Nouveau vs. NVIDIA Linux vs. NVIDIA Windows 8.1

[phoronix]

Phoronix: Nouveau vs. NVIDIA Linux vs. NVIDIA Windows 8.1

A few days back I delivered benchmark results of a 13-way graphics card comparison comparing the OpenGL gaming performance between Ubuntu Linux and Microsoft Windows 8.1. In that article there were multiple AMD Radeon and NVIDIA GeForce graphics cards tested using the binary drivers for each operating system. On Wednesday I then shared results with the Radeon Gallium3D driver compared to Catalyst on Linux and Windows while today are some of the early Nouveau results compared to the proprietary NVIDIA GPU drivers on Windows and Linux.

http://www.phoronix.com/vr.php?view=19292

[Vim_User]

So on the latest generations of Nvidia hardware the closed source Nvidia driver is in many cases faster on Linux than on Windows. A great point for Linux gaming and Nvidia, but it already shows that the often sub-par performance of AMD's closed drivers is not caused by Linux, but by the driver itself.

My two conclusions on that:
- Nvidia, give the nouveau developers more information, so that they can work out the issues with their drivers
- AMD, get the developers of your closed drivers the whip, so that they come up with an equally performant driver

[kokoko3k]

Stop gputest, please

gputest benchmark is always giving results that differs way too much from all of the others.
Please, stop wasting time and replace it with some unigine benchs!

[MatthewPL]

So on the latest generations of Nvidia hardware the closed source Nvidia driver is in many cases faster on Linux than on Windows. A great point for Linux gaming and Nvidia, but it already shows that the often sub-par performance of AMD's closed drivers is not caused by Linux, but by the driver itself.

Sorry, but anyone can tell that Linux is better for gaming depends only on FLOSS games? No offence, but I don't think FLOSS games are the best benchmark. Especially if we now have Steam on Linux.

[brosis]

gputest benchmark is always giving results that differs way too much from all of the others.
Please, stop wasting time and replace it with some unigine benchs!

Thanks for the tests!

I support this, although in less aggressive position. GPUtest triangle is basically glxgears, which is worthless (except when comparing same environiments, but different clocks f.e.).
The rest of Gputests is somehow very strange and unrealistic. Unigine or any modern 3D (heck, even furmark) closed or open, would do better. Primary reason - its modern, native and crossplatform; rest plays no role.
PS
Yes,Michael, I know why you don't bench Steam.

- AMD, get the developers of your closed drivers the whip, so that they come up with an equally performant driver

You can't be fscking serious!?

[Vim_User]

Sorry, but anyone can tell that Linux is better for gaming depends only on FLOSS games? No offence, but I don't think FLOSS games are the best benchmark. Especially if we now have Steam on Linux.

Still the same games are used in both benchmarks, so the relations between Windows and Linux should be the about the same if the game is the cause. If the game itself performs better on Linux it should do it in both cases, for AMD and Nvida hardware.

[dimko]

I also thought of it

gputest benchmark is always giving results that differs way too much from all of the others.
Please, stop wasting time and replace it with some unigine benchs!

Probably thing is badly optimized for Linux. Probably not GPU bottlenecked.

At the end of the day i have seen openCL tests of bitcoin mining software, difference was so small...

[schmidtbag]

- Nvidia, give the nouveau developers more information, so that they can work out the issues with their drivers
- AMD, get the developers of your closed drivers the whip, so that they come up with an equally performant driver

Nvidia could care less about the FOSS drivers. I'm sure the only reason they released any information was so people would stop complaining to them, and to help improve their tegra platform (since they don't seem to care about making closed source drivers for that).

As for AMD, they don't have the money or marketshare to really care about giving catalyst any more attention than it already has, and the catalyst drivers already have too much room for improvement on Windows. They basically just ensure that it works and they don't seem to touch it otherwise. As soon as the radeon and radeonSI drivers catch up, it would make sense for AMD to move the linux catalyst devs over to the FOSS driver development and shut down the catalyst drivers for good. Giving the FOSS drivers undivided attention could easily make AMD the best choice for linux as a whole (when you take into consideration price, performance, politics, and features).

[verde]

Please replace all those uninteresting games and benches (ex. Gpu test) with Unigine benchmarks, TF2 and Dota which are free games and another 2-3 games you can buy with less than 15-20€ all together (L4D2, HL2, S. Sam 3) and then the benchmark will be really interesting.

This is how hardware/software reviews work.

On topic now, I am glad that nvidia drivers are on par with Windows on my GPU (GTX460). Nouveau is a disaster and I don't think will ever reach the AMD Gallium quality, features and performance.

With a little more effort, AMD gallium drivers could have been official drivers against fglrx.

[sarmad]

What I still don't understand is why hardware manufacturers like AMD and nVidia keep their drivers closed source. If they are making profit from the hardware only, why close the software? What is so secret about it that they have to keep it closed source? In fact, I think open sourcing it should help them reduce the cost since the community would be doing some of the work on their behalf. Am I missing something?