VMD-L Mailing List

From: John Stone (johns_at_ks.uiuc.edu)
Date: Fri Oct 22 1999 - 13:12:52 CDT

> hi,
>
> in reading parm & crd (trajectories) files (from amber 5.0),
> which work perfectly on unix/Xwindows machines,
> our windows 98 pc versions invariably crash.
> In fact the problem arises in reading the last set of coordinates
> (indipendently on the number of data sets, 3, 160, 200...)
> could anyone of you give me a reason for that?

This is likely to be caused by one of two things:
  1) memory allocation/deallocation bug in VMD that does something that is
     illegal in Windows, but is legal in Unix. (we have found a number of
     these in other parts of the code)

  2) endianism (aka byte order) of the files being read in.
     This has been an issue for some of the other file formats
     that VMD supports as well. Since a lot of people are trying
     out the Windows version, these problems are finally being uncovered.

If you can give me Parm/CRD files that cause this problem, then I'll
fix the bug in VMD. Someone else reported this same problem previously,
but never gave us a copy of the file(s) that crashed the program, and since
we don't run Amber much here, we don't have good test materials here
presently. (One of my assistants sent email to the Amber group asking
for current Amber sample files but never got a reply...)

If you could post the Parm/CRD file that crashes the Windows version of
VMD on a web page, or FTP site, I will grab it and get this problem fixed.

Thanks for the bug report!!

For others on VMD-L reading this, please _do_ send us bug reports
on the Windows version which are not already listed in the "release notes"
web page for the Windows version.

Thanks,
  John Stone
  vmd_at_ks.uiuc.edu 

-- 
Theoretical Biophysics Group   Email: johns_at_ks.uiuc.edu
Beckman Institute              http://www.ks.uiuc.edu/~johns/
University of Illinois         Phone:  (217) 244-3349
405 N. Mathews  Ave              FAX:  (217) 244-6078 
Urbana, IL 61801, USA          Unix Is Good For You!!!