[CP2K-user] CUTOFF and REL_CUTOFF
Pierre-André Cazade
pierre.a... at gmail.com
Thu Sep 12 13:24:43 UTC 2019
Dear CP2K users,
Following the tutorial to tune the CUTOFF and the REL_CUTOFF for two of my
systems, I find that quite large value of the CUTOFF (700 Ry for beta
glycine and 900Ry for quercetin) are necessary to achieve a decent level of
precision (about 1.e-6). I am a bit surprised considering my systems are
made from C, N, O, H. On top of that, for beta glycine, the last grid is
not used that much, unlike what is recommended in the tutorial. Therefore,
I have a few questions:
1/ Are these large cutoffs normal, even for small atoms? Is the cutoff more
dependent on the crystal structure than the atom type?
2/ Is it normal to have still a large number of gaussians distributed on
the first grid? Should I always aim for having a larger number of gaussians
on the coarsest grid than on the thinnest?
3/ How do one decide on the number of grids? Is 4 kind of a magic number or
should one aim to tune this as well? How does it affect
accuracy/performance?
I attached the input file I used to generate the series of CUTOFF and
REL_CUTOFF for both systems as well as the results I obtained.
Thank you for your help.
Best regards,
Pierre
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20190912/482dbb04/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: quercetin_cutoff.inp
Type: chemical/x-gamess-input
Size: 5624 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20190912/482dbb04/attachment.inp>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: quercetin_cutoff.dat
Type: application/octet-stream
Size: 11596 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20190912/482dbb04/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bglycine_cutoff.dat
Type: application/octet-stream
Size: 11596 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20190912/482dbb04/attachment-0001.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bglycine_cutoff.inp
Type: chemical/x-gamess-input
Size: 4222 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20190912/482dbb04/attachment-0001.inp>
More information about the CP2K-user
mailing list