Here is the latest instalment of Steam's Hardware Survey, as usual we do our monthly thing and compare it and talk about it and make sure you know not to use it as a hard figure.
Linux-only results for July 2014
Ubuntu 14.04 LTS 64 bit 32.77% -3.75%
Linux 3.10 64 bit 7.75% +0.66%
Linux Mint 17 Qiana 64 bit 7.45% +2.04%
Ubuntu 14.04 LTS 5.57% -0.46%
Ubuntu 14.04.1 LTS 64 bit 5.26% +5.26%
Ubuntu 12.04.4 LTS 64 bit 5.05% -0.39%
Other 36.15% +0.71%
Linux Overall Total: 1.11%, so we lost 0.09%
Last Month: 1.20%
My thoughts on it
As expected small movements that don't really mean much! I do wonder what distribution "Linux 3.10 64 bit" is as it's now the second most popular around. Any thoughts to share?
Things to note
Distro-hoppers will mess up the statistics, Linux users are far more likely to switch between different distributions than say a Windows user due to how many we have. So, the Linux-only stats will be quite turbulent.
Just be aware these results will probably not be that accurate as we don't know how they do their percentage results, they could be rounding up, rounding down or truncating the percentages. So a 0.5% could actually be nearly 0.6% as it could be 0.59% but they could do no rounding and simply truncate it.
Also it won't be every Steam user (it's a survey not a full statistic), so it can be as simple as not actually showing the survey to many Linux people.
Let's face it we know they don't survey 100% of their user base, only a small fraction of it, so bear that in mind as well.
Linux users typically shy away from any form of DRM, so you can bet there is a large bunch of Linux users who just simply refuse to use Steam.
Final Note: Look at it this way, Steam has around 75 million active users, 1% of 75 million is about 750,000 people. What developer wouldn't want to hook into a market that big? We are still under-represented in my eyes.
We will continue to be under-represented until we have more developers talking up Linux, and while this survey doesn't really mean that much having it rise to Mac levels would be a big breakthrough for us. Article taken from GamingOnLinux.com.
Linux-only results for July 2014
Ubuntu 14.04 LTS 64 bit 32.77% -3.75%
Linux 3.10 64 bit 7.75% +0.66%
Linux Mint 17 Qiana 64 bit 7.45% +2.04%
Ubuntu 14.04 LTS 5.57% -0.46%
Ubuntu 14.04.1 LTS 64 bit 5.26% +5.26%
Ubuntu 12.04.4 LTS 64 bit 5.05% -0.39%
Other 36.15% +0.71%
Linux Overall Total: 1.11%, so we lost 0.09%
Last Month: 1.20%
My thoughts on it
As expected small movements that don't really mean much! I do wonder what distribution "Linux 3.10 64 bit" is as it's now the second most popular around. Any thoughts to share?
Things to note
Distro-hoppers will mess up the statistics, Linux users are far more likely to switch between different distributions than say a Windows user due to how many we have. So, the Linux-only stats will be quite turbulent.
Just be aware these results will probably not be that accurate as we don't know how they do their percentage results, they could be rounding up, rounding down or truncating the percentages. So a 0.5% could actually be nearly 0.6% as it could be 0.59% but they could do no rounding and simply truncate it.
Also it won't be every Steam user (it's a survey not a full statistic), so it can be as simple as not actually showing the survey to many Linux people.
Let's face it we know they don't survey 100% of their user base, only a small fraction of it, so bear that in mind as well.
Linux users typically shy away from any form of DRM, so you can bet there is a large bunch of Linux users who just simply refuse to use Steam.
Final Note: Look at it this way, Steam has around 75 million active users, 1% of 75 million is about 750,000 people. What developer wouldn't want to hook into a market that big? We are still under-represented in my eyes.
We will continue to be under-represented until we have more developers talking up Linux, and while this survey doesn't really mean that much having it rise to Mac levels would be a big breakthrough for us. Article taken from GamingOnLinux.com.
0 Likes
Some you may have missed, popular articles from the last month:
The comments on this article are closed.
I never get the survey while I am running Linux, I always get it when I "happen" to log into windows.
0 Likes
Quoting: liamdaweWow, so Arch could possibly be Steam's number 2 distro right now. That's a lot of users.
Debian also doesn't install lsb by default. At least I didn't have it in my Debian testing. It has lsb-release and lsb-base though.
