Announcement

Collapse
No announcement yet.

Steam Linux Usage Drops During August

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    Originally posted by Bucic View Post
    Every decent portal should boycott Valve's 'statistics' until they publish proper, transparent statistics.
    *COUGH*There is a bug keeping Linux users form getting surveyed https://github.com/ValveSoftware/ste...ux/issues/2286 *COUGH*

    *COUGH*Larabel never does any damn research*COUGH*

    *COUGH*You're a fool to trust anything he posts on face value alone*COUGH*

    I really need to see a doctor about this cough...

    Comment


    • #32
      well, I use dual-boot, and something I've noticed is that last month they asked me for information while I was using linux and I gave it, but when I was using windows they does not ask me any info. This month, they asked me for information when I was using steam in windows, but when I was using linux steam did not asked me anything, which leads me to believe that there are some problems getting information of PC's that use dual-boot.

      Comment


      • #33
        No, you can use Xorg-Edgers instead of updating your whole distro

        Originally posted by johnc View Post
        Do you really have to upgrade the whole OS just to get the latest video driver?

        This is a serious weakness if true.
        If your distro isn't so old as to be unsupported by the Xorg-Edgers PPA, you can run the current Mesa drivers and Xserver using it. Also install the latest kernel.
        You can install later kernels in older distros. If you are just after kernel features (AMD power management, etc) you may not need to change Mesa or X. For max
        performance you want the latest versions of all three.

        You MIGHT be able to get away with using a current Mesa/X/kernel in a distro too old for the Xorg-Edgers PPA to
        support (add the oldest xorg-edgers PPA entry to /etc/apt/sources.list), but if you do this be SURE you can run ppa-purge from the console and deal with a broken
        X server before trying this in an unsupported older version. For that matter, Xorg-Edgers does occasionally get a bad update, be sure you know what you are doing
        when using PPAs in general. Since I have system backup partitions and snapshot images, I have little to worry about even if something gets REALLY screwed up. If you hack
        your install you should do the same.

        I used the PPA versions of X and Mesa for years, ever since my Evergreen (Radeon HD 5***) cards were not yet supported in Ubuntu Natty/early development Oneiric in 2011.

        Comment


        • #34
          Originally posted by johnc View Post
          The results are pretty pathetic but not surprising. It seems to match the overall desktop trend. It's interesting to see that the unreleased Windows 8.1 has about as much marketshare on Steam as all of Linux combined.
          Devs and pirates. It's not uncommon for pirates to take the dev releases from Microsoft and strip out the self lockout after it's EOL date.

          I've done this for those pesky BIOS updates on OEM mobos that only seem to work from within Windows.

          Comment


          • #35
            Assuming it is a good random survey with a good methodology, small groups (those under 10% for most surveys) will tend to have a large variation in reported numbers and those figures will be less reliable than the larger groups. That coupled with the 0.6% or so under the other category (most of which is lilkely a linux of some type) makes the month to month changes hard to interpret.

            Likely causes in decline are bad graphics drivers for everyone not using the NVIDIA binary blobs, lack of games (AAA titles), and decent windows only titles dragging many users back to wine and dual boot (most real gamers will dual boot for the best experience of a game sory wine). Lastly most linux folks are older males, it is summer, we still work (http://www.pcworld.com/article/25301..._insight_.html). A large proportion of the windows gaming crowd are younger people who have the summer off and have way more time to play games. An influx of windows users due to seasonality can also shift those numbers.

            The least likely cause mentioned in the thread is the Unity/Gnome Shell issue. Ubuntu alone makes up aprox 75% (.75% of the less than 1%) of reported linux steam users. If that was an issue, Ubuntu's domination of the steam linux crowd would not be so complete.

            Comment


            • #36
              Originally posted by Luke View Post
              If your distro isn't so old as to be unsupported by the Xorg-Edgers PPA, you can run the current Mesa drivers and Xserver using it. Also install the latest kernel.
              You can install later kernels in older distros. If you are just after kernel features (AMD power management, etc) you may not need to change Mesa or X. For max
              performance you want the latest versions of all three.

              You MIGHT be able to get away with using a current Mesa/X/kernel in a distro too old for the Xorg-Edgers PPA to
              support (add the oldest xorg-edgers PPA entry to /etc/apt/sources.list), but if you do this be SURE you can run ppa-purge from the console and deal with a broken
              X server before trying this in an unsupported older version. For that matter, Xorg-Edgers does occasionally get a bad update, be sure you know what you are doing
              when using PPAs in general. Since I have system backup partitions and snapshot images, I have little to worry about even if something gets REALLY screwed up. If you hack
              your install you should do the same.

              I used the PPA versions of X and Mesa for years, ever since my Evergreen (Radeon HD 5***) cards were not yet supported in Ubuntu Natty/early development Oneiric in 2011.
              Ahh, thanks for the info. Whenever I use xorg-edgers I'm always careful to only target the package I need, and then disable the PPA. lol. This protects me from accidentally installing a beta driver or something.

              Comment


              • #37
                Originally posted by Kivada View Post
                *COUGH*There is a bug keeping Linux users form getting surveyed https://github.com/ValveSoftware/ste...ux/issues/2286 *COUGH*

                *COUGH*Larabel never does any damn research*COUGH*

                *COUGH*You're a fool to trust anything he posts on face value alone*COUGH*

                I really need to see a doctor about this cough...
                Valve hasn't confirmed any such bug yet. It's just in review.

                Comment


                • #38
                  Originally posted by johnc View Post
                  Do you really have to upgrade the whole OS just to get the latest video driver?

                  This is a serious weakness if true.
                  Of course you don't.

                  Comment


                  • #39
                    Originally posted by Daktyl198 View Post
                    Don't forget about the whole "The survey doesn't pop up on Linux" issue for people who have an install on both Windows and Linux.

                    There are several people who report the survey popping up when they log into steam on Windows several times (not filling it out) but it never once popping up on Linux, even though their "account" was selected.
                    I have linux only steam installation, and last time i got survey pop-up was in january.

                    Comment


                    • #40
                      Originally posted by schmidtbag View Post
                      Seriously, are all of you running P4s?
                      Jup... my p4 runs better than ever with Linux, thank you.

                      (steam not so much... still waiting for pacman to arrive)

                      Comment


                      • #41
                        Originally posted by allenmaher View Post
                        Assuming it is a good random survey with a good methodology, small groups (those under 10% for most surveys) will tend to have a large variation in reported numbers and those figures will be less reliable than the larger groups. That coupled with the 0.6% or so under the other category (most of which is lilkely a linux of some type) makes the month to month changes hard to interpret.

                        Likely causes in decline are bad graphics drivers for everyone not using the NVIDIA binary blobs, lack of games (AAA titles), and decent windows only titles dragging many users back to wine and dual boot (most real gamers will dual boot for the best experience of a game sory wine). Lastly most linux folks are older males, it is summer, we still work (http://www.pcworld.com/article/25301..._insight_.html). A large proportion of the windows gaming crowd are younger people who have the summer off and have way more time to play games. An influx of windows users due to seasonality can also shift those numbers.

                        The least likely cause mentioned in the thread is the Unity/Gnome Shell issue. Ubuntu alone makes up aprox 75% (.75% of the less than 1%) of reported linux steam users. If that was an issue, Ubuntu's domination of the steam linux crowd would not be so complete.
                        Hence why they should just count all active users in the last 30 days and what hardware/os they logged in on with hard numbers. As it stands a random sample can come back with 100% of the users being on Windows.

                        I've been using Steam since the public launch and have only ever used it on Linux and have never once been asked to join the survey even though I never turn off Steam.

                        Comment


                        • #42
                          Originally posted by Daktyl198 View Post
                          Don't forget about the whole "The survey doesn't pop up on Linux" issue for people who have an install on both Windows and Linux.

                          There are several people who report the survey popping up when they log into steam on Windows several times (not filling it out) but it never once popping up on Linux, even though their "account" was selected.

                          *disclaimer* Just what I've heard. Don't kill the messenger...
                          Well that happened to me too. Some time ago, I had been using Steam for Linux for MONTHS. The survey didn't pop up a single time. One day I decided to install Steam on Windows to play some PC games I bought a while ago. The first time I started Steam on Windows the survey popped up. Just an hour before that I was using Steam for Linux.

                          Comment


                          • #43
                            My conscience is clear

                            I use Steam on Linux nearly daily. Since Dota 2 was released, I didn't boot to Windows even 1 time

                            But if it comes to purchasing new titles, there's not much to choose. I wanted to buy one title few days ago, but it was Windows/OSX only.

                            Disclaimer: I'm now writing this from Windows, but I'm at work, not at home

                            Comment


                            • #44
                              Originally posted by Kivada View Post
                              Hence why they should just count all active users in the last 30 days and what hardware/os they logged in on with hard numbers. As it stands a random sample can come back with 100% of the users being on Windows.

                              I've been using Steam since the public launch and have only ever used it on Linux and have never once been asked to join the survey even though I never turn off Steam.
                              I have been on since early beta, average 30+hours a week play time and have 86 games (with linux steam versions) 89 dlc... times surveyed on linux zero. I tried Dear Ester on steam/wine 1 time before the HIB linux version was released (crap frame rate on my laptop, never tried steam on wine again)... and the survey popped. I don't have a copy of windows to boot (and would not if I did) so I don't know what the prevalence of survey requests are. Could it be coincidence yes, but I see why people doubt it. Confirmation bias is hard to get past. My experience is an anecdote and may be an artifact of chance. A collection of anecdotes like you see on this thread carries as much weight as a single anecdote, which is to say none statistically speaking.

                              A 100% survey would bring up privacy issues (remember Ubuntu and amazon searches) and if it were optional would lead to a lot of anoyed gamers and a lot of non responses from privacy minded individuals (the tin foil hat club included). Does linux have more than it's share of tin foil hat wearing people? You read the forums here, I will let you decide on how to interpret the anecdotal collection of rants in here.

                              I want the numbers to go up rather than down... and they are not. So i am likewise disapointed.
                              "Statistics are used much like a drunk uses a lamppost: for support, not illumination." -- I found multiple attributions for the quote so I am not sure what the actual source is. But the point is clear, look to statistics to tell you what you don't know and don't want to know, because we already convince ourselves of what we think we know.

                              Comment


                              • #45
                                Originally posted by verde View Post
                                Ubuntu 13.10 is using 3.11 already..
                                K, then i will change to 13.10!

                                Comment

                                Working...
                                X