Announcement

Collapse
No announcement yet.

AMD GPU-PRO 16.20.3 Beta Linux Driver Released

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

  • #41
    I had HUGE screen tearing with the open source AMDGPU driver (not sure on DRI version, default?) but when I installed AMDGPU-PRO drivers it went away. This is at 4K, which seems to cause issues and I'm starting to think the driver somehow detects the wrong Hz settings for the screen at 4k (might do it manually sometime).

    I'm using Ubuntu 16.04 MATE with the Compton composer, screen tearing is minimal but it can be noticeable at times so I will mess with that later on.

    Anyone having crashing should considering forcing their cards into high&performance states, maybe even force the fan into 255 (max) setting just to eliminate overheating as a problem because my first 390x card (this is the 2nd one) had major issues if I didn't do that (its in videocard heaven now).

    Comment


    • #42
      I would like to see these drivers support freesync, my 49" 4k freesync monitor would surely benefit.

      Comment


      • #43
        Personally, I think it is pretty funny that you people are trying to install the drivers on a card (390/390X) that is not listed as supported by the driver and then complain about it... and also it is a Beta driver, if you install it, you should expect problems with it.

        Also, my personal experience is that AMD has been pretty clear from the beginning (since 1 year ago?) that the whole amdgpu-project should support tonga (285/380/380X) and fiji (Furies) in the first place...

        Comment


        • #44
          Originally posted by vein View Post
          Personally, I think it is pretty funny that you people are trying to install the drivers on a card (390/390X) that is not listed as supported by the driver and then complain about it... and also it is a Beta driver, if you install it, you should expect problems with it.

          Also, my personal experience is that AMD has been pretty clear from the beginning (since 1 year ago?) that the whole amdgpu-project should support tonga (285/380/380X) and fiji (Furies) in the first place...

          Is it funny that we want the best performance for our hardware? The Radeon 300 series cards including the R9 390 were released on June 18, 2015, taken from wiki. As you can see those cards aren't even 1 year old. Perhaps AMD wants to reimburse people for misselling products? I don't think so, than maybe a working set of drivers would do. Also, pro drivers are nothing but 90% open source drivers plus 10% closed source and they have been at it for ages.

          Comment


          • #45
            it's not working dual screen setup on r9_290 second screen connection causes the system freeze

            Comment


            • #46
              Originally posted by alex79 View Post
              Is it funny that we want the best performance for our hardware? The Radeon 300 series cards including the R9 390 were released on June 18, 2015, taken from wiki. As you can see those cards aren't even 1 year old. Perhaps AMD wants to reimburse people for misselling products? I don't think so, than maybe a working set of drivers would do. Also, pro drivers are nothing but 90% open source drivers plus 10% closed source and they have been at it for ages.
              More like 90% closed source and 10% open source, if you count lines of code.

              Not sure where you get "at it for ages" from, the hybrid stack is a relatively recent project with just a Vulkan preview and VI beta release so far. We talked about the overall roadmap a year or so ago at XDC but at the time the main development focus was getting the initial all-open stack ready.

              Can you please recap the issues you are seeing ? AFAICS you are installing a beta driver that has not yet been QA'ed on your hardware, finding that it works pretty well except for tearing, but I wasn't able to figure out if you tried the setup changes that had been recommended by other posters (vein and atomsymbol among others).
              Test signature

              Comment


              • #47
                Originally posted by bridgman View Post
                Not sure where you get "at it for ages" from, the hybrid stack is a relatively recent project with just a Vulkan preview and VI beta release so far. We talked about the overall roadmap a year or so ago at XDC but at the time the main development focus was getting the initial all-open stack ready.
                It's the internet, times goes faster here
                Internet time is like dog years so 1 year in real life equals 7 on the net.
                It's always like that on the net people that have had there iphone 6S for years and so on.

                Comment


                • #48
                  Ahh, dog years. So yes, I agree with alex79 then. Thanks
                  Test signature

                  Comment


                  • #49
                    Originally posted by Qaridarium
                    What is the future time-line for the Vulkan implementation to be opensource and dropped into the amdgpu/radeon driver?
                    Don't think we have one yet but working on it.

                    Originally posted by Qaridarium
                    I do have a GCN1.0 card the 7850 what is the future timeline for the AMDGPU/Radeon driver ? and what is the timeline for the AMDGPU-Pro driver for the GCN1.0 cards?
                    Radeon already supports it of course, so no timeline there. For amdgpu, the timing is going to depend a bit on how much community developers are able to contribute. Right now we're trying to get a "basically working" all-open stack out there.

                    Originally posted by Qaridarium
                    Why is AMD using this toxic development way : closed source first and then maybe opensource? In my opinion the other way around it would be much better way of development opensource first and then maybe closed source...
                    I don't think we are, so it's kinda hard to answer "why we are doing it".
                    Test signature

                    Comment


                    • #50
                      Originally posted by Qaridarium
                      We all know and I think you also know that the opensource code is to be tweaked to be lean as possible means less lines of code with much higher efficiency per line of code. and the closed source code i know this and i think you know this to is that closed source code is FATT with pointless lines of code without any public review to make it lean and to tweak it in less lines of code.

                      so your numbers are Biased ! and we all know this! we all know the closed source lines of code BIAS. (some are paid money per lines of code)
                      Tinfoil hat alert

                      Probably fair to say that the open source code is more efficient in terms of functionality per LOC but that is more a function of the development environment (small number of relatively senior developers working without a lot of time pressure vs large number of developers working to specific business timeline requirements), but nobody gets paid per line of code and never has AFAIK.

                      Remember that we're talking about replacing all of Mesa with closed source code, not just the pipe driver and winsys bits. The numbers really do work out the way I said, and factoring in open vs closed development models wouldn't change the numbers that much.
                      Test signature

                      Comment

                      Working...
                      X