Here Are The Steam Survey Results For Linux During August 2016

[Amarildo]

I think I ran TF2 with the 2013 firmware with llvmpipe.

The same performance (between 10 to 40 FPS) was reproduceable when running with the latest firmware but with "radeon.dpm=0". I didn't test much in this condition, but I couldn't see any crashes for the small period of time tested. I'll do more testing now.

glxinfo: http://pastebin.com/raw/9XCNgZtL

[Amarildo]

Just finished playing for +40 min without a single issue, with `radeon.dpm=0`. Could `si_dpm.c` (in the Kernel driver) be the problem?

Here's a "git diff" between si_dpm.c from kernel 3.16 and 4.7.2: http://pastebin.com/raw/UzZFfYgp

I'll definitely look into this.

[Passso]

The survey system is very buggy. A few weeks ago I got 4 of them in a row. Literally. As soon as I clicked the finish button, another one popped up, and so on.

YOU are the +0.02%

[nomadewolf]

Come on, man. Steam boxes were DOA. Even Valve doesn't care about the project.

At this point if people want to see Linux gaming they have to support other platforms like Android (maybe Nintendo?) and hope there's a trickle-down effect.

I'm not really sure...
If that would be completely truth they would've stopped SteamOS development by now...
So, only time will tell

[vsteel]

Debian, steam for at least two years. No survey ever requested.
Numbers are crap.

I never got the numbers either for a long time but then I had to reinstall steam. I reinstalled from the SteamPowered website and not the repos. After reinstall I received the survey for the first time. I had gone years before that without ever once getting the survey. Though I have never received it since. I also have 3 kids who game on Linux on their own accounts and they have never received it either.

The numbers are definitely low.

[duby229]

The thing about linux desktop marketshare is that the only metric that can be used is website hits... Valves hardware survey is not a poll. I did manage to get one once about a year or so ago. But it gets triggered on login, so if you don't logout of steam or reboot often your chances of getting the survey lessen. That's why it's not a poll. On the other hand website hits are agnostic, you visit a website and the website records the agent string. Of course that method relies on every desktop linux user visiting the same website in the same poll time frame. We know that doesn't happen.

Anyway, desktop linux marketshare must be less than 3%, but it's definitely higher than 2% globally which are mostly government or education workstations. The amount that are gamers, definitely less than 1%. And unfortunately there is no possible way to get numbers any more accurate then that.