[MITgcm-devel] changes in exch2

[Jean-Michel Campin]

Hi Dimitris,

Thanks for doing the update. I talked briefly to Chris and
he does not disagree with leaving the 510 files in contrib.

To generate the 2 exch2 files with missing tiles (generated with 
a "blanklist.txt" file), well, you just need the corresponding
blanklist.txt filei in the dir where you run the matlab "driver.m"

I had the impression that it would be more tricky, because I did
not have the "blanklist.txt" file for the cs32 example with
missing tile (s176t_8x4) but for the 2 cs510 configs with missing
tiles (1500t_17x51 & s205t_85x85) the blanklist.txt file is
checked in, so it should be easy.

Let me know if you have a problem with those config.

And if you agree, I can remove the cs32 config files
(s12t_16x32, s24t_16x16, s24t_32x8, s24t_8x32 & s6t_32x32)
from MITgcmContrib/high_res_cube/code-mods, since they are
no longer up to date.

Cheers,
Jean-Michel

On Fri, Aug 01, 2008 at 06:02:59PM -0700, Dimitris Menemenlis wrote:
> OK.  I have updated the CS510 exch2 files in the Contrib branch
>
> Two of the configurations use blankllist and I have not yet updated  
> them.
> Is there an example or instructions somewhere on how to update the exch2
> files when there is a blanklist?
>
> D.
>
> Dimitris Menemenlis
> DMenemenlis at gmail.com
>
> On Jul 31, 2008, at 6:03 PM, Jean-Michel Campin wrote:
>
>> Since those files are sitting in the Contrib branch, they can
>> stay there.
>>
>> Chris, what is your opinion ? and what should we do with the one
>> you put in global_ocean.cs32x15/code_alt ?
>>
>> But it would be good to update them anyway.
>> (+ it should be faster now, 1 less exch for UV, not yet tested with  
>> cs510)
>> Do you use "blanklist" ? If not, it's relatively straighforward
>> with the matlab script.
>
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel