Firefly and PC GAMESS-related discussion club

 
Learn how to ask questions correctly  
 
 

 

Alex Granovsky
gran@classic.chem.msu.su

you are using incorrect syntax. The correct one could be:

firefly.exe -f -i Mn-QDPT-sex.inp -o C:\FIREFLYPLL\Mn-QDPT-sex.out TEMP1 "TEMP2" -np 2  

The -np command line option should be the last one (actually, it is
parsed by MPI, not by Firefly itself). Note the use of -o option
instead of output redirection. IN addition you can use -t option
to specify the root of working directories, see the list of command
line options here:

http://classic.chem.msu.su/gran/gamess/comm_line.html

Kind regards,
Alex Granovsky






On Sun Apr 29 '12 4:13pm, Wibren Oosterbaan wrote
-------------------------------------------------
>Dear Alex,

>Thanks a lot. I have managed to run the example file on 1 core using:
>firefly.exe -f -i Mn-QDPT-sex.inp > C:\FIREFLY\Mn-QDPT-sex.out

>but I fail to get it running on multiple cores. For example, if I run:
>firefly.exe TEMP1 "TEMP2" -np 2 -f -i Mn-QDPT-sex.inp > C:\FIREFLYPLL\Mn-QDPT-sex.out
>
>
>I get this error message:

> CAN NOT OPEN FIREFLY INPUT FILE. THIS FILE EITHER SHOULD BE NAMED 'INPUT',
> OR IT SHOULD BE DEFINED USING THE 'INPUT' ENVIRONMENT VARIABLE.

> FOR EXAMPLE, YOU CAN DO SOMETHING LIKE FOLLOWING:

>     COPY D:\MYDIR\MYTASK.INP INPUT
>     C:\FIREFLY\FIREFLY.EXE >MYTASK.OUT

>     OR

>     SET INPUT=D:\MYDIR\MYTASK.INP
>     C:\FIREFLY\FIREFLY.EXE >MYTASK.OUT

>But if I follow that advise, and use

>SET INPUT=C:\FIREFLYPLL\Mn-QDPT-sex.inp
>firefly.exe TEMP1 "TEMP2" -np 2 -f > C:\FIREFLYPLL\Mn-QDPT-sex.out
>
>
>I get:
>**** ERROR READING VARIABLE SCFAC    CHECK COLUMN 47
>  O2     8   4.82611   -1.00280    7.50310     @O_6-31G.bas                      
> ....V....1....V....2....V....3....V....4....V....5....V....6....V....7....V....8

>Also in other ways I don't seem to get it working. What am I doing wrong?

>Best regards,
>Wibren
>
>
>On Sun Apr 22 '12 6:45pm, Alex Granovsky wrote
>----------------------------------------------
>>Dear Wibren & Luca,

>>another solution is to use input file preprocessing feature of Firefly
>>as shown in this example (note it must be used together with -i command line option).
>>Please let us know if you need more details or comments on this.

>>Kind regards,
>>Alex Granovsky
>>
>>
>>
>>On Mon Apr 16 '12 3:13pm, Luca Maidich wrote
>>--------------------------------------------
>>>Good morning Wibren,
>>>yes, it is possible to define more than one basis set for each atom in a single BASIS.LIB file as long as they do not have the same name, i.e. not the same string to call them as in your attached file.
>>>The problem arises when you try to load these two different basis set on the same type of atom, in your case carbon. As far as I know the only workaround is to manually define the basis set in the $DATA deck, like in the attached files (NewLinearDep), I know that it could be annoying with many atoms.
>>>In your case the input could be shortened because the 6-31G(d) basis set is stored internally in firefly, have a look at the attached files (NewLinearDep2).
>>>I know that this sounds like a "quick and dirty" solution but at least works :)

>>>Kind Regards
>>>Luca