[MITgcm-devel] Re: [Fwd: (no subject)]

David Ferreira dfer at ocean.mit.edu
Thu Mar 2 12:16:48 EST 2006


Ed Hill wrote:

>On Thu, 2006-03-02 at 11:25 -0500, David Ferreira wrote:
>  
>
>>I tried the rdnctiles.m and it works.
>>However, I noticed it runs about 2 times slower than rdmnc.m.
>>    
>>
>
>Hi David,
>
>I'm moving this over to the devel list since I'm hoping others might
>want to try rdnctiles().  Or, even better, get inspired by its output
>and start extending some of the plotting and analysis routines to work
>with the generalized handling of tiles.
>
>So what is the exact command that you're using to call it?  I think I
>know whats causing most of the overhead.  By default, rdnctiles() reads
>all of the metadata and I think rdmnc() mostly ignores it.  Making this
>a non-default option may give a big speedup.
>
>Ed
>  
>
I use typically:
Ocn  = rdmnc([dirnam,'DiagOcnTave.*']);
or
Ocnb = rdnctiles([dirnam,'DiagOcnTave.*']);

david



-- 

Ferreira David               Tel : 617 253 7967 
EAPS Room 54-1515,
Massachusetts Institute of Technology,
77 Massachusetts Avenue
Cambridge, MA, 02139




More information about the MITgcm-devel mailing list