[molpro-user] Convergence thresholds in UHF-SCF calculations

Amrit Jalan amrit.jalan at gmail.com
Fri May 18 17:04:50 BST 2012


Kirk,
I am using 2010.1.

-Amrit

On Fri, May 18, 2012 at 12:00 PM, Kirk Peterson <kipeters at wsu.edu> wrote:

> Amit,
>
> what version of Molpro are you using?  This syntax works as advertised
> with 2010.1.
>
> best,
>
> Kirk
>
>
> On May 18, 2012, at 7:52 AM, Neeraj Rai wrote:
>
> Amrit,
>
>   My guess is that the accu is smaller than the default (accu=10), so
> probaly the defaults kick in. Can you try 11 or 12 and see what it does?
> Some of the thresholds can be changed by "gthresh". see the manual for
> different options..
>
> On Sat, May 12, 2012 at 2:37 PM, Amrit Jalan <amrit.jalan at gmail.com>wrote:
>
>> Hi,
>>
>> I am trying to change the convergence thresholds in a UHF-SCF calculation
>> but haven’t had much luck. I tried the “accu” keyword after the UHF
>> directive but the program reverted to the default value. My input is given
>> below:
>>
>>
>> geomtyp=xyz
>>
>> geometry=ts2.xyz
>>
>>
>> basis = aug-cc-pVDZ
>>
>> uhf,accu=5
>>
>>
>> The output has this:
>>
>>
>> 1PROGRAM * UHF-SCF       Authors: W. Meyer, H.-J. Werner
>>
>>
>>
>>  NUMBER OF ELECTRONS:      24+   24-    SPACE SYMMETRY=1    SPIN
>> SYMMETRY=Singlet
>>
>>  CONVERGENCE THRESHOLDS:    1.00E-06 (Density)    1.00E-07 (Energy)
>>
>>  MAX. NUMBER OF ITERATIONS:       60
>>
>>  INTERPOLATION TYPE:            DIIS
>>
>>  INTERPOLATION STEPS:              2 (START)      1 (STEP)
>>
>>  LEVEL SHIFTS:                  0.30 (ALPHA)   0.30 (BETA)
>>
>>
>>
>> Any info on where the above convergence thresholds are coming from and
>> how they can be changed would be very helpful. Also, I have noticed that
>> the defaults change depending on the calculation being run. If this is so,
>> is there someplace where this information is available?
>>
>>
>>
>> Thanks,
>> Amrit
>>
>> _______________________________________________
>> Molpro-user mailing list
>> Molpro-user at molpro.net
>> http://www.molpro.net/mailman/listinfo/molpro-user
>>
>
>
>
> --
> Regards,
> Neeraj.
>
>
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20120518/ed5bec9e/attachment.html>


More information about the Molpro-user mailing list