Steam Survey Purports A 0.28% Linux Gaming Marketshare For February

[nuetzel]

Everyone knows all the Windows XP figures are people running old Wine prefixes. As of a few months ago, Wine now creates a Windows 7 prefix by default, so many of the Windows 7 figures will be Wine users too.

And since the recommended approach is to have one game per prefix, most people would have many wineprefixes with Steam installed, whereas they likely only have a single GNU/Linux native Steam installation. For example, I currently have 131 wineprefixes on my GNU/Linux installation that each have Steam installed within, but only one copy of Steam for GNU/Linux. Obviously I'm going to end up getting the Steam survey more than 100x as much for my wineprefixes than for the native Steam installation which I would rather it count (but almost never actually does).

And that's the only reason why it *looks* like GNU/Linux Steam usage is under 1%. Of course the GNU/Linux sales are still significantly less than 1%, but that's only because Wine is just that good.

Than we badly need something like browser identification string for Wine.
System
OS (!)
etc.

What do you think?

[lowflyer]

Not that I believe in the value of the Steam survey, but none of the other surveys you included are Gaming-oriented, so i don't think they're related to the topic.

So you think that *all* users have their own separate machine to run their games on? Gimme a break! Stop making silly arguments for arguments sake!

[Jabberwocky]

So you think that *all* users have their own separate machine to run their games on? Gimme a break! Stop making silly arguments for arguments sake!

The problem is it's not very easy to pull stats on it (unless you are looking at Windows 10 results only and you are working at Microsoft) which is the base of the whole Steam survey issue. Without proper stats your argument is the same as geearf's.

Having two PCs one for Linux and one for Windows is becoming very popular in my circles, also a lot of people doing pci-e passthrough setups now that Ryzen is affordable option and it's relatively easy, no more kernel patching and most of the process is documented. I was surprised to see how many people adopted setups like that since December 2017 community support is also really good if you get stuck https://www.reddit.com/r/VFIO/ . If you are not a hardcore gamer APUs is also a good option for a second gaming PC, I bought 2 last week (2400G & 2200G).

Edit: Just to be clear. I'm also not saying everyone has second PC for Windows+Steam.

[lowflyer]

The problem is it's not very easy to pull stats on it (unless you are looking at Windows 10 results only and you are working at Microsoft) which is the base of the whole Steam survey issue. Without proper stats your argument is the same as geearf's.

My argument is only: disregard the steam survey as it is a clear outlier.

You do not need statistics "to prove" an outlier and only then disregard it. A quality check should be an integral part of *any* statistics process. With the steam surveys figures it's pretty blatantly obvious that something's wrong. I'm not the only one pointing that out, other folks have made similar claims in the past based on their direct experience with the survey, also on phoronix. If you read my original post, you'll see that I did provide some stats.

Besides creating heated discussions on phoronix, this should IMHO at least trigger somebody at Valve to investigate the reasons.

[geearf]

So you think that *all* users have their own separate machine to run their games on? Gimme a break! Stop making silly arguments for arguments sake!

No I think the proportion of Linux users that play games is different from the one of Windows users, so having a certain share of Linux users in general or in specific areas does not say that much in regards to how big a share you have in another specific area.
Of course, feel free to again not understand the point I am making and yet complain about it.

[lowflyer]

No I think the proportion of Linux users that play games is different from the one of Windows users, so having a certain share of Linux users in general or in specific areas does not say that much in regards to how big a share you have in another specific area.
Of course, feel free to again not understand the point I am making and yet complain about it.

Dear geearf, I believe we are mostly on the same plane here so we shouldn't need to argue. I am just trying to tell you that it is very silly to draw an argument because "none of the other surveys you included are Gaming-oriented". You topple that with the argument above: "the proportion of Linux users that play games is different from the one of Windows users". Note that you start your sentence with "I think...". Are you really trying to tell the world that most of the Windows users are using their machine only for gaming? If you've ever worked in a company with more than 20 employees you should know better. And the statistics I've provided confirm that point.

Sure, if you really want to, you can use the figures of the steam survey to claim "all that windows is good for is gaming" But this will be a rough (and a rather short) ride.

By the way, my first post is the one with the most likes in this thread (at least for now), so I'm not the only one with this opinion.

[geearf]

Dear geearf, I believe we are mostly on the same plane here so we shouldn't need to argue.

There's never a need to argue, but debating can be productive I think.

Note that you start your sentence with "I think...".

Well yes, I like to separate facts from opinion (when I remember at least).

Are you really trying to tell the world that most of the Windows users are using their machine only for gaming? If you've ever worked in a company with more than 20 employees you should know better. And the statistics I've provided confirm that point

Sure, if you really want to, you can use the figures of the steam survey to claim "all that windows is good for is gaming" But this will be a rough (and a rather short) ride.

No, again you're imagining things I am neither saying nor implying.

By the way, my first post is the one with the most likes in this thread (at least for now)

Congratulations!

Again I am not saying that the steam survey shows a correct picture or that your figures are necessarily wrong, so yes we may agree on the outcome, but not so much on how we got there.

[boltronics]

Than we badly need something like browser identification string for Wine.
System
OS (!)
etc.

What do you think?

Wine actually already includes this information in the OS information which is collected by Steam. Unfortunately Steam simply ignores it AFAICT.