[Users] Refinement levels 3 to 5.

[Roland Haas]

Hello Bilal,

> When I change, Carpet::max_refinement_levels = 5 (instead of 3). and
> I also changed,
> 
> CarpetRegrid2::num_centres  = 1
> #CarpetRegrid2::num_levels_1 = 3
> #CarpetRegrid2::radius_1[1]  = 1
> #CarpetRegrid2::radius_1[2]  = 1
> 
> CarpetRegrid2::num_levels_1 = 5
> CarpetRegrid2::radius_1[1]  = 5
> CarpetRegrid2::radius_1[2]  = 5
> CarpetRegrid2::radius_1[3]  = 5
> CarpetRegrid2::radius_1[4]  = 5
You are correct in changing both Carpet::max_refinement_levels which
controls the maximum possible number of refinement levels as well as
CarpetRegrid2::num_levels_1 which controls that actual number of
refinement levels used.

> My simulation runs successfully. but fourth and fifth levels shown
> only nan as follows,
> 
> INFO (CT_MultiLevel): === Level 2 ===
> INFO (CT_MultiLevel):  * Equation #0: 10 iterations, norm of final residual = 2.5741620337e-05
> INFO (CT_MultiLevel):  * Equation #1: 10 iterations, norm of final residual = 8.7340004508e-06
> INFO (CT_MultiLevel):  * Equation #2: 10 iterations, norm of final residual = 8.7156879010e-06
> INFO (CT_MultiLevel):  * Equation #3: 10 iterations, norm of final residual = 8.6982806224e-06
> INFO (CT_MultiLevel): === Level 3 ===
> INFO (CT_MultiLevel):  * Equation #0: 1 iterations, norm of final residual = nan
> INFO (CT_MultiLevel):  * Equation #1: 1 iterations, norm of final residual = nan
> INFO (CT_MultiLevel):  * Equation #2: 1 iterations, norm of final residual = nan
> INFO (CT_MultiLevel):  * Equation #3: 1 iterations, norm of final residual = nan

> What else should I fix?

I am not really familiar (at all, never having used it) with
CT_MultiLevel itself. However looking at the parfile
repos/ctthorns/CT_MultiLevel/par/constraints_spherical.par you can see
that it sets:

CT_MultiLevel::topMGlevel          = 2

and that parameter is described in the thorn's documentation (in the
doc directory or online)
http://einsteintoolkit.org/thornguide/CTThorns/CT_MultiLevel/documentation.html
as:

--8<--
The only other parameter which must be set to ensure correct operation
is CT_MultiLevel::topMGlevel, which tells the solver which is the finest
refinement level that covers the entire domain in which the equation is
to be solved. All levels below and including this will be used in the
classical multigrid sense (e.g., in a V- or FMG-cycle). The ones above
will be treated as local AMR boxes, and be solved via a multilevel
prescription; presently, this involves simply interpolating the
solution from the topMGlevel refinement level at the end of the
multigrid part and relaxing it progressively, from coarser to finer, on
all the local grids. 
--8<--

so since you changed max_refinement_levels from 3 to 5 I would also
change topMGlevel from 2 to 4.

Yours,
Roland

-- 
My email is as private as my paper mail. I therefore support encrypting
and signing email messages. Get my PGP key from http://pgp.mit.edu .
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
Url : http://lists.einsteintoolkit.org/pipermail/users/attachments/20200826/68864005/attachment.bin