The ABINIT Users Mailing List ( CLOSED )

[marc.torrent@cea.fr]

Dear Aman,

Shortly, some answers to your questions :

1. You can look at tests_v4 #04 to #06 (in Abinit's directory); but these are 
premilinary tests for developpers.

2. No ! (as said in web page).

3. As mentioned in release notes of Abinit :
"the PAW part of ABINIT is available for BETA TESTING" (see B3. in release 
notes).
However, we made some tests... and they still are to be confirmed.
For example, we obtained good results for Oxygen (Ecut=12 Ha for PAW ; 
Ecut=30 Ha for norm-conserving).. and several other materials.
Cut-off energies may be quite smaller than those in norm-conserving 
pseudopotentials (but you necessarily have to test it by yourself).
CPU times may be 15% higher than in norm-conserving case (IF you use the SAME 
Cut-off - and of course you will not).


Marc Torrent
CEA-Bruyeres-le-Chatel - France

=======================================================

Dear All -

I have a quick question on the PAW implementation of ABINIT.

1. Is there any example file to go through in the list of tests?

2. Can I use the files already present in the
http://www.abinit.org/ABINIT/Psps/PAW (it says for versions > 4.2.1) with
4.1.4?

3. If I use the PAW pseudo potential file generated from USPP2Abinit utility,
OR from the peroidic table links above mentioned, what might be the typical
ranges of "ecut" and the typical CPU times (an example) compared with the
norm-conserved Psps?

Thanks very much in advance

Best,
--
Aman