Re: how to use seeds?

Hello,
first of all, I have to apologize, the EVENTDAT section of manual reports
an obsolete feature. The numbers printed in the eventdat are the initial
seeds of the random sequence and the total number of random seeds
extracted up to that point, they are NOT a complete initialiser. To obtain
the complete seed one should initialize and call the random generator for
the given number of times. The possibility to skip random numbers from
input cards has been removed a few years ago when the random number
generator has been changed.

 I propose two solutions:

1) IF your "global" run did not take too long, re-do it as a series of
runs with one particle each.(scoring can be taken out)

                 rfluka -M998 -N0 xxx

 In this way you will automatically get the complete seed after each
event, in the files ranxxx001.dat....ranxxx999.dat. Then, you can re-run
the input on a specific seed: for instance, for event 25

      rfluka -N24 -M25 xxx

the random ranxxx025.dat will be taken.
If you have more than 999 events, you can go on by changing names, for
instance:
 cp xxx.inp xxxa.inp
 cp ranxxx999.dat ranxxxa001.inp

2) if 1) takes too long, tell me, I'll prepare an utility that skips a
given number of randoms and writes down the desired full seed.

ement in eventdat the possibility to dump the full seed at each event

Ciao

Paola

> Hello,
>
> How do I use the seed information in order to rerun a certain event?
> My problem is the following: I have simulated a sample of events and
> scored the energy deposit in each event using the EVENTDAT card. Now I
> would like to reproduce some of these events (and only these events)
> in order to study them in more detail. How do I do this?
>
> The manual says that one can do this using information about the
> random number generator seeds in the EVENTDAT output (ISEED1, ISEED2,
> SEED1, SEED2, SOPP1, SOPP2), but I don't understand how to use it.
>
>
> Thank you,
>
> Konstantin
>

Paola Sala
INFN Milano
tel. Milano +39-0250317374
tel. CERN +41-227679148
Received on Tue Jan 13 2009 - 10:28:09 CET