Announcement

Collapse
No announcement yet.

Previewing The Radeon DPM Performance On Linux 3.11

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

  • #16
    Originally posted by Serafean View Post
    PS : A special compliment goes to bridgman for concentrating on development and not on tweaking the old PM code.
    Actually that credit should go to Alex (and probably Dave too). I wanted us to tweak the old PM code a bit more to improve things while we were trying to get approval for DPM release, but I kinda got outvoted.

    Originally posted by Serafean View Post
    Taking the flak from all of us must have been a bit harsh
    Yeah, guess I still get credit for taking the flak
    Last edited by bridgman; 07-16-2013, 03:01 PM.

    Comment


    • #17
      Originally posted by GreatEmerald View Post
      The performance difference shouldn't be nearly like that on dedicated cards. But it's not supposed to be, the DPM code is for energy consumption and thermal management.
      It's going to depend on the card. Guessing we'll see similar improvements for the GCN cards, since they also default to pretty low clocks, and *some* of the earlier discrete GPUs, particularly mobile parts.

      Comment


      • #18
        Gonna try this out later on my HD4200 IGP. Setting power profiles never seemed to stick and previous attempts at dpm was failure as well.

        Comment


        • #19
          Originally posted by Veerappan View Post
          Maybe that's why the article title started with "Preview".

          Also, looks like Urban Terror is bottlenecked somewhere other than the GPU (or at least the GPU/VRAM clocks didn't make much of a difference).
          That was the most interesting of the lot (besides seeing the general degree of improvement). At the 1080 there was only around a 15% difference, but at 1600 it jumped to better than 70%, with the dpm'd driver experience only a slight fps change between resolution.
          So the clock speed made the biggest difference when it comes to larger textures. I'm guessing this is down purely to the increased memory speed?

          Comment


          • #20
            Originally posted by Hamish Wilson View Post
            As a question, is there any major difference in performance between the new DPM code and the old static PM on high? And if so, will this apply to all the supported cards?
            There might be a difference with some GPUs that have boost states, although I don't know if any dedicated GPUs have them.

            Comment


            • #21
              finally, open source drivers that will work with a laptop (without melting it)

              Well, better late than never :-) I expect that in about 6 months when this starts making it into Linux distros that I will be able to start recommending AMD hardware again. I hope your CPU people have some tricks up their sleeves for catching up with Intel.

              Is CrossFire support coming next for those of us with both discrete components and APUs?

              Thank you!!
              Last edited by edsdead; 07-16-2013, 05:54 PM. Reason: typo

              Comment


              • #22
                A small open source gpu revolution!

                Probably the most exciting preview article of all times as far as it concerns the graphics world!
                Think now that the one of the 2 industry leaders has an out of the box open source almost full functional driver for most of its GPUs!
                I am very happy that the Lintel graphics are becoming obsolete and we now can use real GPUs from cheap and cool ones to APUs and to graphic hungry beasts!!
                I don't think someone with a Radeon HD 6970 would care that much for the power consumption... But he/she would care to have a driver that rises the clocks to maximum and provide high performance!
                Crossfire, Overdrive I think are the next goals for this category of GPUs and not only...
                We also have UVD usage among all the others, Radeon SI must be now under the heavy development procedure and inherit the features from its older brothers!

                Big thanks to all AMD open source team and every other contributor!

                As for nVidia, Linus words are always in my mind since they bury nouveau... We don't need their proprietary advantage and open source huge disadvantage anymore!
                Last edited by djdoo; 07-16-2013, 08:29 PM.

                Comment


                • #23
                  Originally posted by djdoo View Post
                  Probably the most exciting preview article of all times as far as it concerns the graphics world!
                  Think now that the one of the 2 industry leaders has an out of the box open source almost full functional driver for most of its GPUs!
                  I am very happy that the Lintel graphics are becoming obsolete and we now can use real GPUs from cheap and cool ones to APUs and to graphic hungry beasts!!
                  I don't think someone with a Radeon HD 6970 would care that much for the power consumption... But he/she would care to have a driver that rises the clocks to maximum and provide high performance!
                  Crossfire, Overdrive I think are the next goals for this category of GPUs and not only...
                  We also have UVD usage among all the others, Radeon SI must be now under the heavy development procedure and inherit the features from its older brothers!

                  Big thanks to all AMD open source team and every other contributor!

                  As for nVidia, Linus words are always in my mind since they bury nouveau... We don't need their proprietary advantage and open source huge disadvantage anymore!
                  I'm not totally disagreeing with you Djdoo... but there's n othing wrong with Intel graphics. They are more than enough for the average (non-gamer) user since Sandy Bridge. Some light gaming is fine on them. If you want to do some more serious gaming then yeah, you need AMD or Nvidia but to say that they are "obsolete" is definitely an overstatement. Not to mention that HAVING the discrete cards suck a fair bit of power which is bad for laptop users.

                  Comment


                  • #24
                    Originally posted by darkbasic View Post
                    Please bench a10-5800K because a10-6800k is way too immature to be interesting.
                    They are almost exactly the same chip with different clocks. Both use r600g, not radeonsi.

                    Comment


                    • #25
                      I donít understand why this APU is always benchmarked overclocked. It makes the results higher than what they should be (I suppose).
                      Also I didnít understand the mention of Catalyst in between two graphs that have nothing to do with Catalyst.

                      Comment


                      • #26
                        Originally posted by stqn View Post
                        I donít understand why this APU is always benchmarked overclocked. It makes the results higher than what they should be (I suppose).
                        Where did you get that from?

                        Comment


                        • #27
                          Originally posted by Vim_User View Post
                          Where did you get that from?
                          From the table in the first page of the article. I canít copy-paste but it says 4.7GHz, while the normal frequency is 4.1GHz (with turbo boost up to 4.4GHz). The previous articles about this APU also used overclocked numbers.

                          Comment


                          • #28
                            We got a problem! I had great success with this kernel on my A8-3870K TV PC but when I install it on my laptop I get no screen!

                            The laptop is a dv6z-7000 from HP. It has an AMD A8-4600 APU with 7660G GPU as well as a 7730M discrete GPU. Internally, the display is connected to the APU and uses DMA to move frames to the iGPU when using the dGPU, but it looks like the new driver ignores the APU's GPU entirely and just inits the discrete one, which of course has no screen. The laptop is cool and quiet now, I'll give them that, but picking the wrong GPU is kind of a big deal.

                            Comment


                            • #29
                              I have been waiting for this day since the day I got my 5770. Even though catalyst is finally somewhat tolerable, I still can't wait to give this a whirl.

                              Comment

                              Working...
                              X