Bad condition number R_COND

[Nilesh Dhumal]

Hello User, 

I installed the latest version of cp2k (2323.1) on our cluster.  

I am getting the Bad condition number R_COND error for my every input file. We have 8.2 version installed on the cluster. The simulation finished successfully using the older version (8.2) of cp2k. 

Attached is input and output file. 

Is the issue in input file or compilation ? 

Thanks, 

Nilesh 

[Matt Watkins]

Hello,

if the input previous ran successfully it sounds like a problem with the compilation.

Matt

[centre....@gmail.com]

I change the mixing method  PULAY_MIXING to  DIRECT_P_MIXING.

ITS RUNNING. i RECEIVED THE ERROR FOR THE PULAY_MIXING.

NOT SURE ITS COMPLIATION ISSUE OR METHOD?

[Matt Watkins]

OK, yes, seems to be an issue showing up with PULAY_MIXING and variable cell. (weirdly Broyden and Kerker seem fine).

Practically, you are better using standard mixing and DIIS or OT for most insulating systems.

Matt