18-Way GPU Linux Benchmarks, Including The Radeon RX 460 & RX 470 On Open-Source

[dungeon]

Actually i am not sure that -s always workaround an issue I would put something like this to ensure it is there

xrandr --output XYZ --mode 1920x1080 --fb 1920x1080 --panning 1920x1080

[dungeon]

But the best is to write an article how recently kwin developer removed unredirect fullscreen windows And advise people to run games without composition, etc...

On Windows game set lower resoultion without any issue, but on Linux many DE does not like that... so cry now or workaround it properly

[arakan94]

I still don't understand one thing.. Why are AMD cards tested with open source drivers and Nvidia with proprietary ones? This comparison is pointless - either I want to be using open drivers and I want to see how Nvidia cards perform under them.. OR I don't care, so I will be using proprietary ones and want to see how AMD performs with them!

[dungeon]

Why are AMD cards tested with open source drivers and Nvidia with proprietary ones?

Because that is how author wanted For equality i guess Michael will test AMD proprietary driver vs nVidia opensource driver later on

[arakan94]

Because that is how author wanted For equality i guess Michael will test AMD proprietary driver vs nVidia opensource driver later on

I guess it's his right.. But still.. It's a shame. I don't want to use some closed binary blobs so it's quite crucial for me to see how Nvidia cards perform without them. One friend of mine can't even get it working (755 or so in laptop) and news are that Nvidia don't want use what others do in Wayland..

So it would seem that my only option is to use AMD card (or settle myself with weak sauce Intel).

[starshipeleven]

I still don't understand one thing.. Why are AMD cards tested with open source drivers and Nvidia with proprietary ones?

Because the article is called "18-Way GPU Linux Benchmarks, Including The Radeon RX 460 & RX 470 On Open-Source"

More seriously, because AMD's open driver is also an official driver and people here (and on linux at large) care about how well the open driver goes against NVIDIA's only driver (the proprietary blob).

This comparison is pointless - either I want to be using open drivers and I want to see how Nvidia cards perform under them.. OR I don't care, so I will be using proprietary ones and want to see how AMD performs with them!

1. there is no need to bench Noveau (opensource third-party driver for NVIDIA based off epic reverse-engineering efforts) to know that it runs like crap on 3D loads for a long list of reasons. It's already great that it runs good on 2D and has enough 3D to support accelerated desktop environments.
2. there are already benches with closed AMD closed driver vs NVIDIA driver in Phoronix.

[starshipeleven]

So it would seem that my only option is to use AMD card (or settle myself with weak sauce Intel).

This. AMD is the only one with powerful GPUs and a decent opensource driver.

[Veerappan]

Regarding F1 performance, Nicolai Hähnle just posted v2 of a patch series that could help that game for radeonsi once it lands.

And Marek Olsak just landed some hang fixes for Overlord and Witcher 2 earlier today. Combined with the rest of the performance fixes that have been written recently, I'm very positive about the future of the radeonsi driver in Mesa, and I'll be really interested in the Mesa vs AMDGPU PRO performance picture 3-6 months from now.

[Michael]

I still don't understand one thing.. Why are AMD cards tested with open source drivers and Nvidia with proprietary ones? This comparison is pointless - either I want to be using open drivers and I want to see how Nvidia cards perform under them.. OR I don't care, so I will be using proprietary ones and want to see how AMD performs with them!

Because ask most AMD Linux users what driver they use... open-source.

Ask most NVIDIA Linux users what they use... proprietary driver.

But as mentioned a few times anyway in recent articles, the AMDGPU-PRO vs. NVIDIA proprietary tests are coming shortly.

[LinuxID10T]

Sure I can add Heaven back to the queue. Basically was just doing Valley since it was newer but I have no problems running both.

Metro is always run at whatever resolution the reported system resolution was, in this case 3840 x 2160. Metro Last Light / 2033 Redux don't offer any CLI controls for resolution handling so basically stuck running it at whatever is the native resolution.

You could add an xrandr command into the PTS script to change the resolution. All you should have to do is change the screen resolution before the game starts and it will use that resolution.