[Carpet] Regridding and scheduling in new version of Carpet

[Erik Schnetter]

On Mar 3, 2008, at 16:12:40, Ian Hinder wrote:

> Erik Schnetter wrote:
>
>> The easiest solution would be to set the grid hierarchy in different
>> ways in PREREGRIDINITIAL and in PREREGRID.  I find that setting up  
>> the
>> initial grid hierarchy and modifying the grid hierarchy later on are
>> sufficiently different that using the same method doesn't quite work.
>
> After reading more carefully how CarpetRegrid2 works, I came to the  
> same
> conclusion.  For the record, I now run my regridding thorn in
> PREREGRIDINITIAL and in PREREGRID, but in the first bin I take the
> locations of the centres from parameters, and in the second bin I take
> it from spherical surfaces.  This now works.
>
> For people wanting to continue using CarpetRegrid rather than
> CarpetRegrid2, I want to point out that the behaviour between
> carpet-stable-3 and current Carpet development has not changed.  You
> still have to manually add the buffer zones to the coordinates string
> (or whatever method you use).  So the code that I have to do rotating
> symmetries works fine as it did before; all I have to do now is know  
> how
> many buffer zones to add, as Carpet doesn't have the parameter any  
> more.

I notice that our three groups have three different regridding thorns  
which essentially all do the same thing.

CarpetRegrid2::min_fraction specifies the minimum fraction of refined  
points that need to be present in a single, cuboid box that replaces  
two individual overlapping boxes.  If the fraction is lower, there is  
no single cuboid boxes; instead a concave region is used.  By setting  
this parameter to 1 there is never a single, cuboid box; by setting it  
to 0, there are never concave regions.

Using a single,cuboid box can increase the memory requirements by a  
factor of two.

Note that this condition is applied before the hierarchy is made self- 
consistent and consistent with the symmetry conditions, so that a  
concave region can still result later on.

Note also that CarpetRegrid performs none of these actions and none of  
these checks, so that it is possible to end up with an inconsistent  
grid hierarchy, or a grid hierarchy where many more grid points are  
interpolated from the coarse grid than intended.

>>> The documentation in Carpet/doc/scheduling.tex says that the bins  
>>> are
>>> run in the order
>>>
>>>   [CCTK_BASEGRID]
>>>     [CCTK_INITIAL]
>>>   [CCTK_PREREGRIDINITIAL]
>>>    Regrid
>>>   [CCTK_POSTREGRIDINITIAL]
>>>
>>> and this seems to me to make much more sense, but I understand if  
>>> it is
>>> out of date.
>>
>> Yes, this is out of date.  Could you correct it?
>
> Sure.


Thanks.

-erik

-- 
Erik Schnetter <schnetter at cct.lsu.edu>   http://www.cct.lsu.edu/~eschnett/

My email is as private as my paper mail.  I therefore support encrypting
and signing email messages.  Get my PGP key from www.keyserver.net.



-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 194 bytes
Desc: This is a digitally signed message part
Url : /archives/developers/attachments/20080303/c9d41947/attachment.pgp