0 Likes
FutureSuture Aug 4, 2014
Quoting: ShmerlI do not follow. All I have in my package manager regarding LSB is lsb-release and Steam is able to tell me what operating system version I am running.Quoting: liamdaweWow, so Arch could possibly be Steam's number 2 distro right now. That's a lot of users.Debian also doesn't install lsb by default. At least I didn't have it in my Debian testing. It has lsb-release and lsb-base though.
0 Likes
It's that time of month again...
This is not true at all. The fact that this post exists show the average Linux user cares about this percentage a whole lot, and wouldn't miss a chance to improve it. Nobody running Windows looks at the Windows percentage dropping one 0.1 percent and says "I didn't get a survey this month, it's biased."
Have you ever met a Linux user who has refused the Steam survey? No, you haven't.
It's a survey, but there is a lot of thought that goes into them beyond what you can see. Valve would be able to identify the platform that is connecting to its server without you having to perform the survey.
They would have quotas for each of the three systems (ie, PC, Mac or Linux). The quotas would be based on the known demographics from a) previous surveys, b) actual steam connections. They would keep throwing up survey requests to users connecting from those systems until the respective quotas are filled.
The reason you see more surveys on PC is that more people refuse the survey on that system, and therefore the quotas take longer to fill.
To generate meaningful statistical data you need to correlate it with other sources. Valve is absolutely able to do this. This is done by targeted data collection to ensure it matches the known demographics. The exact same techniques are done when it comes to phone surveys, they use quotas to ensure that the demographics of their sampled data (eg "Are you satisfied with your congressman?") matches the demographics of that area as closely as possible. (In the case of telephone interviews, it is generally older people who respond, younger people typically either don't answer their phone, don't have one or refuse. So they keep dialing until they find some they need.[Citation Required])
In the case of Steam there is a strong bias present that makes Linux and Mac users more likely to respond to the survey than a PC user would. Steam needs to correct for that, and they do. It's not biased, it's proactively unbiased. The numbers are what the numbers are, +/- error.
*Citations: The Literary Digest Assessing The Representativeness Of Public Opinion SurveysAge Demographics & Evening Survey Calls
Quoting: GamewizardThere are many Linux users that are likely not to do the survey, plus in addition to the fact man distributions do not ship with lsb by default makes it even more skewed this way and that.
This is not true at all. The fact that this post exists show the average Linux user cares about this percentage a whole lot, and wouldn't miss a chance to improve it. Nobody running Windows looks at the Windows percentage dropping one 0.1 percent and says "I didn't get a survey this month, it's biased."
Have you ever met a Linux user who has refused the Steam survey? No, you haven't.
Quoting: SmashleyI never get the survey while I am running Linux, I always get it when I "happen" to log into windows.
It's a survey, but there is a lot of thought that goes into them beyond what you can see. Valve would be able to identify the platform that is connecting to its server without you having to perform the survey.
They would have quotas for each of the three systems (ie, PC, Mac or Linux). The quotas would be based on the known demographics from a) previous surveys, b) actual steam connections. They would keep throwing up survey requests to users connecting from those systems until the respective quotas are filled.
The reason you see more surveys on PC is that more people refuse the survey on that system, and therefore the quotas take longer to fill.
To generate meaningful statistical data you need to correlate it with other sources. Valve is absolutely able to do this. This is done by targeted data collection to ensure it matches the known demographics. The exact same techniques are done when it comes to phone surveys, they use quotas to ensure that the demographics of their sampled data (eg "Are you satisfied with your congressman?") matches the demographics of that area as closely as possible. (In the case of telephone interviews, it is generally older people who respond, younger people typically either don't answer their phone, don't have one or refuse. So they keep dialing until they find some they need.[Citation Required])
In the case of Steam there is a strong bias present that makes Linux and Mac users more likely to respond to the survey than a PC user would. Steam needs to correct for that, and they do. It's not biased, it's proactively unbiased. The numbers are what the numbers are, +/- error.
*Citations: The Literary Digest Assessing The Representativeness Of Public Opinion SurveysAge Demographics & Evening Survey Calls
0 Likes
Quoting: sevenits summer, pretty normal it dipsWin and Mac users don't take vacations ? :)
0 Likes
oldrocker99 Aug 4, 2014
1.1% of Steam users is still a very large number of people, which is why devs have been eager to tap this (small, but enthusiastic) market. It's few devs who regret having released games for Linux, and Gabe Newell is still pushing hard for SteamOS games (which, from all comments, run on just about every distribution: Debian and all its derivatives, Fedora and its, Arch and its, and even Slackware.
So much for the FUD known as "too many different distros."
