Thank you very much for your answer, I'm currently gathering more
information about this bug, even if it's quite difficult to nail down. What
I suspect is that the application issue too much commands in a really short
time filling the "command buffer" of the GPU. I can workaround the freezing
thing by:

disabling double_buffering
disabling vsync
scheduling a slow function
using the glFinish function (glFlush does NOT work)

I'm not sure if this is pyglet-specific, other opengl applications seem to
work (but I'm not sure that they're updating so fast as I am, I've tried
glxgears and Avogadro, a molecular editor).


On Tue, Nov 6, 2012 at 4:03 PM, Richard Jones <[email protected]>wrote:

> On 7 November 2012 09:49, Gabriele Lanaro <[email protected]>
> wrote:
> > If I run a test code like the following I experience a window freeze (the
> > close button not working anymore). The application displays the current
> fps
> > for some time and if I, for example, click on the window I experience a
> > freeze.
> >
> > - The problem is reproduced also on a larger scale with a complex
> > application (ie this is not related to the fps display thing).
> > - The problem seems to be solved (there is a performance drop) if at the
> end
> > of the on_draw method I put a pyglet.gl.glFinish() call or, I schedule
> some
> > call that sleeps for some time.
> > - It happens only on a system (Ubuntu) with a [GeForce GTX 550 Ti] (rev
> a1),
> > using the nvidia driver version 304.51
> >
> > Am I doing something wrong?
>
> I don't believe so. Either pyglet's doing something wrong or the
> driver is doing something wrong and it'd be nice if pyglet could be
> more robust in the face of that.
>
>
> > This is a known bug?
>
> I think we can classify this as a "previously-unknown" bug :-)
>
>
> > import pyglet
> >
> > w = pyglet.window.Window()
> > ds = pyglet.clock.ClockDisplay()
> > @w.event
> > def on_draw():
> >     w.clear()
> >     ds.draw()
> >
> > pyglet.app.run()
>
> That's a great, concise example of bad behaviour (it works fine on my
> Mac.) It'd be great if we could nail down exactly what the problem is.
> Even if it is indeed a missing glFinish - though whether that should
> be included in the standard pyglet app run loop is probably
> contentious.
>
>
>      Richard
>
> --
> You received this message because you are subscribed to the Google Groups
> "pyglet-users" group.
> To post to this group, send email to [email protected].
> To unsubscribe from this group, send email to
> [email protected].
> For more options, visit this group at
> http://groups.google.com/group/pyglet-users?hl=en.
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"pyglet-users" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/pyglet-users?hl=en.

Reply via email to