15 Replies Latest reply on Nov 25, 2015 2:32 PM by m0nt3 Branched to a new discussion.

    AMD Catalyst - Mouse cursor corruption

    leonmaxx

      Hello.

       

      I'm here to report again about mouse cursor corruption problem, it's still not fixed.

      With dual-monitor setup (monitors A and B), mouse cursor gets corrupted on monitor A after some time (typically 10 to 30 mins of uptime), on monitor B cursor is normal, not corrupted.

       

      My primary computer specs:

      OS: Ubuntu 14.04 64-bit

      CPU: AMD FX-8350

      M/B: ASUS M5A97 Evo

      GPU: AMD Radeon R9 290 with Catalyst 14.6 Beta (fglrx-14.20)

      Monitors: 2 x DELL UltraSharp U2412M (1920x1200)

       

      Other computer specs:

      OS: Ubuntu 14.04 64-bit

      CPU: AMD Phenom II X6 1045T (oveclocked to 4.1GHz)

      M/B: ASUS Sabertooth FX990

      GPU: AMD Radeon HD 6950 with Catalyst 14.6 Beta (fglrx-14.20)

      Monitors: 2 x DELL UltraSharp U2412M (1920x1200)

       

      The bug is easily reproducible, do some work on monitor B with mouse for about 10-15 mins and then move mouse to monitor A, most of the times it's gets corruption.

       

      Additional testing (All setups running Ubuntu 14.04 64-bit with Catalyst 14.6 Beta):

      AMD Radeon R9 290 with 2 x DELL U2412M - cursor corrupts

      AMD Radeon HD 6950 with 2 x DELL U2412M - cursor corrupts

      AMD Radeon HD 5850 with 2 x DELL U2312HM (1920x1080) - no corruption (worked on this this setup for 1 day and being unable to reproduce bug)

       

      Ask me for additional info and/or testing if You need.

        • Re: AMD Catalyst - Mouse cursor corruption
          leonmaxx

          Catalyst 14.12 Omega - cursor corruption still not fixed!

          First my report of this problem was in this thread: Catalyst 14.3 Beta - Secondary display not working when TearFree enabled. almost year ago, and You still don't fixed this annoying bug!

           

          This was my last purchase of your product, sorry. There is a limit of patience.

           

          BR.

          1 of 1 people found this helpful
          • Re: AMD Catalyst - Mouse cursor corruption
            koenich

            same for me! i have a Radeon 6970HD on Win7_64 with single monitor setup via HDMI.

            what makes me mad is that the problem is known for years. and AMD simply don't bother. i never read any statement by them regarding the problem.

             

            like mentioned: i have AMD Phenom II and Radeon 6970. i will never ever buy a product that has AMD printed anywhere. this is simply a rip off.

            if you can't get things done then don't start them.

              • Re: AMD Catalyst - Mouse cursor corruption

                I will make sure the Catalyst team hears this. Thanks for the report.

                 

                For what it's worth,  from the original post "... do some work on monitor B with mouse for about 10-15 mins and then move mouse to monitor A, most of the times it's gets corruption." is NOT a clear reproducible. Without clear, step by step instructions that make this happen reliably, it isn't likely the QA team can replicate, isolate, and fix this problem. I can EASILY imagine all kinds of scenarios where the driver has nothing to do with why this happens.

                 

                If you have a clear replicatable case, you can report the bug following the instructions at the FAQ thread: Where do I report a Catalyst/Driver issue?

                  • Re: AMD Catalyst - Mouse cursor corruption
                    leonmaxx

                    I was getting this bug 10 times a day. So it's easily reproducible.

                    Video cards affected: Radeon R9 290, Radeon HD 6930/6950.

                    Unaffected: Radeon R9 280, Radeon R7 250, Radeon R9 270, Radeon HD 5850.

                     

                    And I use all of these video cards for development, imagine how it annoys me when I have to reboot PC 10 times a day in middle of process.

                     

                    BR, Leon.

                    • Re: AMD Catalyst - Mouse cursor corruption
                      koenich

                      for me the problem so far only occured while using fullscreen 3D apps like for example Company of Heroes or Dota 2 where a 3D cursor overlay is used. please note: as i said i don't think this is only happening in dual monitor environments since i never used a second one. maybe turning off the second output port of the GFX card could be a help. i don't know though whether the second output can be truely deactivated.

                      for me the problem can't be reproduced since it happens irregularly; sometimes after minutes sometimes after hours of 3D use.

                      but i know that sometimes the broken cursor can be fixed while still in 3D mode by quickly shaking the mouse (cursor) and by moving it to the edges of the screen. it's unbelievable. i know. but it worked several times even if not every time.

                       

                      what i think (i am not dev or any of that sort): in Windows (7 and 8 i believe) in the bottom right corner there is a field called "Show Desktop". when you roll over it, it hides all windows. i think there is the problem. don't misunderstand me. i am no pro, but as far as i know 3D is only an overlay which means that in some cases even in 3D mode the mouse could correspond with the "Show Desktop" field. i think like that because sometimes popups like from JDownloader are visible for few mseconds even in 3D mode when they are not supposed to be visible. even if not in specific i think the corrupted cursor has to do with some roll-over mechanics.

                       

                      i hope this is a help to solve the problem.

                  • Re: AMD Catalyst - Mouse cursor corruption

                    Just a heads up. Don't hold your breath, but.... we have reliably replicated this problem, with help from one of my colleague engineers in Markham, who is in the same building as the driver QA team. Nothing like solid human presence sometimes. He had the same problem, hauled in the QA guys, they played around with it, and VOILA! We have a replicatable case. No predictions on if/when a fix, but a positive step.

                     

                    Thanks for the input and the feedback.

                    1 of 1 people found this helpful