Dcd file and VMD

[Daniele Passerone]

Dear Cp2k developers,

I use FIST on Cp2k version 2.0.0, with a system in a monoclinic cell
that I specify
by giving the three vectors A,B and C.
Indeed, in the output I get

INITIAL CELL ANGLS[deg] = 0.9000000E+02 0.9000000E+02
0.1199999E+03

The corresponding .dcd file contains in the array cell(6) only the
diagonal part of the cell (I read it with a Fortran program), so that
when it is read by vmd the periodic boundary conditions are wrong.
By the way, the
> molinfo 0 set alpha 120
command does not seem to have effect on my version of vmd (not
mentioning the fact that I should give this command for all frames).
It would be great to have the dcd file also with the cell lengths and
angle information...

Thank you in advance
Daniele Passerone

[Laino Teodoro]

Daniele,

we do write the full cell information. I suspect that the format we
employ is not correct.
If you will provide the correct format for the DCD with the cell
information I will be extremely
happy to fix it.

At the moment we write the full lower diagonal part of the cell
matrix (hmat).
cheers
Teo

[Laino Teodoro]

I found a message from Axel with the necessary info (cell info format
in a dcd file) on another mailing list.
I updated consequently the cell info in the dcd. Please try it and
let me know in case of further problems.

Cheers
Teo

On 10 Nov 2008, at 23:53, Daniele Passerone wrote:

[Daniele Passerone]

Thank you very much Teo: now a dcd from a cell with lower diagonal
form will work perfectly in vmd, and pbc are already applied to all
frames without the need of pbctools!

[ilya]

I've just updated my cp2k and found that when I'm trying to read dcd
in vmd now it prints an error

dcdplugin) read_dcdstep: corruption or unrecognized file structure

I decided that maybe it's due to changes discussed in this thread.
Could somebody explain how should i open new dcd.

Thanks,
Ilya.

[Teodoro Laino]

Hi Ilya,

those changes should not affect the DCD format at all.
Can you please explain a bit better the sequence of operations you did
to get
that error message? (restarting? totally new dcd file?)..

Thanks
Teo

[Teodoro Laino]

Ilya,
no need to say anything else. I identified the problem. It is due to
some conflicts in different CVS commits.
I'm working now on a patch. Will send a message when the DCD will be
again working.
Thanks for reporting,
Teo

[Teodoro Laino]

Hi Ilya,

it is now fixed in the CVS.
Thanks for reporting the problem.
Teo