Valve's Steam Survey Numbers For October Just Point To More Inaccuracies

[Dukenukemx]

There are still too many games on Steam that need changes done to make games 100% compatible. Once Proton is good enough that nobody notices a difference between Windows and Linux then we can see some massive gains for users.

[aht0]

I tried playing CSGO. It was laggy as hell and I immediately uninstalled it. PIng seemed to be OK (~40ms), FPS was also OK (>200), but it felt like there was a 0.3 seconds delay between my actions and the reaction of the game. It felt like I was in water. I couldn't win a single duel. No, thank you.

It's called "input lag" and yes, it's a bitch.

[deemon]

This month is another clear example of why rushing to report it isn't ideal. Phoronix said:

Valve have already revised it to say it's 0.83% (0.00% change not the +0.03% Phoronix stated), meaning it's the same as last month and they may well update it again in the next few days.

If you select from the drop down menu "Linux Only" ( https://store.steampowered.com/hwsurvey?platform=linux ) you see better the only linux numbers, so there has been some little movement.

Ubuntu 18.04.3 LTS 64 bit	20.70%	+1.10%
Ubuntu 19.04 64 bit	11.19%	+0.34%
"Arch Linux" 64 bit	10.64%	-0.01%
"Manjaro Linux" 64 bit	9.99%	+0.35%
Other	47.48%	-1.78%

Which sadly in grand picture is still +0.00% for Linux this month. I really wish they tried better to de-tangle the "Other", to what it actually might be.

[Ardje]

I actually got my first spontaneous hardware survey in my life. And it had to be on my WIN2. It did not even report the hardware correctly, like my touchscreen, which works fine, but it reported as "no touch screen".

[deemon]

I actually got my first spontaneous hardware survey in my life. And it had to be on my WIN2. It did not even report the hardware correctly, like my touchscreen, which works fine, but it reported as "no touch screen".

What happened when you restarted Steam in your linux machine before even completing it in windows? Last time I did that I got to fill it in in Linux also... which I did and then canceled windows computer submission (where I don't play anyway).

[Ardje]

What happened when you restarted Steam in your linux machine before even completing it in windows? Last time I did that I got to fill it in in Linux also... which I did and then canceled windows computer submission (where I don't play anyway).

What windows? The only reason windows would run on any of my computers is for reversed engineering a driver, and that was 18 years after I stopped looking at that crap.
This was an actual spontaneous hardware survey on linux.
I don't remember ever starting steam on windows. Purely since I had no official licenses until I got my win and win2. But installing windows on a win2 took a bite of 48+ hours of the less than 14 days I had to send back the $1000 device I bought if I couldn't get it to work on Linux. I also spend another 3 days on revving the .sys file, but then I decided to just use USBPCAP and a few days later, the DDJ SX3's debute on linux was on an odroid xu4, cause I had plenty and mixxx worked perfectly on a xu4 for years ;-).

Eh, but I digress...
My first time ever steam hardware survey. On linux. And the results were incorrect. Nice.

[deant]

I play games on linux EVERYDAY, i just do it using plain wine!

[betam4x]

I tried playing CSGO. It was laggy as hell and I immediately uninstalled it. PIng seemed to be OK (~40ms), FPS was also OK (>200), but it felt like there was a 0.3 seconds delay between my actions and the reaction of the game. It felt like I was in water. I couldn't win a single duel. No, thank you.

Really? I capped at 300 fps and 0.1% mins were well over 100 fps. 4K, 1080ti, Threadripper.

Marketshare percentages are bullshit. 100% of 1 person is...1 person. 0.01% of 20 million is 200,000.

200,000 sales of a $60 game is $12,000,000. Just glancing at the regular sale prices via price tracking sites tells me that a proper RDR2 Linux port would account for 15-20% of overall sales...for the entire game.

Joel over at the fun pumps tried the whole 'no marketshare' numbers game with me. I actually spent several hours JUST USING STEAM DATA and their own sale prices. I also pulled data sources from elsewhere and I discovered that 7 Days to Die has lost approximately 30% of all revenue due to not having working Linux support.

Every CEO/CTO/whatever uses percentages instead of useful statistics like...oh world population or the actually number of people running Linux or OS-X on the desktop (Apple is the single largest OEM out there by the way...by far. They beat the next 3 OEMs combined. Where is their game support?)