment