[MITgcm-devel] Re: [MITgcm-cvs] MITgcm/pkg/seaice CVS Commit

Patrick Heimbach heimbach at MIT.EDU
Sun Dec 24 15:47:27 EST 2006


Hi Jean-Michel,
thanks, I had completely missed this (6*60=360),
and had generated AD output before adding checking.
Should be fixed now.
Now ADM and TLM are almost "equally wrong"
re. gradient checks (still due to seaice_growth).
-p.

Quoting Jean-Michel Campin <jmc at ocean.mit.edu>:

> Hi Patrick,
>
> It seems to me that we would need at least nEVPstepMax=60
> in order to get this lab-sea ad test to run.
> Right now, it stop in SEAICE_CHECK:
> (PID.TID 0000.0001) SEAICE_ALLOW_EVP: need to set nEVPstepMax to >= nEVPstep
> (PID.TID 0000.0001) nEVPstep = INT(SEAICE_deltaTdyn/SEAICE_deltaTevp) =   60
>
> I am going to change the SEAICE_CHECK to write error msg to STDERR
> (instead of STDOUT) so that we get those msg also in the "run.log"
> file (+ saved & visible on the web for automated test on faulks).
>
> Jean-Michel
>
> On Fri, Dec 22, 2006 at 07:18:41PM -0500, Patrick Heimbach wrote:
>> Update of /u/gcmpack/MITgcm/pkg/seaice
>> In directory forge:/tmp/cvs-serv22113
>>
>> Modified Files:
>> 	SEAICE.h
>> Log Message:
>> Set default for nEVPstepMax=20
>> (note that this is only used for ALLOW_AUTODIFF_TAMC;
>> in pure forward it's still just reset to fraction of time-stepping
>> parameters; and in adjoint it's also restricted to that calculated
>> fraction, but needs the hard loop bondary for proper push/pop).
>>
>> _______________________________________________
>> MITgcm-cvs mailing list
>> MITgcm-cvs at mitgcm.org
>> http://mitgcm.org/mailman/listinfo/mitgcm-cvs
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel
>



--------------------------------------------------------
Patrick Heimbach   Massachusetts Institute of Technology
FON: +1/617/253-5259                  EAPS, Room 54-1518
FAX: +1/617/253-4464             77 Massachusetts Avenue
mailto:heimbach at mit.edu               Cambridge MA 02139
http://www.mit.edu/~heimbach/                        USA




More information about the MITgcm-devel mailing list