Intel OpenGL Performance Across 11 Linux Distributions

[phoronix]

Phoronix: Intel OpenGL Performance Across 11 Linux Distributions

When carrying out the 11-way Linux distribution boot time tests I also used that as an opportunity for some fresh OpenGL graphics tests with the Intel system on the variety of Linux distributions benchmarked.

http://www.phoronix.com/vr.php?view=25582

[tildearrow]

Typo:

The 11 Linux distributions tested were Antergoa 17.11

[jhenke]

Would be nice to know how it compares to a Gentoo system.

[dh04000]

Switching from Unity7 to Gnome3 has had a marked effect on the graphics performance.

[andrei_me]

Looking at the graphs, it seems that the distros with updated graphics stack have same/similar performance, and the perf difference is caused by other optimized packages in the system

[Veto]

Switching from Unity7 to Gnome3 has had a marked effect on the graphics performance.

What do you mean? Good or bad? Desktop or fullscreen gaming?
Just curious not having upgraded from 17.04 yet

[andyprough]

You'd have to squint real hard to see much difference in most of those tests. Clear Linux did not perform terribly well for a "distro" whose only apparent purpose in life is to use little tricks to beat real, serviceable Linux distros at performance tests.

[Strunkenbold]

Is performance governor the default for clear linux? Would be nice to know if there the performance diff comes from...

[deant]

Actualy Debian in some cases outperfoms Ubuntu even with older software. But I think that tests should be done on debian testing (buster in tha case), coz I dont think many ppl use debian stable. i for one prefer testing on production and sid for home/fun.

[gamerk2]

You'd have to squint real hard to see much difference in most of those tests. Clear Linux did not perform terribly well for a "distro" whose only apparent purpose in life is to use little tricks to beat real, serviceable Linux distros at performance tests.

Because there's only so much you can do within the graphics stack for GPU intensive tasks; your gain is fundamentally limited by how much GPU performance you're leaving on the table. For these types of tests, you're always talking just a few FPS between being optimized and not.