You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm trying to optimize some DX11 shader bits on a machine that does not have AMD GPU nor a driver installed. Getting really strange results, as if a bunch of shader code is not really used at all. I can reproduce that even with tiny shaders, using currently latest (2.7) RGA:
and a command line like: rga-windows-x64-2.7\rga.exe -s dx11 -p vs_5_0 -f vs -c gfx900 --isa out.txt test.hlsl produces these relevant bits in the output:
We updated the RGA 2.7 Windows release archive with a fix and updated the release notes accordingly.
Please note that gfx1100 is not yet supported as a target in DX11 mode - this will be addressed in a future release.
I'm trying to optimize some DX11 shader bits on a machine that does not have AMD GPU nor a driver installed. Getting really strange results, as if a bunch of shader code is not really used at all. I can reproduce that even with tiny shaders, using currently latest (2.7) RGA:
and a command line like:
rga-windows-x64-2.7\rga.exe -s dx11 -p vs_5_0 -f vs -c gfx900 --isa out.txt test.hlsl
produces these relevant bits in the output:It's as if a shader is actually outputting
(0,0,0,1)
. The same happens with other-c
arguments, e.g.gfx1100
orgfx1030
.The text was updated successfully, but these errors were encountered: