- removed comment
init_3_timelevels seems to fail on Maxwell
If I use the carpet::init_3_timelevels=yes I get the following error WARNING level 0 in thorn CarpetLib processor 0 host quasar.ccrg.rit.edu (line 385 of /home/yosef/ET/Maxwell/Cactus/configs/lazev/build/CarpetLib/gdata.cc): -> Internal error: extrapolation in time. variable=ADMBASE::alp time=1280 times=[2560,4955,2395] WARNING level 0 in thorn CarpetLib processor 0 host quasar.ccrg.rit.edu (line 385 of /home/yosef/ET/Maxwell/Cactus/configs/lazev/build/CarpetLib/gdata.cc): -> Internal error: extrapolation in time. variable=ADMBASE::alp time=1280 times=[2560,4955,2395]
This error message did not occur when I ran with the previous
Keyword: init_3_timelevels
Comments (5)
-
reporter -
- removed comment
Do you want to suggest this as official solution? To my knowledge, only you (RIT) is using these.
-
reporter - removed comment
We no longer use OldStyle BCs. So removing them will affect the RIT codes.
-
- changed status to resolved
- removed comment
I removed the support for OldStyle BCs.
-
- changed status to closed
- edited description
- Log in to comment
The problem is due to an interaction of MoL at Initial and Carpet. The OldStyle boundary routines modify cctk_delta_time, which Carpet interprets as a change in CFL factor. Removing the OldStyle boundary routines seems to fix the problem.