Hi all.
I am experiencing the following problem with Gem.
When I use the dimen message to set window size, I lose hardware acceleration (I'm assuming thats why everything become super-slow.).
This happens even if I set the window size to the default (500x500). Acceleration only works in a fresh PD session when the dimen message has not been used yet.
This is a debian/unstable linux system w/ a GeForce2 card.
Any ideas?
thanks andy
for me, dimen doesn't do anything.
i really expected it to crash, because normally if i breathe, blink, or lapse from a state of trancendental meditation, gem crashes pd.
but this has a lot to do with this message, i guess:
error: GEM: Unable to create single buffer window error: GEM: Error creating const context error: GEM: A serious error occured creating const Context error: GEM: Do not continue!
and it's right, single buffer mode doesnt work. but normal double buffer mode works within the rigid operating parameters outlined above (actually, it normally just crashes when i try to destroy the window, or close the patch).
debian unstable, x4.1, ati mobility m3.
On 10 Dec 2001 21:59:21 -0800 Andrew Water Schmeder andy@eh2o.ws wrote:
Hi all.
I am experiencing the following problem with Gem.
When I use the dimen message to set window size, I lose hardware acceleration (I'm assuming thats why everything become super-slow.).
This happens even if I set the window size to the default (500x500). Acceleration only works in a fresh PD session when the dimen message has not been used yet.
This is a debian/unstable linux system w/ a GeForce2 card.
Any ideas?
thanks andy
hi all !
fine that there are bug-reports, but i am far from knowing why there are these (!) bugs. Gem is being used for (say) 5 years, and i used to think, that (at least) the window-routines (like "dimen") should be somehow stable. well, obviously they are not
pix wrote:
for me, dimen doesn't do anything.
you need to send a dimen-message (like "dimen 1024 768") to the gemwin, before (!) you create the gem-window (with "create"). the dimen-message will not resize the window.
i really expected it to crash, because normally if i breathe, blink, or lapse from a state of trancendental meditation, gem crashes pd.
so i understand, why you don't try to destroy the gemwin, re-dimension it, and create it...
but this has a lot to do with this message, i guess:
error: GEM: Unable to create single buffer window error: GEM: Error creating const context error: GEM: A serious error occured creating const Context error: GEM: Do not continue!
and it's right, single buffer mode doesnt work. but normal double buffer mode works within the rigid operating parameters outlined above (actually, it normally just crashes when i try to destroy the window, or close the patch).
debian unstable, x4.1, ati mobility m3.
it seems to be a driver-problem (oh, gosh; i DO love these answers: it is a driver-problem -- no wait, it might have to do with your computer...) unfortunately i do not have access to your grafix-card (not even having a laptop). by the way, "debian unstable" really does mean "unstable" (i didn't believe this either till yesterday, when i tried to install the new nvidia-drivers which killed my X-server in combination with Gem)
On 10 Dec 2001 21:59:21 -0800 Andrew Water Schmeder andy@eh2o.ws wrote:
Hi all.
I am experiencing the following problem with Gem.
When I use the dimen message to set window size, I lose hardware acceleration (I'm assuming thats why everything become super-slow.).
This happens even if I set the window size to the default (500x500). Acceleration only works in a fresh PD session when the dimen message has not been used yet.
This is a debian/unstable linux system w/ a GeForce2 card.
hmmh, i think i believe it might have something to do, with the 1.0.2313/14-drivers for the geForce.
Any ideas?
thanks andy
i really expected it to crash, because normally if i breathe, blink, or lapse from a state of trancendental meditation, gem crashes pd.
so i understand, why you don't try to destroy the gemwin, re-dimension it, and create it...
but this has a lot to do with this message, i guess:
error: GEM: Unable to create single buffer window error: GEM: Error creating const context error: GEM: A serious error occured creating const Context error: GEM: Do not continue!
and it's right, single buffer mode doesnt work. but normal double buffer mode works within the rigid operating parameters outlined above (actually, it normally just crashes when i try to destroy the window, or close the patch).
debian unstable, x4.1, ati mobility m3.
as i understand gem is producing this errors always when there is no hardware accelleration ? though it lets u work with gem without hwa, its chrashes pd when u try to close the gemwin. or is there some workaraound for this, has somebody gem working without hwa and problems ?
i guess the ati mobility m3 is also a mach64 chip ? there is no x4.1 dri support for this one at the moment, but there exists a branch in the dri cvs which is currently worked on. i lurked a little bit the last days at their developer list as i have the same chip in my notebook, and it seems that there is soon something useful available.
cheers
erich
-
Andrew Water Schmeder -
IOhannes m zmoelnig -
pix -
rat＠telecoma.net