On Thu, 2006-11-30 at 21:56 +0000, Peter Clifton wrote: > On Thu, 2006-11-30 at 15:09 -0500, DJ Delorie wrote: > > > I've not noticed any obvious improvement in performance, nor any > > > obvious redraw artifacts (I presume PCB double buffers internally, > > > however I've not looked! The PC I'm using is quite fast though, and > > > I never found GTK PCB slow. Is it useful for someone with a slower > > > machine available to see if this helps? > > > > Compare with the lesstif build, which uses explicit double buffering > > only. > > > > You can also test some things by forwarding your X connection to a > > remote machine. > > I've tried that as you suggested, and can't notice any difference. I > think the double buffering is server-side anyway, so will likely show up > most on a slow X server. > > I'll try against the lesstif build now.
This surprised me somewhat... Testing against Lesstif, the patch to turn GTK double buffering on the drawing area had basically no effect I could notice, however the GTK version appears far more responsive over a remote X connection than the Lesstif version. Now, it may be that I've not got the hardware / server configuration others found slow, or that the later (2.10.x) GTK versions such as I have are much faster than those initially tested with PCB. Has anyone else compared the GTK HID to Lesstif HID for speed lately? It may have been that the GTK==SLOW complaints were pre-HID, and that the problem is now fixed? Peter C. _______________________________________________ geda-dev mailing list [email protected] http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev
