NVIDIA GeForce: Windows 10 vs. Ubuntu 15.04 Linux OpenGL Benchmarks

[phoronix]

Phoronix: NVIDIA GeForce: Windows 10 vs. Ubuntu 15.04 Linux OpenGL Benchmarks

Earlier this week I ran some Windows 10 vs. Linux benchmarks of Middle-earth: Shadow of Mordor, the latest AAA game that's been ported to Linux. Those results showed the Linux version of this game running much slower than Windows, so while having a Win10 installation around I decided to also run some fresh OpenGL Windows 10 vs. Ubuntu Linux benchmarks on some older titles. Here are those results.

http://www.phoronix.com/vr.php?view=21977

[duby229]

That is exactly why synthetic tests are not valuable. They don't say anything about how your games will play. Only gameplay can show how well a game will play.

[Daktyl198]

That is exactly why synthetic tests are not valuable. They don't say anything about how your games will play. Only gameplay can show how well a game will play.

They don't tell you anything about bugs, tearing, etc, but they do give a pretty good general overview of the kind of performance you should see when playing. Most benchmark modes go through simulations of most things you'll be doing/seeing in the game.

[GreatEmerald]

Something is hilariously broken in your Xonotic test there. Windows scores are flat, increasing the resolution improves performance, and a 780 Ti does better than a 980 Ti...

[andre30correia]

something i like to see, cgo dx windows 10 cgo opengl ubuntu

[Michael]

Something is hilariously broken in your Xonotic test there. Windows scores are flat, increasing the resolution improves performance, and a 780 Ti does better than a 980 Ti...

Waiting to hear back from NVIDIA as they uncovered Xonotic performance issues/improvements in the past.

[Michael]

something i like to see, cgo dx windows 10 cgo opengl ubuntu

Would be interesting, hopefully there's enough supporters after that to allow it to happen.

[duby229]

They don't tell you anything about bugs, tearing, etc, but they do give a pretty good general overview of the kind of performance you should see when playing. Most benchmark modes go through simulations of most things you'll be doing/seeing in the game.

The numbers in this article at least don't agree with you. Gputest results never line up with any game benchmarks. If you are looking for a general overview, synthetic benchmarks can't give you that by their very nature. I'm not a fan of time demos either, but at least those use actual games to render them.

The first rule of benchmarking is to benchmark the game or application you want info on. Synthetic benchmarks will never meet the first rule.

EDIT: Honda's Formula One doesn't say anything at all about Honda's Civic. Same difference here. Synthetic benchmarks are usually made to stress specific bottlenecks and unless your game or app suffers from those specific bottlenecks, they don't matter.

EDIT: Also it is an absolute fact that both nVidia and AMD have been known to cheat on synthetic benchmarks in the past. That effectively makes them useless as a measure of performance.

[gamerk2]

The fact the synthetics give equivalent results would tend to rule out driver side issues.

Again, as I said in the SoM thread, I firmly believe you're seeing a difference in the way threads, in particular, the thread that handles most of the GPU workload, is getting scheduled. If that thread ever gets bumped, you are going to lose significant performance. Windows biases high workload threads to prevent this from happening, Linux does not. Even if the thread gets bumped for 1ms, that's still 6.25% your render window.

Point being, a purely synthetic benchmark may run on just a few threads. A full fledged game, at least on windows, easily pushes 80, even if only a handful do meaningful work. But how those high workload threads get scheduled is of the upmost importance.

[eydee]

Something is hilariously broken in your Xonotic test there. Windows scores are flat, increasing the resolution improves performance, and a 780 Ti does better than a 980 Ti...

People say the current state of nvidia win10 drivers is a disaster. Think of it like running fglrx on linux...