On Mon, Mar 16, 2009, Mark Williamson wrote:
>
> 1)
>
> possible FAILURE: check mdout.tip5p.dif
> /server-home/mjw/code/AMBER/amber11/test/tip5p
> 180,181d179
> < EKCMT = 0.2952 VIRIAL = 9.6699 VOLUME =
> 0.0001
> < Density =
> 0.
> ---------------------------------------
Should be cleaned up, but it's really not an error -- these are RMS
values for a very short run, and print trigger is slightly different in
the new vs. saved results. [i.e., this really corresponds to a rounding
error.] I don't see this problem with gfortran 4.3.2.
> 2)
> possible FAILURE: check tgtmd.out.dif
> /server-home/mjw/code/AMBER/amber11/test/tgtmd/change_target.ntr
targetted MD has always had a big portability problem. Scott may(?)
probably know what sorts of machines the save files came from. Probably
deserves a bugzilla entry.
> 3)
>
> possible FAILURE: check mdout.dif
> /server-home/mjw/code/AMBER/amber11/test/ncsu/smd
> ---
> > ** NCSU-Error ** : expected list value for key 'i', got '<EMPTY>'
> instead
Has been reported on the mailing list several times...I don't recall
that it was ever tracked down. Doesn't show up with gfortran 4.3.2.
>
>
> 4)
>
> possible FAILURE: check smd.txt.dif
> /server-home/mjw/code/AMBER/amber11/test/ncsu/smd
You need to examine the output files for any clues -- the diffs suggest
that the job just stopped(?), or crashed(?). If you haven't done so,
you will need to run the tests by hand to look for more clues.
>
> All compiles and tests were done on my desktop machine (RHEL 4, gfortran
> 4.1.2).
>
....dac
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Wed Mar 18 2009 - 01:10:48 PDT
