Intel Skylake Graphics: Windows 10 vs. Ubuntu 16.04 + Latest Open-Source Driver Code

[SaucyJack]

In fairness, if you look at the corresponding AMD and NVidia Windows/Linux tests the pattern here seems to be as much "Linux game ports are slower" as "drivers are slower". There's enough information across those three reviews for some pretty interesting analysis.

Feral is slower, analysis complete.

[dungeon]

Feral is slower, analysis complete.

Maybe we can propose Feral Interactive to change their logo to turtle

But Intel linux is slow with xonotic too, so it is not just that

[SaucyJack]

Maybe we can propose Feral Interactive to change their logo to turtle

But Intel linux is slow with xonotic too, so it is not just that

In the Intel case of course, but bridgman was talking about the overall picture with the recent benchmarks.

[Michael]

I haven't had time to actually *do* the analysis... it just occurred to me that it was possible with the info Michael had provided... but if I do have time over the weekend I will definitely share what I find.

This weekend I'll try to do an analysis article too... With Openbenchmarking.org I can merge all comparable results from the NVIDIA/AMD/Intel data and normalize the data to the best being 1.0 for each result to see relatively how things compare.

[dungeon]

In the Intel case of course, but bridgman was talking about the overall picture with the recent benchmarks.

It is possible that either all vendor drivers are wrong OR just Turtle Interactive

People talking their OS_X ports are also slowanders (slower then on Windows)... so drivers might fix if someting is very unusual deep slow, but that is what it is - shitty ports are shitty ports, those can't perform on par with original games.

When it comes to drivers from different vendors, well we can only talk who is less slower on xyz slow port, but generaly everybody is slower then original game for win.

For that matter there is one general rule of thumb - if some linux port vs windows game is slower on nvidia, then it will be even more slower on anything else

[boxie]

Yes, that's indeed very peculiar, wondering why that is.

Also: this means that the AMDGPU driver stack is getting better, compared to Windows, than the Intel driver stack, compared to Windows?

http://i.imgur.com/pP9BqOR.png

Granted, the AMD stack has a bit higher deviation, so I suppose they still have to polish some stuff.

This gives an interesting view: AMD on Ubuntu is only 12% slower than on Windows, while Intel is 39% slower than on Windows. Intel's measurements are also more consistent (skew closer to 0 indicates more of a normal distribution).

I think my next computers will be AMD powered...

Nice work on those calculations

[behrangsa]

Does Intel have both open source and closed source drivers for Linux?

[smitty3268]

Does Intel have both open source and closed source drivers for Linux?

I believe they ported their closed source drivers to Android at one point. I have no idea if they could be run on linux as well, but i don't think they are publicly available for people to try.

[smitty3268]

Yes, I wonder why Intel struggles here. The AMD numbers are very different.

http://www.phoronix.com/scan.php?page=article&item=amdgpu-win10-fury&num=2

Is this a regression?

No. It only affect Skylake GPUs, all the other Intel chips run fine.

[bug77]

By 'second class' I was referring to the discrepancy between the love they give to the windoze driver, and the apparent lack of love for the Linux driver. As nvidia has proven with their cross platform driver, there is absolutely no reason for Linux to perform worse than windoze.

Well, Nvidia had to sidestep Mesa to provide that level of support. A move they still catch a lot of flak for.