[molpro-user] MRCISD+Q calculations

[Dawid das]

Evgeniy,

I have already increased and decreased the memory. I tried to give even
3000 M words when I do so I get:
gmainv1 failure to allocate 3000230002

My machine has definetely enough memory to do these calculations.
Is it possible that the molpro is so compiled it cannot handle sufficient
amount of memory?

Dawid


2014-03-02 18:10 GMT+01:00 Kirk Peterson <kipeters at wsu.edu>:

> Dawid,
>
> typically I save stuff like this to records 4000 and above on file 2.
>  Something like:
>
> 1st CI:
> ci;wf,35,2,1;save,4100.2
>
> 2nd CI:
> ci;wf,35,3,1;save,4200.2
>
> transition moment:
> ci;trans,4100.2,4200.2
>
> -Kirk
>
> On Mar 1, 2014, at 9:11 AM, Dawid das <addiw7 at googlemail.com> wrote:
>
> Alright, I have finally managed to run those casscf and mrci calculations.
> But how can I calculate transition moments (oscillatory strength) for
> those mrci wave functions?
>
> http://www.molpro.net/info/2012.1/examples/h2op_mrci_trans.com
>
> In the example above I need to specify the records where my wave functions
> are, right?
> So, how do I know numbers of records?
> What I want to do here is to calculate transition moments or oscillatory
> strengths for
> transitions between states 1-2, 1-3, 1-4 and 1-5.
> How do I do it?
>
> Best wishes,
> Dawid
>
>
> 2014-02-18 15:38 GMT+01:00 Dawid das <addiw7 at googlemail.com>:
>
>> I discovered that I am able to perform HF geometry optimization or CCSD
>> energy calculations.
>> But when I use mrci command I got an error like above. Any idea what can
>> cause it?
>>
>> Best wishes.
>>
>>
>> 2014-02-18 10:27 GMT+01:00 Dawid das <addiw7 at googlemail.com>:
>>
>>
>>>
>>>
>>> Kirk Peterson kipeters at wsu.edu:
>>>
>>>> Not sure what you mean by restricting the MRCI to just singles and
>>>> doubles.  Do you mean singles and doubles from a HF reference of 5
>>>> orbitals/10 electrons into just 5 virtuals? In Molpro you can't restrict
>>>> the number of virtuals you use for correlation in the MRCI.  Could you
>>>> describe more closely what you would like to do here?
>>>>
>>>>
>>> What I meant here is that I have 10 pi electrons on 5 doubly occupied
>>> orbitals. Now I want only 5 occupied and 5 lowest lying virtual orbitals to
>>> be active, that means  the occupation number is more or equal to 0 and less
>>> or equal to 2. All of other occupied orbitals are to stay doubly occupied.
>>> All of other virtual orbitals are to stay not occupied.
>>> Now I was trying to achieve it on a smaller system with 6x6 complete
>>> active space, that means 3 doubly occupied orbitals and 3 virtual oribtals.
>>> There are 22 doubly occupied orbitals all together. I attached my input and
>>> output files. There is obviously something wrong. It looks like problem
>>> with queuing system to me. Nevertheless I'd like you to take a look at my
>>> input. I did not specify the active space in mrci command block because it
>>> will be taken from the last command block, right?
>>>
>>>
>>> Thank you!
>>>
>>> Dawid
>>>
>>
>>
> _______________________________________________
> 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/20140302/17615e4b/attachment.html>