Announcement

Collapse
No announcement yet.

Here's Why Radeon Graphics Are Faster On Linux 3.12

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

  • #91
    AMD Phenom II now fine with "ondemand" governor even if Bulldozer is not

    Originally posted by Luke View Post
    On my FX8120, both Linux 3.11 and Linux 3.12 suffer a GPU performance hit when running the "ondemand" governor in Critter, the only non-CPU limited game I have.
    Earlier I posted my FX-8120 results showing that the difference between "performance" and "ondemand" governors has narrowed but is still significant. This evening I tested the results of the two governors on a Phenom II X4, a 955BE with a Radeon HD5570. Again the high-framerate, non CPU limited game "critter" was used, it can heat that little card all the way to 70C-and run at about 730fps peak on either governor setting.

    On that Phenom II, I found no discernable difference between leaving the governor in "ondemand" or putting it in "performance," no difference at all. Sometime I will have to see if that also applys to video edit rendering.

    Comment


    • #92
      Luke, while technically a graphically very simple GPU accelerated game might be well be non-CPU limited at the extreme, I'm not sure what you're trying to measure. It could well be you're hitting the GPU fillrate limit, exceeding the internal bandwidth of the card and at such high frame-rates I can't see how it's meaningful.

      Comment


      • #93
        To add my worthless opinion:

        Well, things do happen as they do. I never really wanted to speculate where it came from but had hoped that it would be the sum of changes in the free driver stack.
        Anyway.

        Michael probably used a really fast PC to limit bottlenecks from that side and to actually benchmark the GPU. But then it is of course wrong to use any non-performance governor (performance = which keeps the CPU untouched and at full speed).
        Still, for a real-world benchmark you just got to use ondemand since 99% of people will exactly use that one. Everything else would not make sense for all normal people. So at least that would be the effect most people would see. And it is good that a close look was done to find out what exactly caused it.

        So Michael you might learn from that and check with a full-speed-CPU governor again if something looks too strange. Or keep in mind any other possible larger influences on general performance. Still I think testing also on normal-world machines and normal configurations is a fair thing since most people do not have an i7+SSD+16 GB RAM or something combination.
        And please make these text boxes with the testing config larger.

        Comment


        • #94
          Originally posted by 89c51 View Post
          Average joe configuration should be used IMO. Also the user MUST NOT have to care about stuff like that. It should just work.
          Indeed, the OS should magically read average Joe's mind to determine what exactly he wants...

          /facepalm

          Comment


          • #95
            Originally posted by prodigy_ View Post
            Indeed, the OS should magically read average Joe's mind to determine what exactly he wants...

            /facepalm
            ...and apparently most of the settings in software are completely useless and should be eliminated. The default way or now way. Ever. No, not even in that one special case that would be important to _you_.
            Similar to the world of Gnome 3TM
            /sarcasm

            Comment


            • #96
              Originally posted by menfin87 View Post
              You are not interested as a driver developer ? So if a user complains about performance problems, do you just tell him "I don't see it so there is no problem" or do you maybe try to understand what is going on ? You should be glad that someone is testing an average user configuration to spot problems that would be blamed on you.
              If you set the governor to performance and send the benchmark results to me, I am interested. If you set the governor to the minimum CPU frequency, I am interested. If you set the governor to ondemand, I am not interested, because it would be a waste of time for me to try and reproduce your random results which might not be reproducible on my machine at all. In that case, you should probably be sending your results to the cpufreq team and not me.
              Last edited by marek; 10-16-2013, 07:31 AM.

              Comment


              • #97
                So the benchmarks that showed the improvements were with intel cpus: http://www.phoronix.com/scan.php?pag...12_major&num=2

                How come they even use ondemand? Has cpufreq not been superseded by pstate for several kernel versions now?

                Comment


                • #98
                  Ubuntu 13.10

                  It would be nice if Ubuntu can backport that commit in 13.10 (3.11 kernel).

