On Tue, Apr 13, 2010, Adrian Roitberg wrote:
>
> 1. It would make it impossible to reproduce data exactly (let's say for
> debugging) unless the user sets him/herself the seed by hand.
The program prints out what value of ig is chosen, so one can use that
value as input for a re-run to get the same set of random numbers.
So, one does not have to know in advance that a re-run might be desirable;
it's always possible just by copying the seed in the old output file to the
new input file.
>
> 2. The test cases MUST have a seed set by hand, otherwise they would
> never pass.
I think this is killer for right now, since there are hundreds of tests that
would be need to be updated.
.....dac
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Tue Apr 13 2010 - 08:30:02 PDT
