[Mitgcm-support] Re: How are we doing?

[mitgcm-support at dev.mitgcm.org]

Alistair Adcroft wrote:
> 
> Patrick,
> 
> Not meaning to upset you but this should still work for you and
> it now works for me too.
> 
> I've changed the macro to DISABLE_MULTIDIM_ADVECTION
> and I've put in GAD_OPTIONS.h: #ifdef AUTO... #define DISABLE_... #endif
> and updated comments to make it obvious what's going on.

Well, no.
I've modified gad_advection so as to be able to differentiate
it INCLUDING the multi-dim. advection scheme.
However, since this requires a new and additional type of common block
for the innermost checkpointing, I'd like to be able to switch
it off, if I don't want it so as to reduce memory.
So, linking DISABLE_MULTIDIM_ADVECTION to ALLOW_AUTODIFF_TAMC
as you suggest does not make sense.

P.

> This way we don't have to use it in CPP_OPTIONS.h: all the new
> advection schemes are meaningless without out multi-dimensional
> advection (it's part of the method) and most future
> experiments will use one of these schemes. The documentation on
> tracers shows why you need multi-dim with these schemes.
> 
> A.
> 
> Patrick Heimbach wrote:
> > I'm performing some final AD-related tests.
> > Am about to check-in fixed code.
> > As mentioned earlier, I've added a new CPP option
> > ALLOW_MULTIDIM_ADVECTION
> > You may have to define that in one of your verif. experiments

-- 
_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/

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

>>>>>>>      http://web.mit.edu/cms/reconstructions      <<<<<<<