Announcement

Collapse
No announcement yet.

Intel SNA Performance Continues To Be Compelling

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

  • Intel SNA Performance Continues To Be Compelling

    Phoronix: Intel SNA Performance Continues To Be Compelling

    Due to the extreme pace at which Chris Wilson has been releasing SNA architecture updates for Intel's open-source X.Org driver, here are another set of benchmarks of Intel Sandy Bridge HD 3000 graphics when comparing UXA and SNA using yesterday's Git code following the xf86-video-intel 2.20.2 driver release.

    http://www.phoronix.com/vr.php?view=17664

  • #2
    Cool but a reference point would be nice.

    Hey Guys,

    It is great to see the relative gains that Sandy Bridge / Ivy Bridge whatever are doing. And much love to the dudes squeezing all kinds of performance out of it with a new software architecture.

    I would however like to see some kind of baseline against the usual suspects in discrete gpus. Nothing fancy, just the bottom of the range from AMD and Nvidia to give it all some kind of perspective.

    Cheers,
    t

    Comment


    • #3
      That would be embarassing the AMD and Nvidia GPUs. Intel + SNA should beat both regardless of driver.

      Comment


      • #4
        Originally posted by tenzero View Post
        I would however like to see some kind of baseline against the usual suspects in discrete gpus. Nothing fancy, just the bottom of the range from AMD and Nvidia to give it all some kind of perspective.
        I can guarantee you that intel GPUs offer a much better experience than AMD's chipset integrated GPUs. I have no idea how the comparision is like with the newer Llano iGPUs, which is something I would also like to know, but the difference was massive with the chipset ones. That is in real-world desktop usage not synthetic benchmarks.

        Comment


        • #5
          Llano has ~10x the shader power of the fastest IGP (5 x 16-way SIMD vs 2 x 4-way SIMD), and Trinity is higher again (6 x 16-way SIMD). The IGPs are comparable to RV610/620 but without dedicated video memory.

          Comment


          • #6
            Shader power doesn't help driver fallbacks

            Comment


            • #7
              It helps you get to them a lot faster

              That said, if the context here is X-style 2D operations then I agree that is going to be more a function of effective memory bandwidth (2-3x), driver design, and other hardware features (eg 2D accel HW) than shader power (10x). What makes it hard to predict is that desktop operations are increasingly going through 3D APIs (where the performance dependencies are a bit different) rather than X drawing operations, so generalizing is becoming more problematic every year.
              Last edited by bridgman; 07-28-2012, 03:01 PM.

              Comment


              • #8
                Originally posted by devius View Post
                I can guarantee you that intel GPUs offer a much better experience than AMD's chipset integrated GPUs. I have no idea how the comparision is like with the newer Llano iGPUs, which is something I would also like to know, but the difference was massive with the chipset ones. That is in real-world desktop usage not synthetic benchmarks.
                You missed the word "discrete" in the post you quoted. And I assume he meant with proprietary drivers.

                Comment


                • #9
                  @Michael it looks like your CPU was being throttled during those tests, the dmesg does indeed some throttling due to exceeding its safe temperature.

                  For example, on your machine both UXA and SNA should score over 3000 op/s for -putimage500 (even higher for shorter benchmark runs due to turbo). And there are several other cases where your machine underperformed by a factor of 2-3x, consistent with throttling - or another CPU hog running. What would be fantastic were if pts were automatically able to perform system profiling in conjunction with running the benchmark, which would be both a boon for people trying to understand a test result and for spotting anomalous runs.

                  Comment


                  • #10
                    Originally posted by GreatEmerald View Post
                    You missed the word "discrete" in the post you quoted. And I assume he meant with proprietary drivers.
                    I know, but I was just talking about comparable products because that's what I do know. I have read that even some powerful discrete AMD GPUs have problems providing a smooth desktop experience but still, a Radeon HD4200 theoretically has about the same performance as a Intel HD2000, but in reality the intel GPU provides a much better desktop experience. This is irrespective of which drivers are being used on the Radeon (binary or open source, doesn't make a difference in terms of desktop compositing).

                    Comment


                    • #11
                      Originally posted by devius View Post
                      This is irrespective of which drivers are being used on the Radeon (binary or open source, doesn't make a difference in terms of desktop compositing).
                      Well, it does make a difference for me on KDE. KWin won't work with fglrx and will work with radeon (though this is on openSUSE 12.1, I think they made improvements in later KDE versions).

                      Comment


                      • #12
                        Originally posted by tenzero View Post

                        I would however like to see some kind of baseline against the usual suspects in discrete gpus. Nothing fancy, just the bottom of the range from AMD and Nvidia to give it all some kind of perspective.
                        Here you go, IvyBridge (i7-3720qm) in comparison with an Nvidia GTX-550: http://openbenchmarking.org/result/1...SU-1207273SU39

                        Comment


                        • #13
                          Originally posted by ickle View Post
                          Here you go, IvyBridge (i7-3720qm) in comparison with an Nvidia GTX-550: http://openbenchmarking.org/result/1...SU-1207273SU39
                          Very nice. Basically SNA = Good, UXA = Bad. When SNA has regressions they are negligible, but when it performs better, it really performs a lot better.

                          Comment


                          • #14
                            Originally posted by devius View Post
                            Very nice. Basically SNA = Good, UXA = Bad. When SNA has regressions they are negligible, but when it performs better, it really performs a lot better.
                            Right, the regressions tend to be a consequence of choosing one method that gives the better performance elsewhere at a cost. Most of the regressions are in the noise of the measurement, IvyBridge is very sensitive to thermals (in some of those tests the initial run is 2x faster than the final run due to turbo). The only significant regression there is -compwinwin500. The reason for the regression is that last week it was 2x faster due to hitting the Render cache - however that was missing a flush. Having added that flush for correctness, it becomes faster to use the BLT for that particular test, a trivial change already made.

                            But what I find truly fascinating is how competitive we actually are with a discrete GPU that has a good driver, over 4x the fill rate of the igfx and several times the shader flops. With regards to 2D performance the limitation tends not to be SNA (unlike UXA and glamor where they are the bottleneck), but the application - which is as it should be. :-)

                            Comment


                            • #15
                              Originally posted by ickle View Post
                              unlike UXA and glamor where they are the bottleneck
                              Is this statement related to intel hardware only or do you think there are general (significant) bottlenecks connected to Glamor?

                              Comment

                              Working...
                              X