So much for the FUD known as "too many different distros."
0 Likes
I got the survey yesterday after months of not seeing it.... and I was using WINE D:
1 Likes, Who?
Quoting: FutureSutureQuoting: ShmerlI do not follow. All I have in my package manager regarding LSB is lsb-release and Steam is able to tell me what operating system version I am running.Quoting: liamdaweWow, so Arch could possibly be Steam's number 2 distro right now. That's a lot of users.Debian also doesn't install lsb by default. At least I didn't have it in my Debian testing. It has lsb-release and lsb-base though.
It's enough then. Debian has lsb-release as well. lsb is a meta package that pulls in more related stuff (like rpm and etc.). I just answered to a previous post which mentioned lsb package specifically.
0 Likes
oldrocker99 Aug 4, 2014
Inasmuch as .deb files are archives, and can be extracted with the contained directories intact, and considering that some non-Debian distros include dpkg, it is possible to install Steam on just about any distro, as proven by happy Slackware customers
0 Likes
Obvisouly I don't have any data to back any of this up but it's what makes sense to me from my experience, that said.. In the wake of steam being released for linux as well as steamOS betas there were many who tried them and then have still been forced back to windows for most of their gaming needs. It's great what valve is doing and it has helped immensly, but 600 old and indie titles is obviously not going to get AAA title gamers to switch to linux fulltime. I play some other things but most of my gaming is AAA shooter type titles. Of my 140 games I own on steam.. 21 of them are available for linux and good chunk of those are valve source engine titles that do no work with my 5760x1080 res in linux.. ive posted bug reports to their github about it and it's just not received much attention. Im assuming because it doesn't affect the main focus of their steamOS and steambox's being the single big tv in your living room.
I do like that some major game engines are now being built with linux in mind such as UE4 and from what I have seen from the demos runs well and looks damn good. Unity is a pretty good engine but they still need to iron out some flaws in their linux implementation. Metro: LL runs very well for me in linux but does not look anywhere near as good as it's windows DX11 counterpart. Serious Sam 3 is in the same boat as it runs well but doesnt look as good as it's windows counterpart. While I am glad they have native linux ports and they run well, I built this machine because I enjoy the eye candy in games as much as the gameplay. If gameplay was the only factor in my gaming I would play on a console.
All that said though, for the first time ever, I have more games installed in linux than I do in windows. It's a great feeling really, but the reason is because I am playing some older titles and some others that I only tried because of their linux ability. I have found some games that I enjoy that I otherwise would have not tried. I know that opengl has the ability to spit out dx11 level graphics as the UE4 demos look really damn good. When the larger (AAA) game studios start using these technologies to get on par with the dx11 versions and really get on board with the linux movement, I believe adoption will skyrocket. I have HIGH hopes for 2015 personally.
EDIT* I REALLY wish more studios would open source the engines for their older titles and let the awesome linux community sort it out if they are not going to (in any kind of timely manner). The bioshocks, fallouts, dead spaces, F.E.A.R's, are all ones that quickly come to mind. Let us compile the binary ourselves and then copy the content from the disc or from steam.
I do like that some major game engines are now being built with linux in mind such as UE4 and from what I have seen from the demos runs well and looks damn good. Unity is a pretty good engine but they still need to iron out some flaws in their linux implementation. Metro: LL runs very well for me in linux but does not look anywhere near as good as it's windows DX11 counterpart. Serious Sam 3 is in the same boat as it runs well but doesnt look as good as it's windows counterpart. While I am glad they have native linux ports and they run well, I built this machine because I enjoy the eye candy in games as much as the gameplay. If gameplay was the only factor in my gaming I would play on a console.
All that said though, for the first time ever, I have more games installed in linux than I do in windows. It's a great feeling really, but the reason is because I am playing some older titles and some others that I only tried because of their linux ability. I have found some games that I enjoy that I otherwise would have not tried. I know that opengl has the ability to spit out dx11 level graphics as the UE4 demos look really damn good. When the larger (AAA) game studios start using these technologies to get on par with the dx11 versions and really get on board with the linux movement, I believe adoption will skyrocket. I have HIGH hopes for 2015 personally.
EDIT* I REALLY wish more studios would open source the engines for their older titles and let the awesome linux community sort it out if they are not going to (in any kind of timely manner). The bioshocks, fallouts, dead spaces, F.E.A.R's, are all ones that quickly come to mind. Let us compile the binary ourselves and then copy the content from the disc or from steam.
2 Likes, Who?
Patreon
PayPal
See more from me