Announcement

Collapse
No announcement yet.

Radeon Driver Commit Activity Is On The Decline

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

  • Radeon Driver Commit Activity Is On The Decline

    Phoronix: Radeon Driver Commit Activity Is On The Decline

    After delivering development statistics on the Nouveau driver and the Intel driver, here's some numbers looking at the development pace of the xf86-video-ati X.Org driver for Radeon graphics cards...

    http://www.phoronix.com/vr.php?view=MTE0NjE

  • #2
    Quality journalism is on the decline

    Code:
    _________
    |\       |
    |  \     |
    |    \   |
    |      \ |
    |_______|
    Last edited by FireBurn; 07-25-2012, 07:10 AM.

    Comment


    • #3
      Michael, the watermarking on all phoronix images is a disaster. Please update the image to be semitransparent (~30%) instead of solid and it would make sense to make it smaller size, or make it occupy a maximum of 20% of image proportionally. :/

      This is because, for many images you post, the "water"mark very often overlaps over information.

      Example from your post: http://www.phoronix.com/image-viewer...6_ati_git1_lrg

      Thank you!

      Comment


      • #4
        Originally posted by crazycheese View Post
        Michael, the watermarking on all phoronix images is a disaster. Please update the image to be semitransparent (~30%) instead of solid and it would make sense to make it smaller size, or make it occupy a maximum of 20% of image proportionally. :/

        This is because, for many images you post, the "water"mark very often overlaps over information.

        Example from your post: http://www.phoronix.com/image-viewer...6_ati_git1_lrg

        Thank you!
        I second that; it's ridiculous sometimes.

        Comment


        • #5
          All benchmark graphs are completely broken on my iPhone. Have been forever.

          Comment


          • #6
            That they are most likely on the decline, because the bulk of the code has been written. Support for an entire generation of cards requires a lot of code. Adding support (may be only pci-ids) and fixing bug is far smaller and less frequent commits?

            Also isn't r300g nearly done now? Hasn't r300 classic and r600 classic been dropped now? Would also reduce the amount of commits.

            I could be wrong of course.

            Comment


            • #7
              Originally posted by oliver View Post
              That they are most likely on the decline, because the bulk of the code has been written. Support for an entire generation of cards requires a lot of code. Adding support (may be only pci-ids) and fixing bug is far smaller and less frequent commits?

              Also isn't r300g nearly done now? Hasn't r300 classic and r600 classic been dropped now? Would also reduce the amount of commits.

              I could be wrong of course.
              That was my thought. Commit activity looked more or less like your normal standard distribution curve, which you'd expect to see in regards to driver activity over time.

              Comment


              • #8
                Yeah, I think there were really three phases :

                - steady state with UMS (medium activity)
                - transition to KMS (really high activity)
                - steady state with KMS (low activity since the rest of the work moved to kernel driver)

                Comment

                Working...
                X