No announcement yet.

Intel Sandy Bridge Performance Goes Up Again

  • Filter
  • Time
  • Show
Clear All
new posts

  • Intel Sandy Bridge Performance Goes Up Again

    Phoronix: Intel Sandy Bridge Performance Goes Up Again

    At the beginning of the month I reported on a small patch to Mesa that resulted in a huge performance improvement for Intel Sandy Bridge graphics, but Intel's OSTC developers have bumped up the performance of the latest-generation graphics processors even more. With the LLC caching patch-set, which should be committed to the Linux 2.6.40 kernel (not the current 2.6.39 cycle), there are measurable OpenGL performance improvements.

  • #2
    I am sorry AMD. I am fan of yours, but Llano is too late now. When all problems resolve and motherboards become mainstream it will be end of the year already. I will get one of those i3-2100, since I am not much of a gamer, but open video acceleration is much of importance for me.
    I wish Intel, keep the good work with all open source projects, and keep evolving Mesa

    I am still a fanboy, but not at all cost


    • #3
      I'm also cheering Intel. They may be the ones who bring OpenGL 3 to X land. And restart-free graphic switching, and, and, and.

      Unfortunately Intel seam to adopt OpenGL 4.1 really slowly


      • #4
        There is nothing stopping Ubuntu or any other distro from backporting these changes into what ever kernel they decide to ship


        • #5
          This looks extremely sexy.
          Im looking at selling my AMD rig and puchasing 2,5k.
          Opencore is not opensource, proprietary with always half-arsed opensource strategy is not opensource really.


          • #6
            I'm simultaniously happy about this and sad about the older intel gfx still sucking horribly. Is something like this not possible for them too?


            • #7

              Good. I was getting tired of all that AMD openness crap. "Yes, we're open. We're vevy vevy open." Yes, you are, dear. You are wide open.

              How come xf86-video-ati devs can't do some reverse engineering akin to nouveau? How come Fusion/Zacate don't have open source video acceleration much alike SandyBridge. Hell, even Broadcom managed to push open source acceleration code for their Crystal family. SRSLY, Whats wrong with you, guys?

              There. Now my steam pressure level is... lets say... tollerable.


              • #8
                Eyeballing the older Windows VS Ubuntu thread:

                It seems that the Linux driver is on par with and sometimes exceeding the Windows driver!


                • #9
                  i'm an amd fan and i would have to agree with everyone here that amd is just taking way too long. but, keep in mind even their windows drivers are bad. if they can't get their main audience some decent drivers when they release a product, they're not going to do it for linux either. until linux gets attention when products are released, i don't care about using anything new or fancy in it. the nice thing about linux is it's a lot more hardware efficient than windows, but, it takes longer for it to get there. if you don't have anything too demanding, just go with the newest compatible hardware.

                  i'm very happy to see intel putting in this much effort, and i didn't bother to check if Tgui was right but if so, its nice to know linux has competitive graphics performance.

                  considering all the kernel updates lately, i feel like linux should outperform windows in just about every test at this point, as long as we don't use amd or nvidia for the graphics tests. i'm aware ubuntu is considered the default distro due to the most popularity but i really wish phoronix would do separate tests on an optimized distro like gentoo or arch. ubuntu packs in so much unnecessary stuff.


                  • #10
                    Please Michael show more graphics hardware benchmarks as a reference, ati or nvidia (preferably both), this way would be easy to see how the performance of SB compares to different hardware / driver.