[MITgcm-devel] Re: Exch2_ wide stencil

[Dimitris Menemenlis]

> I checked in the exch2_ wide stencil loop plus the extra bits to make
> the numbers the same as exch1_ by zeroing some numbers. Can you try a
> pickup of your low visc run that blows up after with these changes.
> 
> In the tradition of ensuring that changes fix thing - I preditc this
> fix will not change anything.

As you predicted, this fix does not change much.
I have not checked corners in detail yet cause altix
is out of commission, but a test that I ran last night
crashed at about the same time as before.

D.

-- 
Dimitris Menemenlis <menemenlis at jpl.nasa.gov>
Jet Propulsion Lab, California Insitute of Technology
MS 300-323, 4800 Oak Grove Dr, Pasadena CA 91109-8099
tel: 818-354-1656;  fax: 818-393-6720