Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Unify temporal and spatial taps #194

Merged
merged 1 commit into from
Oct 15, 2024
Merged

Unify temporal and spatial taps #194

merged 1 commit into from
Oct 15, 2024

Conversation

kvark
Copy link
Owner

@kvark kvark commented Oct 15, 2024

Unless ReSTIR is multi-pass (e.g. #192), there is really no difference between spatial taps and the temporal tap. This PR unifies them together, which simplifies the code a bit.

Before:
ball-unified-before
unified-before

After:
ball-unified-after
unified-after2

Looks like we are losing a bit of performance. The relevant metric is RT Core Throughput, and it's lower by a bit, probably because the temporal resampling made it a more coherent pattern. But in exchange, the noise level seems reduced on the "move" example.

@kvark kvark marked this pull request as draft October 15, 2024 06:55
@kvark kvark force-pushed the experiment/unified-temporal branch from 715368d to af8884d Compare October 15, 2024 07:00
@kvark kvark added the type: experiment Experimental code label Oct 15, 2024
@kvark kvark marked this pull request as ready for review October 15, 2024 07:08
@kvark kvark merged commit aaa15a5 into main Oct 15, 2024
5 checks passed
@kvark kvark deleted the experiment/unified-temporal branch October 15, 2024 07:08
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
type: experiment Experimental code
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant