No announcement yet.

A First Look At The 2010 Linux Graphics Survey Results

  • Filter
  • Time
  • Show
Clear All
new posts

  • A First Look At The 2010 Linux Graphics Survey Results

    Phoronix: A First Look At The 2010 Linux Graphics Survey Results

    Earlier this month we started once again our annual Linux Graphics Survey in which we poll our readers about their choices and opinions concerning graphics cards, display drivers, and other graphics / X.Org related features of the Linux desktop. While this survey is still going on through the end of September -- so you still have time to participate -- here are the results from the first 6,300 people to submit their responses. We are publishing the results so far since there is the X Developers' Summit this week in Toulouse and some of these findings may prove to be useful during those discussions.

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    "Only about 5% users are building from Git snapshots."

    Just wanted to point out that I selected "distribution supplied / repository", even though the package manager of my distro fetches sources from Git, builds and installs those. It still counts as "repository".

    So in other words, the survey assumes that distro supplied is different from Git sources.


    • #3
      Michael, The pie charts you provided was more difficult to read than just plain text.
      Could you refrain from using pie charts in the future?


      • #4
        Figures flawed?

        Hmmm... strange. One should expect that you get the amount of usage of graphic hardware brands when adding up the percentages of the specific drivers. Yet, this does not work, at least for Intel and AMD hardware. Even when adding the figures for the VESA driver, the percentage of drivers which work with AMD (Intel) hardware doesn't reflect the percentage of AMD (Intel) hardware in the pie plot above -- it's far too low.


        • #5
          I find some of the results to be very surprising.

          20% of the users use Linux for "enthusiast gaming"? This can't be right.

          Similarly, less than 10% of the users caring about licensing would be very depressing if correct.


          • #6
            Remember that some (or all?) of the questions allowed multiple answers. I think I ticked licensing but it wasn't the only one I ticked. I think I even chose enthusiast gamer along with mainstream user. I mainly play older games on Linux but I'd like to play newer ones too if the driver situation improves.


            • #7
              Then those pie charts are flat-out wrong and misleading. Pie charts can only be used if the percentages sum up to 100.


              • #8
                I agree. For most of the questions, it is likely that only one answer was given. Fair enough. But statements like "Less than 10% of the users are interested in licensing." are just completely false. Michael, even if the charts are misleading, at least get your facts straight in the text.


                • #9
                  ok, survey is done. whats now? will you send it to bridgman or something?


                  • #10
                    This would have been an opportunity to cross-reference data - this is unfortunately not possible with the reduced charts. Example:
                    - how many users of multiple displays actually want hot-plug? How many need hot-plug, but use only 1 monitor (i.e. plug in the projector for a presentation)?
                    - compare the percentage of binary vs. oss users by interests. Are people interested in video accel more likely to use binary drivers? Do "professional" users prefer the performance of binary drivers or the robustness of the OSS stack?
                    - are gamers more likely to pick nvidia over ati for wine gaming? By how much? Are there gamers with intel hardware?

                    Write a script to cross-reference everything that sounds reasonable and make a graph for each (labeled "GPU vendor for people who picked 'gaming enthusiast'" etc). Pick the graphs that deviate from the global average and point out the differences. I'm sure there are plenty of interesting results in there.