The argumentparser does not handle the auto logic of config values set to None. Please consider this for improvement.
Patch partially applied in r3761:
print_bool
to the display string (and not the argument value which is always expected to be a boolean)
OpenGL
enabled by default on platforms that support it - I'm not sure we are ready for this... does this work ok for you? have you seen any visual corruption or any bugs?
We could enable it if it passes all the tests, assigning to ahuillet to review the status as I can't find any outstanding opengl issues.
Replying to antoine:
Patch partially applied in r3761:
- the other part of the patch makes
OpenGL
enabled by default on platforms that support it - I'm not sure we are ready for this... does this work ok for you? have you seen any visual corruption or any bugs?
No bugs yet, but I haven't used it excessively.
As far as I can tell there still are issues with Windows 7. Do you want me to check with latest release?
Please check, see also #372
afarr/SmO: is it usable on windows, or should it still be disable by default?
(I am tempted to enabled it on posix)
Smo and I are both having issues (like with osx, #379, logs posted at comment 14) with sound dropping with the windows client (r3908). It doesn't seem to be opengl related, but it's hard to decide how well the opengl is working when sound keeps dropping. (Seems ok?)
...it's hard to decide how well the opengl is working when sound keeps dropping
How so? Why not just leave the sound off for testing GL?
FYI: for testing, the best thing I have found for stress testing is to run one or both of:
mplayer somefile.avi
while true; do gtkperf -a; sleep 0.1; done
Well, once sound issues are set aside, OpenGL is working very nicely with win client (r3971). .avi file plays well, gtkperf displays errors elegantly (don't seem to have it installed), was even able to watch silent youtube video at the same time as playing a .avi file, with only the smallest bit of video chop on either.
Looks good.
I'm still not 100% convinced that OpenGL
rendering is 100% stable (not necessarily because of our code - could well be (py)gtkgl
or the drivers or X11 or a combination thereof - but I am keeping it enabled by default for this release... we'll see what falls out.
this ticket has been moved to: https://github.com/Xpra-org/xpra/issues/367