VMD-L Mailing List

From: Stefano Piana (piana_at_power.curtin.edu.au)
Date: Tue Jun 28 2005 - 20:37:15 CDT

John Stone wrote:

>Peter,
> Yes, they do limit your graphics capability. Does your VMD run ok
>without them? Or does it die? If you can run without them, that's
>what you want to do. If not, then we can find a more minimalistic
>set of options to cut back just the feature(s) that cause you trouble.
>
>I've found that on a colleague's ATI-based laptop, just disabling one
>OpenGL feature will make VMD run just fine on his ATI chip /w Linux.
>
>
Hi John,
I have tried with the VMDSIMPLEGRAPHICS 1 but still VMD crashes. I am
trying to track down the cause of the crash and, although it is likely
to be related to the ATI driver (if I run with a standard vesa driver
everything runs fine) I am a bit puzzled.
As I mentioned before, in my case VMD opens normally. If I load a
molecule with no covalent bonds I can load up to a million atoms and
display them in fancy VDW mode without any problem.
If I load a molecule with only a few covalent bonds (less than 100) then
VMD works fine.
If I load a molecule with more than ~100 covalent bonds then either VMD
crashes instantly or it starts sucking up all the memory and CPU until
the system hangs and requires a hard reboot.
I can see no difference between versions 1.8.2 and 1.8.3.
Suggestions?

Stefano