CAS/RCAS sharpening: verify placement vs TAA/FSR (currently post-AA but pre-tonemap in HDR); option to sharpen at the end of the chain #434
Labels
No labels
amd
avatars
bug
builder
cannot-reproduce
crash
epic
feature
from-viewer
gl-removal
known-limitation
monitoring
parity
perf
platform
priority
critical
priority
high
priority
low
priority
medium
provenance
rendering
rt
status
in-progress
status
testing
status
up-next
triage
ui
upstream
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
sky_line/Slipstream#434
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Verify where CAS / RCAS sharpening (
RenderCASSharpness, Preferences > Graphics / Lightbox) runs relative to TAA and FSR and the rest of the post chain, and consider letting users sharpen at the end (post-TAA/FSR and post-tonemap / post-bloom).What the code does today (to verify in RenderDoc / with captures)
clean_rcas.comp) runs on the TAA-resolved image as it's written intomHdr(instead of the plain copy). The TAA history stays unsharpened, which is correct: no feedback amplification (llcleanrenderervk.cpp~14320-14364,#183).mFsr->record(cmd, f, &sharpened), ~14519);mRcasRanmakes the final pass skip its CAS.RenderVKSharpenOnce(B1/F4a) ensures sharpening happens only once (it used to double-sharpen, and RCAS used to run without TAA).So with TAA / FSR, sharpening is post-TAA/FSR but pre-tonemap, in HDR, before exposure / tonemapping, bloom, colour grading, the final pass, and any FXAA/SMAA.
Questions
RenderCASSharpness? It should map the same slider to FSR's sharpness parameter consistently; check the mapping and the 0 = off behaviour.Possible change
Done when