>>>
>>> Not against, but as a regular pymol user, I won't be able to load
>>> trajectories without converting back to text format...
>>>
>>> Gerald, who really needs to learn VMD someday...
>>
>> no,no, The proper warning is Gerald, who will learn that netcdf saves
>> tons of time and disk space !
>
> I am not that convinced by the disk space: when I gzip my text file, I
> usually get smaller files that the netcdf ones. However, I do agree that
> they are more convenient to handle and faster to process with cpptraj.
>
Not just cpptraj. On a cluster here at SDSC - that uses a lustre based parallel file system:
250K atom system - ntwx=100 (needs to be short for aMD reweighting).
ioutfm = 0 - 2.7 ns/day
ioutfm = 1 - 31.6 ns/day
I wonder how many people are leaving an order of magnitude in performance on the table because they are still writing ascii trajectories?
I plan to implement Dan's suggestion - if people still want ASCII mdcrd they can set ioutfm=1 - I will leave ntxo at the current default.
All the best
Ross
/\
\/
|\oss Walker
---------------------------------------------------------
| Associate Research Professor |
| San Diego Supercomputer Center |
| Adjunct Associate Professor |
| Dept. of Chemistry and Biochemistry |
| University of California San Diego |
| NVIDIA Fellow |
|
http://www.rosswalker.co.uk |
http://www.wmd-lab.org |
| Tel: +1 858 822 0854 | EMail:- ross.rosswalker.co.uk |
---------------------------------------------------------
Note: Electronic Mail is not secure, has no guarantee of delivery, may not be read every day, and should not be used for urgent or sensitive issues.
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Fri Feb 26 2016 - 15:00:08 PST
