[Bug 98993] dosbox artefacts when using opengl

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Mon Dec 5 01:22:02 UTC 2016


https://bugs.freedesktop.org/show_bug.cgi?id=98993

            Bug ID: 98993
           Summary: dosbox artefacts when using opengl
           Product: Mesa
           Version: git
          Hardware: Other
                OS: All
            Status: NEW
          Severity: normal
          Priority: medium
         Component: Drivers/Gallium/radeonsi
          Assignee: dri-devel at lists.freedesktop.org
          Reporter: aaronbpaden at gmail.com
        QA Contact: dri-devel at lists.freedesktop.org

Created attachment 128338
  --> https://bugs.freedesktop.org/attachment.cgi?id=128338&action=edit
example image

I've got an OLAND era radeon card, and I'm using mesa-git. I can confirm the
issue exists using both the radeonsi and amdgpu kernel drivers. I'm using the
latest SVN revision for dosbox.

Using these config options:
[sdl]
fullresolution=desktop
output=opengl

[render]
aspect=true

which should maintain the desktop's display mode, use opengl for the output and
scale it to fit the screen while maintaining the correct aspect ratio.

Using these options, there are flashing artefacts outside of where the
application is rendering on my hardware. The artefacts are sometimes complete
garbage and other times they are noticeably the garbled remnants of a
previously opened application (for example, a youtube video or a previously
opened dosbox game).

The issue is far more pronounced on Xorg. Using wayland, I do occasionally get
a 1-pixel-wide line at the edge of dosbox's framebuffer, but it isn't flashing.

It's also unlikely to be an issue with Mutter, because I've tested it out in
openbox and got the same problem. IRC said it was most likely an issue on
mesa's end.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/dri-devel/attachments/20161205/8f57effb/attachment.html>


More information about the dri-devel mailing list