Ceilidh ... Re^5: How to speed up the HESSIAN calculations

Firefly and PC GAMESS-related discussion club

Learn how to ask questions correctly

We are NATO-free zone

Re^5: How to speed up the HESSIAN calculations

Dear Thomas,
Thank you for a prompt reply. I am glad to know that I have already specified most of the important keywords. Regarding your suggestion of FDIFF=.t., I am unsure about it, as I am unsure that even after OPTTOL=10E-7, the geometry has converged or not... At 10E-7, it did converge though... But unsure whether it is the global minimum or not...For that I am running Hessian calculations now.

On Wed Jan 9 '13 11:39am, Siddheshwar Chopra wrote
--------------------------------------------------
>Dear Pasha,
>Yes true Sir. If I am not wrong, both the runtyp=hessian and hessend=.t. mean the same???

>Regards,

>On Tue Jan 8 '13 8:00pm, Solntsev Pasha wrote
>---------------------------------------------
>>Dear Siddheshwar.

>>One more thing. You don't need specify runtyp=hessian and hessend=.t. simultaneously. Use only runtyp=hessian for hessian matrix calculation.
>>
>>
>>PS.

>>On Tue Jan 8 '13 3:01pm, Siddheshwar Chopra wrote
>>-------------------------------------------------
>>>Dear Thomas,
>>>As asked by you, I am hereby attaching the SAMPLE.inp for your reference. Please suggest me what to do? This sample is the smallest of all.

>>>Regards,

>>>On Tue Jan 8 '13 11:51am, Thomas wrote
>>>--------------------------------------
>>>>Dear Siddheshwar,

>>>>Could you be a bit more specific? For example, could you post the input file you've been using so far for Hessian calculations? Otherwise, your question is too general for us to be helpful. :-)
>>>>
>>>>
>>>>Kind regards,
>>>>Thom
>>>>
>>>>
>>>>
>>>>
>>>>On Tue Jan 8 '13 7:29am, Siddheshwar Chopra wrote
>>>>-------------------------------------------------
>>>>>Dear Sir,
>>>>>I wish to know if there are any ways I could SPEED up the HESSIAN calculations?

>>>>>Regards,

Wed Jan 9 '13 12:04pm

This message read 876 times