Since the code originally was attempting to mimic ipl/procs/random.icn,
which had a one-second resolution, the initialization seed would change
each second. Good enough for normal human variation from run to run, not
good enough for a script to execute frequently and vary from run to run.
Since Windows clock was traditionally 54ms one might imagine that
modifications using ordinary clock C library calls might still not issue a
different seed for script-launched adjacent executions.
I certainly endorse the notion of using a process id or its equivalent to
further spread out the seed initialization, and agree that whatever
improvement we come up with, specialists will always have
reason to resort to their higher end C math libraries for better random
numbers or better seeds,
so I hear what folks are saying about a desire to enhance the C library
calling capability.
Cheers,
Clint
------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
_______________________________________________
Unicon-group mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/unicon-group
