Intel's Clear Linux Shows Much Potential For Optimized Graphics Performance

[phoronix]

Phoronix: Intel's Clear Linux Shows Much Potential For Optimized Graphics Performance

When it comes to CPU workloads, stunning in our Linux distribution comparisons has been Intel's Clear Linux distribution. This Intel Open-Source Technology Center project has led many of our distribution / OS comparisons with Intel engineers investing heavily in performance optimizations via AutoFDO, LTO-optimized binaries, aggressive compiler flags by default, and more. But how does the OpenGL performance compare for Clear Linux? Here are some graphics benchmarks and in select cases the results are quite a surprise.

http://www.phoronix.com/vr.php?view=23058

[Passso]

If "real" games get that boost from it, Clear Linux may be the next distro base for players... or SteamOS

[Michael]

If "real" games get that boost from it, Clear Linux may be the next distro base for players... or SteamOS

Will be trying their latest build later today that hopefully has SDL all integrated and will be trying it from a Core i5 6500.

[bug77]

If "real" games get that boost from it, Clear Linux may be the next distro base for players... or SteamOS

As I read the graphs, games don't win, things that are CPU-bound do.

[FireBurn]

Does that prove that Gentoo would in theory be faster if you configure it correctly?

[karolherbst]

Does that prove that Gentoo would in theory be faster if you configure it correctly?

or if you would use ICC

[microcode]

Cool, it would be nice to start porting over these build insights to Arch Linux PKGBUILDs; the difference in performance is compelling. On Clear Linux, it seems that the Intel driver trounces the Windows drivers in performance.

[schmidtbag]

Cool, it would be nice to start porting over these build insights to Arch Linux PKGBUILDs; the difference in performance is compelling. On Clear Linux, it seems that the Intel driver trounces the Windows drivers in performance.

I was thinking the same thing. It'd be interesting to see a comparison.

[carewolf]

Almost looks like xonotic is using CPU rendering. That often gives a trivial 2-4x speedup when using proper -O3 -march=native compared to -O2. We had huge of rendering speeds-up when we enabled -O3 on QtGui two years back. Especially for everything that didn't already have hand-optimized routines.

[rabcor]

Can we have Dirty Linux? From AMD?