Dear Dr. Henning Heiberg-Andersen,

Did you set the MAXMEMORYUSAGE key also in the EPR input?
That is crucial. If this does not solve the problem, you should add
something like

VECTORLENGTH 10
ALLOW SMALLBLOCKS

in your EPR input. This will make the calculation slower (because only
10 integration points instead of the default 128 are treated
simultaneously,
which is not optimal because of some extra overhead), but it will make
sure that the EPR program needs significantly less memory.
More precisely, it will use 12.8 times less memory for those arrays
related to the integration grid. In the EPR code some of the largest
arrays are of this type.

I hope this helps.

Best regards,
Stan van Gisbergen

On Sep 16, 2005, at 11:26 AM, Henning Heiberg-Andersen wrote:

>
> Dear ADF users,
>
> I am trying to calculate the EPR g-Tensor for a spin 1/2 radical with
> ~ 100 atoms with the NMR/EPR program. No matter how I set the
> MAXMEMORYUSAGE key, I get the message
>
> IRALOC: insufficient memory in the system
>
> Any recommendations?
>
> Best regards,
> Henning Heiberg-Andersen
>
>
Dr. S.J.A. van Gisbergen      
Scientific Computing & Modelling NV
Theoretical Chemistry, Vrije Universiteit
De Boelelaan 1083
1081 HV Amsterdam
The Netherlands                                
vangisbergen_at_scm.com  
http://www.scm.com

Please note NEW FAX AND TELEPHONE NUMBERS:
T: +31-20-5987626    
F: +31-20-5987629      
Received on 2005-09-16 11:34:48