Firefly and PC GAMESS-related discussion club

Learn how to ask questions correctly

Alex Granovsky
gran@classic.chem.msu.su

>   Command KDIAG=0 in documentation FireFly_input_rev002
> is new as it is allocated by red colour. Actually, this command
> not new, and for it, possibly, the new code, than at Gamess (US)
> (look the text for KDIAG=1) is written. I do not have last version
> Gamess (US), but it is interesting, and in this program is not
> displayed KDIAG=0 in group $guess?

>   If it is not displayed, this error in a code has passed in
> FireFly from Gamess (US) and code FireFly should be corrected.


With Firefly, there are two input groups where one can enter KDIAG:

 $system kdiag=

and

 $guess kdiag=

The first one controls system-wide behavior, but with some exceptions.
One of them is the initial guess procedure, so that the second
option allows one to specify diagonalization routine used by the
inital guess procedure. This affects time spent in the initial
guess code but does not affect its results. This is why this option
is not printed in the output file. This is by design and we do not
plan to print out all of thousands alailable options which do not
affect results of calculations but can be used to fine-tune the code.

> Otherwise what for then listing of commands in OUTPUT is necessary?
> I answer itself: to check up, whether took FireFly this command in
> work, instead of has ignored it.

That's not true. The listing is necessary to print out the
most important options. However, if the input group was was
read in, and there is no warning or error message(s) in the
output file on syntax errors in the input, one can be sure
that all keywords of the onput group were recognized. This
does not automatically mean all of them will be in effect
and will not be modified by Firefly, as some combinations
of keywords are illegal and in most cases Firefly automatically
fixes this.

You are right that non-recognized input groups (note, not keywords!)
are silently ignored at present. Once again, this is by design.
One can verify whether the particular input group was recognized
running exetyp=check and searching output for "check shows" comments.

There is also the way how to get the complete list of all keywords
of any available input group in the case one is interested to dig
a bit deeper.  One just need to put here some illegal keyword, e.g.:
$group abc=defgh $end

>   So, unfortunately, I have spent a lot of time, yet have not found
> out that FireFly ignores many commands from code Chem (Bio) 3D
> (CSGamess CambridgeSoft Corp.). Because of it there were my
> questions at the Forum:

> http://classic.chem.msu.su/cgi-bin/ceilidh.exe/gran/gamess/forum/?C35d7f1aa6eEV-7389-47-00.htm

> http://classic.chem.msu.su/cgi-bin/ceilidh.exe/gran/gamess/forum/?C34df668afbHW-7391-1334+00.htm


Why did you expect Firefly to be compatible with CSGamess or GAMESS (US)? It's not GAMESS (US) nor CSGamess - it's Firefly.

Regards,
Alex Granovsky