Firefly and PC GAMESS-related discussion club

 
Learn how to ask questions correctly  
 
 

 

Alex Granovsky
gran@classic.chem.msu.su

Finally, with Firefly 8.0.0 RC 31 we were able to solve this issue.
The output file is attached to this post. There were several problems
with both your input file and with Firefly. In a nutshell, the problems
were related to the use of DLCs with the $force purify=1,
$force prtifc=1, and $force decomp=1 options. These options cannot be
used with DLCs and/or natural internals or symmetry coordinates.
This was most likely poorly documented in the manual.

Kind regards,
Alex Granovsky

P.S. I'd suggest using $force vibsize of at least 0.002 or 0.003 Bohr.
Your use of vibsiz=0.001 Bohr results in a considerable numerical
noise in the hessian and frequencies as 0.001 Bohr seems to be too
small for reliable differentiation. In addition, use of tighter nconv=8 would be a good idea.







On Wed Sep 19 '12 6:20pm, Serge Pisarev wrote
---------------------------------------------
>Dear Alex,

>The #7341 calculations have shifted the imaginary frequency down to 824.67i, nevertheless, it is still imaginary (the output is attached again). Should I consider the fault in 71G #5618 calculations that had indicated the minimum?
>I should also mention that the optimization proceeded in 20 steps for the #5618 calculations, while #7317, #7339 and #7341 all required only 6 steps and gave slightly lower MP2 energies.

>With best regards --- Serge.

>On Mon Sep 17 '12 9:10pm, Alex Granovsky wrote
>----------------------------------------------
>>Dear Serge,

>>sorry it's our fault. Please try Firefly v. 8.0.0 RC 31 build # 7341

>>Kind regards,
>>Alex

This message contains the 348 kb attachment [ NO-CC_fixed.rar ]