[gmx-users] -ndec 30
feenstra at chem.vu.nl
Fri Nov 14 09:07:03 CET 2003
Nguyen Hoang Phuong wrote:
> the problem is that I would like to change the coordinates of some atoms
> in the last step of the fisrt run before using this for the next run. That
> why I convert the trr file to gro file (g96 as David said is not possible
> for me because it doesn't contains velecities) and change coordinates in
> this gro file. How can I keep the precision of the velocities in this gro
> file as same as in the trr file?
The .gro file format specifies 3 decimal places (or maybe four, I'm not sure),
anyway way less than even single precision. As a modification, trjconv supports
writing higher precision, and all gromacs tools should support reading that.
I have never checked what the limit is here, but since -ndec 7 is about single
precision, -ndec 14 (or 15 if you want to be safe), should be enough!
You could check this by doing the following steps:
1) trjconv -ndec to convert .trr to .gro
2) trjconv again from .gro back to .trr
3) tpbconv to convert from .trr to .tpr
4) gmxcheck -s1 -s2 to compare 'old' and 'new' .tpr.
(you maybe need tpbconv to generate also your 'old' .tpr from .trr)
(maybe gmxcheck can compare old and new .tpr straight away, I'm not sure)
Increase -ndec till gmxcheck shows no differences (or possible, a few last
round-off errors in last decimal remain).
| | |
| _ _ ___,| K. Anton Feenstra |
| / \ / \'| | | Dept. of Pharmacochem. - Vrije Universiteit Amsterdam |
|( | )| | | De Boelelaan 1083 - 1081 HV Amsterdam - Netherlands |
| \_/ \_/ | | | Tel: +31 20 44 47608 - Fax: +31 20 44 47610 |
| | Feenstra at chem.vu.nl - www.chem.vu.nl/~feenstra/ |
| | "If You See Me Getting High, Knock Me Down" |
| | (Red Hot Chili Peppers) |
More information about the gromacs.org_gmx-users