Strange behavior with Max_SCF of GEO_OPT and ENERGY runtype

[Natalie Austin]

Hello,

I've noticed for a few RUN_TYPE GEO_OPT calculations even though the SCF 
did not converge (reached MAX_SCF=500 iteration set), since the convergence 
criteria was met the geometry optimization was complete. I've worked with 
Turbomole and Gaussian and in those programs if the SCF doesn't converge 
within the MAX_SCF iteration I've set, the calculation will terminate 
without continuing the optimization.
I find it strange that even if the SCF has failed CP2K gives an output that 
represents what you'd see in a normal termination. I found this to be the 
case for a few RUN TYPE ENERGY calculations as well. Would anyone be able 
to explain why this is the case? I've attached an example input and output. 

Thanks,

Natalie
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20151120/974cd628/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: cu54ni_cn6_m3_new_300k.inp
Type: chemical/x-gamess-input
Size: 2120 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20151120/974cd628/attachment.inp>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: cu54ni_cn6_m3_new_300k.out
Type: application/octet-stream
Size: 2820219 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20151120/974cd628/attachment.obj>