This archive web page is obsolete!

Please refer to the new mailman archive! Re: running simulations stepwise

[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: running simulations stepwise


I agree and it was to some extent discussed at the workshop in relation to the request for logging neutron events. I think that logging events is the first step and should be carried out in a format that it can be used to resimulate one part of the instrument with a different setup and to carry out analyses that have not been done during the simulation. However, I am currently not sure how complicated it will be to modify the source code to include this feature.

I will also try to summarize some of the suggestions given at the workshop in some more detail at the web-page. It would be quite useful for future discussions.

Best regards,


Peter Link wrote:

> Hello,
> throughout of some discussion I had here with Ulrich, the problem came up, whether McStas should include the possibility to use the result of one simulation as input of an other one. The application in my mind is to separate the simlulation of the primary spectrometer ( beam-hole, monochromator) from the secondary part (sample, analyser etc.) in order to gain in computing time, while optimising the secondary part. In the workshop summary there has been mentionend some history logging feature. This might be the high-end solution. I think in a first step, one could think to sample the 'useful' neutrons data (xyz,vx...) into a file which serves as 'virtual' source.
> Best regards, Peter
> ***************************************
> Dr. P. Link
> Dreiachsenspektrometer PUMA
> IPC Uni G÷ttingen
> Aussenstelle am FRM-II
> 85747 Garching
> Tel. 089 289 14622
> Fax 089 289 54622
> ***************************************

Per-Olof ┼strand
Dept. of Chemistry, University of Copenhagen
and Materials Research Department, Ris° National Laboratory or and