Guys I’m losing my ■■■■ mind here trying to get shaders to compile at a reasonable pace!
I’ve seen a bunch of “Fixes” and sure the XGE one unstucks shaders so they actually compile but beyond that - no changes to ANY of the following settings seems to have any effect!
Specs for reference: Ryzen 1950x (3.8 Ghz) 64GB RAM (3200mhz) Samsung EVO 970 1TB
----------------------------------------------------------BaseEngine.ini:----------------------------------------------------------
[DevOptions.Shaders]
bAllowCompilingThroughWorkers= True / False
MaxShaderJobBatchSize= 10 → 24
WorkerProcessPriority= 0 → 2
bUseVirtualCores = False / True
CookerMemoryUsedInGB = 48 (or any)
MemoryToLeaveForTheOSInGB = 16 (or any)
MemoryUsedPerSCWProcessInGB = 0,4 - 2
MinSCWsToSpawnBeforeWarning = 8
bForceUseSCWMemoryPressureLimits = False / True
------------------------------------------------------ConsoleVariables.ini:------------------------------------------------------
(Commented / Uncommented no change to the actual value)
r.ShaderCompiler.JobCache= 0
r.XGEShaderCompile = 0 (only one that seems to have any effect!)
r.Shaders.AllowCompilingThroughWorkers=0
r.Shaders.SkipCompression=1
I know it used to be faster but have no ■■■■ clue what changed -
Also - no matter settings, it refuses to use all my cores / threads and even when it uses more of them - its still the same (shader comp) speed - as far as I can tell !?
Update:
Just realized it doesnt even launch any ShaderCompilerWorkers now - again NO matter the settings !?!?
Any and all advice is welcome
Thanks in advance