The ABINIT Users Mailing List ( CLOSED )

["Anglade Pierre-Matthieu" <anglade@gmail.com>]

Hi,

Strange :
- Abinit allways goes down to the end except in case there is an error
(because of the compiler, the system, a bug ...)
- Error always produce an error message. Yet it is not always in the
log file and can "popup" in the standard error stream.
If you can't get any error message even after following my previous
advice (did you tried ?), and you are sure Abinit stopped, I guess no
one will be able to help you. Sorry.
Still a piece of advices : for molecule it could be wise to use a code
aimed at molecules like BigDFT or any other that use space localized
basis set.

Regards

PMA

On Sat, Jun 14, 2008 at 6:11 PM, Maryam jamali <mar_jamali@yahoo.com> wrote:
> Hi,
>
> The job was run with 2 different computer: a) 4 GB of RAM, 2.4 GHz CPU   b)
> 80 GB of  RAM, 2.4 GHz CPU.
> in both of them the same problem was watched:  the job is stopped and Idid
> not see any error in log and output files.
>
> Thank you in advance.
>
> Maryam
>
> ----- Original Message ----
> From: Anglade Pierre-Matthieu <anglade@gmail.com>
> To: forum@abinit.org
> Sent: Saturday, June 14, 2008 11:21:28 AM
> Subject: Re: [abinit-forum] The convergence in ecut "stopped again"
>
> Hi,
>
> How powerfull is your computer ? With such a big cell you may be going
> beyond the capabilities of your computer. Do you get any message in
> the log file or in the standard error output ?
> (If you do not know the answer, you can run again your job with "bash;
> abinis < my.files &> log-error" and look at the end of the file
> "log-error").
>
> regards
>
> PMA
>
> On Fri, Jun 13, 2008 at 9:45 PM, Maryam jamali <mar_jamali@yahoo.com> wrote:
>> Dear all,
>>
>> I have inserted my molecule in a supercell and I want to find optimized
>> ecut. When I increase ecut or acell, the job is stopped without any
>> error.Also I added "iscf 5"  to input file, but it did not work and the
>> job
>> is stopped again after DATASET 3. What does go wrong with the job? I do
>> not
>> know what I should do to avoid this trouble.
>>
>> Thank you in advance.
>>
>> You can see the input as follows.
>>
>>
>>
>> iput file:
>>
>> # tube molecule in a big box
>>
>>  ndtset 5
>>
>> ixc 11
>> acell 30 30 5 angstrom
>> angdeg 90 90 120
>>  ecut: 5  ecut+ 5
>>
>>    natom 32
>>
>>    xangst 2.198749        2.198749        -1.7625
>>          1.189919        2.872833        0.3525
>>          2.198749        2.198749        -0.3525
>>          1.189919        2.872833        1.7625
>>          0.000000        3.109431        -1.7625
>>          -1.189919        2.872833        0.3525
>>          0.000000        3.109431        -0.3525
>>          -1.189919        2.872833        1.7625
>>          -2.198749        2.198749        -1.7625
>>          -2.872833        1.189919        0.3525
>>          -2.198749        2.198749        -0.3525
>>          -2.872833        1.189919        1.7625
>>          -3.109431        0.000000        -1.7625
>>          -2.872833      -1.189919        0.3525
>>          -3.109431        0.000000        -0.3525
>>          -2.872833      -1.189919        1.7625
>>          -2.198749      -2.198749        -1.7625
>>          -1.189919      -2.872833        0.3525
>>          -2.198749      -2.198749        -0.3525
>>          -1.189919      -2.872833        1.7625
>>          0.000000      -3.109431        -1.7625
>>          1.189919      -2.872833        0.3525
>>          0.000000      -3.109431        -0.3525
>>          1.189919      -2.872833        1.7625
>>          2.198749      -2.198749        -1.7625
>>          2.872833      -1.189919        0.3525
>>          2.198749      -2.198749        -0.3525
>>          2.872833      -1.189919        1.7625
>>          3.109431        0.000000        -1.7625
>>          2.872833        1.189919        0.3525
>>          3.109431        0.000000        -0.3525
>>          2.872833        1.189919        1.7625
>>
>> toldff 5.0d-5
>> ntypat 1
>> znucl 6
>> typat 32*1
>>
>> nkpt 1
>> nstep 50
>>
>> iscf 5
>> diemac 12.0
>>
>
>
>
> --
> Pierre-Matthieu Anglade
>
>
> Compose Email:
> anglade@gmail.com
> Add to Contacts
>



-- 
Pierre-Matthieu Anglade