OpenGL - MESA or Proprietary driver?


Oct 13, 2016
OK, i went on doing a self test benchmark of my GPU HD7750 2GB DDR3 with
and get the score, mostly for comparison on 3D rendering.

glmark2 2014.03
OpenGL Information
GL_RENDERER: Gallium 0.4 on AMD CAPE VERDE (DRM 2.45.0 / 4.7.6-200.fc24.x86_64, LLVM 3.8.0)
GL_VERSION: 3.0 Mesa 12.0.3
[build] use-vbo=false: FPS: 1884 FrameTime: 0.531 ms
[build] use-vbo=true: FPS: 2793 FrameTime: 0.358 ms
[texture] texture-filter=nearest: FPS: 2523 FrameTime: 0.396 ms
[texture] texture-filter=linear: FPS: 2514 FrameTime: 0.398 ms
[texture] texture-filter=mipmap: FPS: 2634 FrameTime: 0.380 ms
[shading] shading=gouraud: FPS: 2491 FrameTime: 0.401 ms
[shading] shading=blinn-phong-inf: FPS: 2502 FrameTime: 0.400 ms
[shading] shading=phong: FPS: 2499 FrameTime: 0.400 ms
[shading] shading=cel: FPS: 2498 FrameTime: 0.400 ms
[bump] bump-render=high-poly: FPS: 1977 FrameTime: 0.506 ms
[bump] bump-render=normals: FPS: 2768 FrameTime: 0.361 ms
[bump] bump-render=height: FPS: 2744 FrameTime: 0.364 ms
[effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 2444 FrameTime: 0.409 ms
[effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 1960 FrameTime: 0.510 ms
[pulsar] light=false:quads=5:texture=false: FPS: 2449 FrameTime: 0.408 ms
[desktop] blur-radius=5:effect=blur:passes=1:separable=true:windows=4: FPS: 1090 FrameTime: 0.917 ms
[desktop] effect=shadow:windows=4: FPS: 1381 FrameTime: 0.724 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 404 FrameTime: 2.475 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 479 FrameTime: 2.088 ms
[buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 500 FrameTime: 2.000 ms
[ideas] speed=duration: FPS: 864 FrameTime: 1.157 ms
[jellyfish] <default>: FPS: 1861 FrameTime: 0.537 ms
[terrain] <default>: FPS: 253 FrameTime: 3.953 ms
[shadow] <default>: FPS: 1719 FrameTime: 0.582 ms
[refract] <default>: FPS: 347 FrameTime: 2.882 ms
[conditionals] fragment-steps=0:vertex-steps=0: FPS: 2814 FrameTime: 0.355 ms
[conditionals] fragment-steps=5:vertex-steps=0: FPS: 2826 FrameTime: 0.354 ms
[conditionals] fragment-steps=0:vertex-steps=5: FPS: 2846 FrameTime: 0.351 ms
[function] fragment-complexity=low:fragment-steps=5: FPS: 2848 FrameTime: 0.351 ms
[function] fragment-complexity=medium:fragment-steps=5: FPS: 2838 FrameTime: 0.352 ms
[loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 2845 FrameTime: 0.351 ms
[loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 2829 FrameTime: 0.353 ms
[loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 2807 FrameTime: 0.356 ms
glmark2 Score: 2067

From what i know for MESA drivers the 3D is being handled from software rather than hardware. That means 3D is mainly being handled from CPU. For GPU handling is better to use the proprietary drivers.

But the current MESA drivers i have installed support OpenGL 4.3 and from Monday i think they will get an update supporting Vulkan drivers (my GPU is compatible).

The proprietary drivers are the version Crimson Edition 15.12 from 12/18/2015 and are not being updated any more.

What do you think is better for 3D? Don't tell me to download the proprietary and run the benchamark once more because installation is the a pain in the ...

ps1. For Nvidia GPU it seems that they get higher scores with proprietary than MESA ones.

ps2. Also this can be an easy guide for self testing different GPU models "power" from the user it self. You can always install a Linux OS (free software) as dual boot on your Windows PC or run it as virtual machine. You test your current GPU model, you plug the new one, test and compare result (be sure to compare with the same set of drivers both GPU models).

A user with Nvidia proprietary drivers and GPU model GeForce GTX TITAN Black/PCIe/SSE2 got a score of 15943 (compared to mine 2067...) but that was back in 2014, and i have no idea if MESA drivers have surpass proprietary in 3D.