Skip to main content


The current Steam Survey numbers seem...off.

Reads like Pop!_OS and Linux Mint weren't being counted before.

It also shows a total increase for Linux over last month, despite the overall percentage being lower. Whut.
This entry was edited (1 year ago)
in reply to Liam @ GamingOnLinux 🐧🎮

Personally, I expect people to move more and more from #Windows10 to #Windows11 due to the bs requirement for a #TPM for #CensorBoot...

https://www.youtube.com/watch?v=_Ua-d9OeUOg
in reply to Liam @ GamingOnLinux 🐧🎮

PopOS and Mint where probably included in the other field, instead of being identified separately.
in reply to mrAX

@mrax Together their % is more than what Other dropped by, so that doesn't even relate
@mrAX
in reply to Liam @ GamingOnLinux 🐧🎮

@mrax (plus if they only moved out of other, it would / should only still show their monthly change)
@mrAX
in reply to Liam @ GamingOnLinux 🐧🎮

@mrax couldn’t this be explained by both moving them out of other and a general increase in adoption of each?
@mrAX
in reply to Liam @ GamingOnLinux 🐧🎮

This is based on random samples, so you must expect an inherent error and instability to the numbers. Using a single data point (one month) is not a good idea.
in reply to Mivey

@mivey uh no, it's generally fine, it's not instability, this month just doesn't make any sense
in reply to Liam @ GamingOnLinux 🐧🎮

Oh, right, the sums over the percentages seem off. Perhaps Valve updated the underlying math somehow? It be strange for an algorithm to make "mistakes" when adding up numbers
in reply to Liam @ GamingOnLinux 🐧🎮

Well, all in all the #linux percentage is higher so I guess we can count it as a win? 🤷‍♂️
in reply to Liam @ GamingOnLinux 🐧🎮

@fauzruk how is it not a win? I’m confused - is there something I’m missing? Is Linux Marketshare not at an all time high once again?
in reply to Liam @ GamingOnLinux 🐧🎮

Steam could adjust the data collection for Flatpaks. The host system info is available under `/run/host/os-release`.
in reply to Liam @ GamingOnLinux 🐧🎮

I feel like I've seen this kind of error a lot with OS stats that I'm not sure how we trust any of it tbh. I'm just hoping eventually it all gets ironed out. Wish they didn't put flatpak as a distro tho... That just seems so.... Odd. Would think flatpak would pass in the actual distro but that's just me....
in reply to Liam @ GamingOnLinux 🐧🎮

If I had to guess steam has some internal threshold to count as others. Lets say 3 %. And both Popos and Linux Mint were below that last month. Lets say they were both 2.86%. But they grew their marketshare since last month (or error is around 1%). This would make this table and It will decrease others by (2.86 *2) = 5.71.
in reply to Liam @ GamingOnLinux 🐧🎮

I wondered if they counted Flatpak separately. I was using that on Arch, then when I switched to Fedora recently, I kept using the Flatpak. Hard to beat the convenience when using proprietary software.
in reply to Liam @ GamingOnLinux 🐧🎮

This encouraged me to spend the time to upgrade from Mint 20.3 to Mint 21 so i could be in the list next month. Cheers.

Lo, thar be cookies on this site to keep track of your login. By clicking 'okay', you are CONSENTING to this.