VMD-L Mailing List

From: Luca Gelisio (luca.gelisio_at_gmail.com)
Date: Fri Sep 25 2009 - 09:10:23 CDT

John,

  Are you running out of memory on your machine?
> That's what it looks like to me from the errors you're getting.
>

I'm not running out of machine memory (6Gb RAM + 12Gb swap)

A second issue here is that you've got 12 million atoms, which goes
> beyond the limits of the normal PSF file format. We have built
> structures this large, but we had to use our own file formats rather
> than PSF/PDB because various field widths overflow with the standard
> PSF/PDB files. What do you plan to do with the system next?

I need only atomic coordinates (I usually export them in .xyz format, I've
just to know where is every atom and what's its scattering power, that's if
is O or H)... I'm simulating x-ray diffraction patterns, I'm working at
Trento University, Italy.

> If you're
> going to simulate with NAMD, you'll have to do what we've been doing
> and use the other file formats. I have special builds of VMD and psfgen
> that can emit files in the other formats for setting up huge simulations
> like this.
>
> If you tell us more about what you're attempting to do,
> I can either provide you with the needed VMD/psfgen builds that enable
> output to the .js file format or give you alternative suggestions.
>

Many thanks, VMD is really an incredible and powerful tool,

luca