- Splitting symbol generation and shader "debug info" generation into two different cvars. This should now allow us to generate symbols post-packaging without generating different shader variants. This should also enable us to always generate symbols and still have the final deduplicated shader sizes.
- r.Shaders.Symbols is now the primary way to enable shader symbols for debugging.
- r.Shaders.GenerateSymbols can be enabled to always generate symbols without writing them to disk.
- r.Shaders.WriteSymbols can be used to force write symbols if they were previously generated.
- r.Shaders.ExtraData can be used to generate shader names for engine runtime systems to use.
- NotifyShaderCooked replaced with NotifyShaderCompiled, prep for upcoming non-cook symbol writing.
- PC DXC shader compiles should only generate debug info when requested, not all the time.
- PC FXC shader compiles should generate debug info when requested, not just when r.Shaders.Optimize is 0.
#jira none
#rb arciel.rekman, lukas.hermanns
#preflight 611ad9035e737200015992db
[CL 17186155 by christopher waters in ue5-main branch]
Change default behavior to allow a null login with empty credentials, as this matches 99% of the platforms
#rb rob.cannaday
[CL 17124087 by ben zeigler in ue5-main branch]
* 9 short jittered rays are traced from the voxels of a camera-aligned volume texture, and on miss the rays interpolate from a world space Radiance Cache just for Translucency GI. The Radiance Cache is populated only where the Lumen Translucency Lighting Volume will sample, and generates mips to avoid introducing noise when sampled.
* A spatial filter runs on the volume texture after tracing to smooth out noise
* Quality of Lumen GI on Translucency and Volumetric Fog is significantly higher than before, while cost at defaults went from .28ms to .51ms on 2080Ti
* HZB occlusion is used to avoid tracing for any voxels whose trilinear footprint is fully occluded by opaque
* Important cvars: r.Lumen.TranslucencyVolume.TraceFromVolume, r.Lumen.TranslucencyVolume.TracingOctahedronResolution, r.Lumen.TranslucencyVolume.RadianceCache, r.Lumen.TranslucencyVolume.RadianceCache.ProbeResolution
* Hardware Ray Tracing support for Lumen Translucency Volume
* Implemented a single dispatch mip generation for Radiance Cache
#preflight 6111f73f0ccb840001a1f260
#ROBOMERGE-SOURCE: CL 17118715
#ROBOMERGE-BOT: (v855-17104924)
[CL 17118791 by daniel wright in ue5-main branch]
Enable with r.VT.Residency.Notify
Default VT residency mip map on for default pools.
This is overridable per project and per pool.
#rb none
#preflight 61128bb272fe2c00013a846c
#ROBOMERGE-SOURCE: CL 17116695
#ROBOMERGE-BOT: (v855-17104924)
[CL 17116736 by jeremy moore in ue5-main branch]
was High
CompressEffortLevel_NoEditor=Normal
so that Editor and NoEditor make the same texture by default
makes the cooked texture quality slightly lower, but faster
#rb none
[CL 17107563 by charles bloom in ue5-main branch]
Due to the prevalence of SSDs, the Boot cache no longer offers the performance benefits that it once did, and its startup cost now often cancels out its performance gains.
Additionally, there are cases where different values are being written for the same key, which is incorrect but has not been fixed; and the Boot cache ignores the second and subsequent writes, which can cause out-of-date content to be loaded from the Boot cache in a future session.
#rb Zousar.Shaker
#rnx
[CL 17085965 by Devin Doucette in ue5-main branch]
- Add CVar (r.Vulkan.Allow64bitShaderAtomics) to forcefully disable 64bit image atomics. Currently defaults to 0 (until other components are in).
- Check driver version for Vulkan on NVIDIA, a fix was added for the 64bit image atomics that is required for Nanite. (currently Windows only, Linux TBD)
#rb Rolando.Caloca
#fyi mihnea.balta,dmitriy.dyomin,brandon.schaefer
[CL 16982971 by jeannoe morissette in ue5-main branch]
- Add a few DataDrivenPlatformInfo ShaderPlatform flags for METAL_SM5.
- Attribute dual-source blending targets.
- Match the RWBuffer type for RWScreenProbeRadianceSHDirectional between the HLSL and C++ code.
- Update PF_FloatRGB buffer format to RG11B10Float.
#rb Daniel.Wright, Richard.Wallis, Lukas.Hermanns
#jira UE-117253
#rnx
[CL 16972126 by will damon in ue5-main branch]
