VMD-L Mailing List
From: John Stone (johns_at_ks.uiuc.edu)
Date: Wed May 05 2004 - 16:46:01 CDT
- Next message: John Stone: "Re: variable atom number and dynamic bonds"
- Previous message: Tim Isgro: "Re: Include HETATM into *.psf file"
- Next in thread: Dallas Warren: "Rotating and translating molecules .."
- Reply: Dallas Warren: "Rotating and translating molecules .."
- Reply: Dallas Warren: "Catching error info before it shuts down ..."
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Hi,
I thought I'd write a followup to the previous VMD-L postings about weird
problems with the recent crop of ATI drivers on Linux (versions 3.7.x..).
Testing VMD with ATI's 3.7.6 drivers (released in March 2004) reveals
they have various bugs. Some of bugs are new (vs 3.2.x anyway) and some
of which match bugs I've seen before.
For reference/comparison, our ATI test system is an:
AMD Athlon 1333, with an ATI FireGL X1-128 video board
Running RedHat 9, with Linux kernel 2.4.20-8
- Maximizing the graphics window causes X to get very slow/hang and
yields a blank black window, and/or crashes VMD. This is certainly a
bug in the ATI drivers since the problem doesn't occur on any other cards.
- The "snapshot" rendering method produces corrupt output regardless
whether antialiasing or other features are enabled within VMD.
I believe this is also a graphics driver bug, the problem didn't occur
with older revisions of the ATI drivers, i.e. 3.2.x as I recall.
- Stereo support still seems very buggy, as I get visible display
corruption regardless of whether VMD is even running, random parts
of the display contain "garbage" as I run/move standard X-windows
apps like 'xterm'. When VMD starts, it does not detect stereo from
the OpenGL driver anyway, so it is unusable presently. This would also
seem to be a bug in the driver.
I tried various combinations of settings with the ATI 'fglrxconfig'
utility but I was unable to come up with any combination that cured
the problems listed above thus far. One good thing about the new ATI
drivers is that some old bugs I had seen where geometry was actually
drawn incorrectly (or not drawn at all for that matter) have been fixed.
If anyone has solutions for the 3 problems I list above, let me know.
I plan to report the bugs to ATI (again) if I don't find a solution shortly.
I recommend that anyone else that finds bugs in the ATI Linux drivers
report them on their Linux drivers feedback form:
http://apps.ati.com/linuxDfeedback/index.asp
John Stone
vmd_at_ks.uiuc.edu
-- NIH Resource for Macromolecular Modeling and Bioinformatics Beckman Institute for Advanced Science and Technology University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801 Email: johns_at_ks.uiuc.edu Phone: 217-244-3349 WWW: http://www.ks.uiuc.edu/~johns/ Fax: 217-244-6078
- Next message: John Stone: "Re: variable atom number and dynamic bonds"
- Previous message: Tim Isgro: "Re: Include HETATM into *.psf file"
- Next in thread: Dallas Warren: "Rotating and translating molecules .."
- Reply: Dallas Warren: "Rotating and translating molecules .."
- Reply: Dallas Warren: "Catching error info before it shuts down ..."
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]