gEDA-dev: PCB rendering performance
der Mouse
mouse at Rodents.Montreal.QC.CA
Tue May 13 21:31:37 EDT 2008
I hope nobody has been putting many brain cycles into the PCB rendering
slowness I mentioned a little while back. Quite by coincidence, I
found very compelling evidence that the problem isn't in PCB, not even
if my HID counts as part of "PCB".
I was trying to track down another problem. Specifically, if I display
tut1.pcb with the "whole board" zoom setting of 510 and it's fine, set
zoom to 100 and it draws a small part, as expected, then the machine
crashes. (I hasten to add, I do not consider this PCB's fault. No
matter _how_ buggy it may be, it shouldn't be able to do that; it's
running with no access to /dev/mem or any such thing that could justify
a crash.)
While trying to figure out what's with the crash, I ran PCB under a
very simple server+mterm+twm session instead of my usual full-fledged X
session. The redraw that has usually taken some 45-60 seconds
completed in maybe one or two seconds.
The really bizarre thing is, when I try it now under my usual session,
it is a little slower, but still only maybe two to three seconds.
There's a delay of about five seconds before it starts drawing, and
that might be worth looking at, but until I figure out WTF is going on
with this graphics replot speed bizarrity I don't think there's any
point in even thinking about other speed issues. I'm trying to imagine
what possible relevant difference there could be between a freshly
booted machine and one that's been up for a while.
/~\ The ASCII der Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents.montreal.qc.ca
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
More information about the geda-dev
mailing list