Hello !
[note: I have been trolled by Giebl. Feel free to skip this one]
Sorry if it looked like this, but i do not want
to troll anyone here. I respect everybody here.
CU
Mikael Eriksson:> On Tue, May 09, 2006 at 09:58:21AM -0700, Nick Moffitt wrote:
I assure you, I have methodically gone through my game and fiddled
with GetSurface and SetVideoMode flags alone and in combination.
Each time I run the game both with firefox running on another
desktop, and again without. The session with a firefox going runs
smoothly, while the framerate without firefox is much lower. I’m
baffled!Does other SDL based things run faster to?
I have just demonstrated the effect with several pygame applications.
These were using rect-based updates instead of attempting to
double-buffer, so things are getting curiouser and curiouser.
Also, my pygame demos use the pygame clock to delay down to 30fps, but
are having trouble keeping up with that framerate only when firefox
is not running! So I don’t think it’s the sleep problem.
In fact, with firefox going I see smooth playback of my pygame examples
EVEN WHILE encoding a DVD track into ogg theora (which is currently dog
slow, and maxes out the CPU to get barely 3fps for encoding). So the
scheduler shouldn’t be worried about pygame grabbing all the cycles.
For the record, this is a Thinkpad T40 running Ubuntu dapper drake.
I’ve turned on a few DRI-related switches in my xorg.conf, and would be
happy to provide it on request.
–
Information gladly given, but safety requires Nick Moffitt
avoiding unnecessary conversation. @Nick_Moffitt
Also, my pygame demos use the pygame clock to delay down to 30fps, but
are having trouble keeping up with that framerate only when firefox
is not running! So I don’t think it’s the sleep problem.
At this point, it really sounds like an X server bug to me.
–ryan.