Re: [ESPResSo-devel] Re: [ESPResSo-users] RDF function in large systems

[Ulf Schiller]

Hi,

Axel Arnold wrote:
> On Tuesday 08 February 2011 09:13:50 Olaf Lenz wrote:
>>>> All integers are so far 32 Bit, so that is in general possible. With
>>>> 450,000 particles the total count of samples is certainly too large.
>>>> You can just change the variable cnt to long long in the calc_rdf
>>>> function, that should fix it.
>> Would it make sense to modify these variables to be long int in general?
>
> I guess so, I don't think it will make things drastically slower.

One could also use a #define'd datatype, e.g.

#define ANALYSIS_DT_INT <int | long int>

to change it easily.

I came across a similar issue when using very large lattices in LB,
where I actually needed a 64bit integer index for an array. There, I
used #define to keep it portable.

Best,
Ulf

--
Dr. Ulf D. Schiller                         Building 04.8, Room 231a
Institute of Complex Systems (ICS-2)        Phone:  +49 2461 61-1765
Forschungszentrum Jülich, Germany           Fax:    +49 2461 61-2850

------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich
Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzender des Aufsichtsrats: MinDirig Dr. Karl Eugen Huthmacher
Geschaeftsfuehrung: Prof. Dr. Achim Bachem (Vorsitzender),
Dr. Ulrich Krafft (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt,
Prof. Dr. Sebastian M. Schmidt
------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------