Merrick
04-11-2014, 03:12 PM
Hello,
I am working with some legacy code that actually uses RAN1 from the NR First Edition Fortran Version. I saw on your "Distressing Rumours..." page that there were " slight, but genuine, problems in the First Edition [random number generator] routines [that] are completely fixed in the Second Edition."
I looked in the forums and I Googled a bit but I was unable to find any discussion of what issues existed for the First Edition routines (was RAN1 problematic? how?) and saw no discussion in the Second Edition of why RAN1 in the First Edition and ran1 in the Second Edition are so different.
Is there any place where you discuss what problems existed in RAN1 in the First Edition that resulted in the change to ran1 in the Second Edition, or is ran1 just an improvement, either in periodicity or execution time?
Thanks...
I am working with some legacy code that actually uses RAN1 from the NR First Edition Fortran Version. I saw on your "Distressing Rumours..." page that there were " slight, but genuine, problems in the First Edition [random number generator] routines [that] are completely fixed in the Second Edition."
I looked in the forums and I Googled a bit but I was unable to find any discussion of what issues existed for the First Edition routines (was RAN1 problematic? how?) and saw no discussion in the Second Edition of why RAN1 in the First Edition and ran1 in the Second Edition are so different.
Is there any place where you discuss what problems existed in RAN1 in the First Edition that resulted in the change to ran1 in the Second Edition, or is ran1 just an improvement, either in periodicity or execution time?
Thanks...