Announcement

Collapse
No announcement yet.

When It Works, Intel Core i5 2500K Graphics On Linux Are Fast!

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

  • When It Works, Intel Core i5 2500K Graphics On Linux Are Fast!

    Phoronix: When It Works, Intel Core i5 2500K Graphics On Linux Are Fast!

    After a month of headaches for Intel and myself, there are now Sandy Bridge graphics benchmark results from the Intel Core i5 2500K under Linux to finally publish. Sandy Bridge was a tough launch for Intel in terms of the Linux coverage with the media having problems building a working driver stack and then when I finally got my hands on a CPU, I ran into an entirely different set of show-stopping problems. The developers still have not solved the biggest original issue yet, but Intel sent out a new motherboard and another CPU and it happens to "just work" nicely under Linux. When using the latest bits of their open-source Intel Linux graphics code, the performance on the Core i5 2500K is actually quite impressive compared to other open-source Linux drivers.

    http://www.phoronix.com/vr.php?view=15669

  • #2
    Comparison with current-gen Intel IGP?

    Michael, would be great if you could do a comparison with current Core IronLake IGP to have a better idea what to expect when upgrading from current gen to SandyBridge.

    With my Core i7 920 (IronLake), I noticed a HUGE improvement between 4x and 10x in some GL apps after upgrading to Mesa-git compared to Mesa 7.9.

    Tip for mesa-git adventurers: if not building in /usr/lib, don't forget to set LIBGL_DRIVERS_PATH otherwise the older DRI driver will still get loaded by the newly built libGL (which works but do not give *that* big performance increase).

    Comment


    • #3
      How did you test integrated graphics with an ASUS P8P67-M PRO board? P67 has got no graphics support at all.

      Comment


      • #4
        Originally posted by Kano View Post
        How did you test integrated graphics with an ASUS P8P67-M PRO board? P67 has got no graphics support at all.
        Err P8PH67.
        Michael Larabel
        http://www.michaellarabel.com/

        Comment


        • #5
          That does not exist, maybe a P8H67-M PRO? Did you try booting via UEFI, that's what i would have done.

          Comment


          • #6
            Goodness. 12-self-referencing links in six consecutive paragraphs. This is a record to my knowledge, even for phoronix.

            Comment


            • #7
              Neat article. I'm not sure how to interpret the last (OpenBenchmarking.org) graph though.

              Comment


              • #8
                Originally posted by mattst88 View Post
                Neat article. I'm not sure how to interpret the last (OpenBenchmarking.org) graph though.
                Because there is nothing else like that on the net... It's just a tease till end of month when the heatmap to be explained
                Michael Larabel
                http://www.michaellarabel.com/

                Comment


                • #9
                  Originally posted by Michael View Post
                  Because there is nothing else like that on the net... It's just a tease till end of month when the heatmap to be explained
                  I'm sure there are lots of meaningless graphs on the net Michael

                  I'm almost sure this is showing where the performance is relative to other cards. If the red line is on the right does that mean it's the best card benchmarked?

                  Trying to figure out what the black lines mean though

                  Comment


                  • #10
                    Don't buy the Intel 6-series H67/P67 chipsets!

                    The problem occured in Intelís 6-series H67/P67 chipsets that has two sets of SATA ports. One set handles four 3 Gbit/s drives and the other works with a pair of 6 Gbit/s drives. The transistor of concern is located in the PLL (phase lock loop) clock tree of the 3 Gbit/s controller. The circuit was biased at too high a voltage for the design and this resulted in an excessively high leakage current. This in turn changes the system's characteristics and causes the controller to fail. The other controller is unaffected as well as it has its own PLL.

                    Comment


                    • #11
                      Originally posted by FireBurn View Post
                      I'm sure there are lots of meaningless graphs on the net Michael

                      I'm almost sure this is showing where the performance is relative to other cards. If the red line is on the right does that mean it's the best card benchmarked?

                      Trying to figure out what the black lines mean though
                      I guess this is quantiles at 33% and 66%
                      so when the red line is between the two black lines, you can say that's a mid-range hardware
                      I think it's quite interesting ! espacially because those quantiles are not stable at all across the different tests
                      OpenBenchmarking.org is going to be a great playground for wannabe statisticians !

                      Comment


                      • #12
                        Originally posted by xav_19 View Post
                        I guess this is quantiles at 33% and 66%
                        so when the red line is between the two black lines, you can say that's a mid-range hardware
                        I think it's quite interesting ! espacially because those quantiles are not stable at all across the different tests
                        OpenBenchmarking.org is going to be a great playground for wannabe statisticians !
                        Yes.

                        There are markers at the 33% and 66% percentiles. That represents (from all the results that Openbencharking.org has - ie: people who are playing with PTS3).

                        So from that graph, you can immediately say that the system (and software used) is a mid-range system (between the 33% and 66% percentile). You can also see that the system is a poor choice against other hardware in the GLSL based games/benchmarks (nexuiz and lightsmark).

                        Yes, people wanting stats are going to have a field day .

                        Comment


                        • #13
                          Originally posted by FireBurn View Post
                          I'm sure there are lots of meaningless graphs on the net Michael

                          I'm almost sure this is showing where the performance is relative to other cards. If the red line is on the right does that mean it's the best card benchmarked?

                          Trying to figure out what the black lines mean though
                          They are 33 & 66 percentile. The shades represent frequency of a particular result (white being more common). It allows a human to easily make a "low/medium/high" judgment really quickly. Since it effectively normalizes the data as well, we can turn around and make some generalizations of a system against the population of systems currently out there.

                          Yes, a percentile rank of 100 (far right) indicates that it is one of the fastest systems out there for that benchmark.

                          Comment


                          • #14
                            How is a barely mid-range card "terrific"? Some of the other cards running mesa beat it handily; the ancient r300 card beats it by a long shot every time.

                            There are some chipsets under some open source graphics drivers for which I'd call the performance for some games "terrific", but I didn't see any of that come out of the Intel IGP in these tests....

                            Comment


                            • #15
                              Originally posted by allquixotic View Post
                              How is a barely mid-range card "terrific"? Some of the other cards running mesa beat it handily; the ancient r300 card beats it by a long shot every time.

                              There are some chipsets under some open source graphics drivers for which I'd call the performance for some games "terrific", but I didn't see any of that come out of the Intel IGP in these tests....
                              Historically, integrated graphics have been horrific, and would barely hold up against the lowest of the low discrete cards. In some benchmarks (quake3 based), the system performs faster than some reasonably capable cards with the OSS drivers. With one particular test, vdrift, the system is faster than some of cards with Catalyst drivers.

                              For something that is effectively given away for free, it's done a pretty good showing.

                              Further, when looking at the last page, the card sits roughly in the middle for a lot of the results in openbenchmarking.org. Historically, they would all be "Low" performing systems in the bottom 33 percent.

                              Comment

                              Working...
                              X