[molpro-user] bug of analytic gradient calculation

[Andy May]

Wenli,

Thanks for the info, I've reproduced the problem and made a bug report:

https://www.molpro.net/bugzilla/show_bug.cgi?id=4666

The problem doesn't exist in our master version, but unfortunately it's 
not clear (to me) which commit was responsible for fixing it.

Best wishes,

Andy

On 08/04/15 14:21, ZOUWL wrote:
> Hi Andy,
>
> Attached please find the output files of molpro 2010. Maybe it has
> been corrected in the 2012 version. Thanks!
>
> Best regards,
> Wenli
>
> On Tue, Apr 7, 2015 at 11:11 AM, Andy May <MayAJ1 at cardiff.ac.uk> wrote:
>> Wenli,
>>
>> I tried to reproduce the problem with your input, but it ran correctly. Can
>> you send the full output file from the input that fails, and let me know
>> which version of Molpro you are using?
>>
>> Thanks,
>>
>> Andy
>>
>>
>> On 05/04/15 14:53, ZOUWL wrote:
>>>
>>> Dear developers,
>>>
>>> In the following input, if a molden file is saved between the MULTI
>>> and OPTG steps, the optimization calculation will crash. It seems that
>>> the gradients are missing.
>>>
>>> ***,test
>>> memory,20,m
>>>
>>> basis=3-21g
>>>
>>> geometry={
>>> C
>>> O 1 r1
>>> }
>>>
>>> r1 = 1.3 angstrom
>>>
>>> hf;
>>>
>>> {multi;natorb,2352.2;dm,2352.2;}
>>>
>>> !put,molden,molpro.mol;orb,2352.2;
>>>
>>> optg
>>> ---
>>>
>>> Best regards,
>>> Wenli
>>>
>>> _______________________________________________
>>> Molpro-user mailing list
>>> Molpro-user at molpro.net
>>> http://www.molpro.net/mailman/listinfo/molpro-user
>>>
>>
>