                  Comment


                  • #99
                    Originally posted by marek View Post
                    If you set the governor to performance and send the benchmark results to me, I am interested. If you set the governor to the minimum CPU frequency, I am interested. If you set the governor to ondemand, I am not interested, because it would be a waste of time for me to try and reproduce your random results which might not be reproducible on my machine at all. In that case, you should probably be sending your results to the cpufreq team and not me.
                    Perfectly fair request IMO. But, the underlying issue is the CPU should not have downclocked this much in the first place. I'm not saying this is the radeon driver's fault - I'm pretty certain this was an issue with cpufreq this entire time. In Windows or Mac, you don't have to mess with the governor (or whatever they call it) and their tests, to my knowledge, are not skewed. I understand why you don't want test results of ondemand, but the matter of the issue is that shouldn't be skewing the results in the first place. In other words, while disabling ondemand proves more reliable results, the tests should come out overall the same with ondemand enabled. I'm thinking kernel 3.12 is the first time the ondemand governor is working the way it SHOULD have.

                    On the other hand, as ChrisXY has mentioned, many of these governors are obsoleted anyway.

                    Comment


                    • Originally posted by ChrisXY View Post
                      So the benchmarks that showed the improvements were with intel cpus: http://www.phoronix.com/scan.php?pag...12_major&num=2

                      How come they even use ondemand? Has cpufreq not been superseded by pstate for several kernel versions now?
                      On any OTHER distro (not *Buntu based) Yuuuuuuup. But Canonical hasn't gotten their heads out of their asses yet and switched yet. Someone said that it was because of bugs in the driver but im finding it hard to believe that THEY are hitting bugs that no one else has.

                      Comment


                      • Originally posted by Ericg View Post
                        On any OTHER distro (not *Buntu based) Yuuuuuuup. But Canonical hasn't gotten their heads out of their asses yet and switched yet. Someone said that it was because of bugs in the driver but im finding it hard to believe that THEY are hitting bugs that no one else has.
                        There was a bug in kernel 3.10 where the intel_pstate driver used the turbo frequency as the base performance frequency, and that used to heat things up more than it should have. But that issue was solved in 3.11, but Ubuntu hasn't enabled it back just yet.

                        Comment


                        • Originally posted by Michael View Post
                          Then it should be fixed or the default setting changed within the kernel.
                          Agreed! Everyone I know uses the default "ondemand" governor. While I'd like to use the 'performance' governor, the power consumption prevents me from doing so. Almost every user is going to be using 'ondemand'. A good user experience demands that things just work with the default settings. If thats not the case, then someone needs to change the defaults!

                          Comment


                          • Originally posted by GreatEmerald View Post
                            And will make your FPS dip to 30 if 60 can't be sustained, instead of just 59... No, you should have VSync on only for games you know will never dip below 60 (or whatever your refresh rate may be).
                            Umm. No. It would probably be 58 fps not 30. I know cause I had NS2 and with vsync on the fps ranged from 45 - 60 fps. Vsync just means that it wont display partially rendered frames, it will finish rendering the current frame, display it, and then render the most current frame, display it, etc, etc.

                            Comment


                            • Originally posted by GreatEmerald View Post
                              There was a bug in kernel 3.10 where the intel_pstate driver used the turbo frequency as the base performance frequency, and that used to heat things up more than it should have. But that issue was solved in 3.11, but Ubuntu hasn't enabled it back just yet.
                              Interesting that they hit such a bug, I didn't hit that bug on Arch. Wonder if it got pulled in when they were making their 'changes' to the kernel.

                              Comment


                              • Originally posted by gururise View Post
                                Agreed! Everyone I know uses the default "ondemand" governor. While I'd like to use the 'performance' governor, the power consumption prevents me from doing so. Almost every user is going to be using 'ondemand'. A good user experience demands that things just work with the default settings. If thats not the case, then someone needs to change the defaults!
                                It's not a default setting in the kernel. It's a setting forced by a startup script in Ubuntu and you can't turn it off (unless you delete the script manually). So this is obviously Canonical's fault. Too bad we didn't get benchmarks with multiple distributions, we'd at least see how bad Canonical is. Generally, "ondemand" is okay for most people, giving you approx. 75-80% performance out of your CPU. You'll never get 100% unless you use up all CPU cores (very unlikely for most people, even gamers). However, there are applications where ondemand is absolutely unacceptable. Those are gaming, benchmarking, and running computationally-intensive applications where the running time matters (compiling too).

                                Comment

                                Working...
                                X