gistfile1.txt

Results for Sebastian Aaltonen's buffer tester https://github.com/sebbbi/perftest 
From Intel Haswell GT2 (i3 4010-U).   It was necessary to change the threadgroup count to 64x64 down from 1024, or the test would TDR.

Load R8 invariant: 2.106ms
Load R8 linear: 13.438ms
Load R8 random: 6.053ms
Load RG8 invariant: 2.105ms
Load RG8 linear: 12.763ms
Load RG8 random: 6.229ms
Load RGBA8 invariant: 2.105ms
Load RGBA8 linear: 11.197ms
Load RGBA8 random: 5.863ms
Load R16f invariant: 2.105ms
Load R16f linear: 11.756ms
Load R16f random: 6.159ms
Load RG16f invariant: 2.105ms
Load RG16f linear: 11.190ms
Load RG16f random: 5.990ms
Load RGBA16f invariant: 2.105ms
Load RGBA16f linear: 10.141ms
Load RGBA16f random: 3.114ms
Load R32f invariant: 2.105ms
Load R32f linear: 11.321ms
Load R32f random: 5.653ms
Load RG32f invariant: 2.105ms
Load RG32f linear: 10.000ms
Load RG32f random: 2.911ms
Load RGBA32f invariant: 4.202ms
Load RGBA32f linear: 8.090ms
Load RGBA32f random: 4.341ms
Load1 raw32 invariant: 0.952ms
Load1 raw32 linear: 0.922ms
Load1 raw32 random: 0.946ms
Load2 raw32 invariant: 1.005ms
Load2 raw32 linear: 1.043ms
Load2 raw32 random: 1.684ms
Load3 raw32 invariant: 1.138ms
Load3 raw32 linear: 1.789ms
Load3 raw32 random: 3.457ms
Load4 raw32 invariant: 1.258ms
Load4 raw32 linear: 1.857ms
Load4 raw32 random: 3.661ms
Load2 raw32 unaligned invariant: 1.006ms
Load2 raw32 unaligned linear: 1.092ms
Load2 raw32 unaligned random: 1.776ms
Load4 raw32 unaligned invariant: 1.212ms
Load4 raw32 unaligned linear: 2.115ms
Load4 raw32 unaligned random: 4.103ms
Tex2D load R8 invariant: 1.090ms
Tex2D load R8 linear: 1.088ms
Tex2D load R8 random: 1.280ms
Tex2D load RGBA8 invariant: 1.084ms
Tex2D load RGBA8 linear: 1.111ms
Tex2D load RGBA8 random: 1.293ms
Tex2D load R16F invariant: 1.086ms
Tex2D load R16F linear: 1.087ms
Tex2D load R16F random: 1.278ms
Tex2D load RGBA16F invariant: 1.089ms
Tex2D load RGBA16F linear: 1.208ms
Tex2D load RGBA16F random: 1.895ms
Tex2D load R32F invariant: 1.089ms
Tex2D load R32F linear: 1.095ms
Tex2D load R32F random: 1.280ms
Tex2D load RGBA32F invariant: 4.205ms
Tex2D load RGBA32F linear: 4.209ms
Tex2D load RGBA32F random: 4.212ms