comp.lang.idl-pvwave archive: archive

Home » Public Forums » archive » Re: Random selection

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Re: Random selection [message #51242]

Sun, 12 November 2006 05:07

greg michael
Messages: 163
Registered: January 2006

Senior Member

Kenneth P. Bowman wrote:

> The problem with this approach is that it is actually rather likely
> that you will generate two (or more) numbers that round to the
> same 3-digit integer.
>
> Sorting floats, on the other hand, involves no loss of precision.
>
> If you really want to avoid duplicates, generate double-precision
> random numbers
>
> r = RANDOMU(seed, 1000, /DOUBLE)
>
> Cheers, Ken

I'm not sure doubles are going to help - you'll get the same problem at
3 digits. And if you use the sort method, it doesn't really matter if
you get two the same - they'll still map to unique indices.

many greetings,
Greg

Report message to a moderator

[Message index]

		Re: Random selection By: greg michael on Sun, 12 November 2006 05:07
		Re: Random selection By: Julio[1] on Sun, 12 November 2006 04:57
		Re: Random selection By: Kenneth P. Bowman on Sat, 11 November 2006 20:09
		Re: Random selection By: David Fanning on Sat, 11 November 2006 14:09
		Re: Random selection By: Jean H. on Sat, 11 November 2006 13:25
		Re: Random selection By: David Fanning on Sat, 11 November 2006 13:08
		Re: Random selection By: Jean H. on Sat, 11 November 2006 12:45
		Re: Random selection By: David Fanning on Sat, 11 November 2006 05:30
		Re: Random selection By: greg michael on Sat, 11 November 2006 05:28
		Re: Random selection By: David Fanning on Sat, 11 November 2006 05:23
		Re: Random selection By: JD Smith on Mon, 13 November 2006 10:46

Previous Topic:	Re: colorbar positioning in postscript
Next Topic:	canonical correlations

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Oct 08 17:00:29 PDT 2025

Total time taken to generate the page: 0.00401 seconds