From: Leandro Martínez (leandromartinez98_at_gmail.com)
Date: Sat Nov 25 2006 - 05:08:28 CST
Hi Brian, Jonathan and Hyonseok,
Thank you very much for the answers. I will update my dcd file reader
to account for this possibility.
Best regards to all.
On 11/25/06, Hyonseok Hwang <danggi_at_northwestern.edu> wrote:
> Hi, Leandro
> From NAMD 2.6, the default value of 'DCDUnitCell' option is 'Yes',
> which means that the information of the cell size (=simulation box) is
> automatically written in each frame if you use periodic boundary
> condition. Until 2.5, 'No' was default value. The cell size information
> includes the x, y, z length of the simulation box and three angles. If I
> remember correctly, each variable has double size and there are dummy
> integers in front and behind the cell size information. So total size is
> 4 + 6 * 8 + 4 = 56 Bytes. The 56 Bytes are taken in front of the
> coordinations in each frame. So you have to consider this if you want to
> read a DCD file from 2.6. If you want more detailed information on DCD
> file structure, please see the "dcdlib.C" among the NAMD source codes.
> I hope this will be of help.
> Leandro Martínez wrote:
> > Hi all,
> > I had fortran programs to read the dcd file format generated by version
> > 2.5 of namd.
> > Now it seems that in version 2.6 there has been a chenge in the DCD
> > format and
> > the programs do not seem to be able to read the coordinate files anymore.
> > Previously I read the dcd files using:
> > open(10,file=dcdfile,status='old',form='unformatted')
> > read(10) dummyc, nset, (dummyi,i=1,8), dummyr, (dummyi,i=1,9)
> > read(10) dummyi, dummyr
> > read(10) ntotat
> > do i = 1, nframes
> > read(10) (x(i,j),j=1,ntotat)
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:42:52 CST