From: "Geo." 

well a simple solution to that would be

int n = (rand()/RAND_MAX)*99999

That should give me a pretty random distribution of the right most 4
characters, theoretically with no bias.

Geo.(there actually is a bias dependent on the time you run the program)

"Paul Ranson"  wrote in message
news:41b58a1f{at}w3.nls.net...
> It depends how random you want your results to be. 'rand' returns what
> should be an even distribution between 0 and 32767, so a simple '%10' has
a
> slight bias away from 8 and 9, if you look at the second and third more
> significant digits then they will have a bias towards 0 and also away from
> 7, 8, 9. If 'rand' returns 8 you will get 0, 0, 8 etc.
>
> Paul
>
> "Geo"  wrote in message
news:41b4ecad$1{at}w3.nls.net...
> >> Another point, probably unimportant. You could extract the three
> >> digits that you want from one call to rand, as below:
> >>   int n = rand();
> >>   itoa((n%10),a,10);
> >>   itoa(((n/10)%10),b,10);
> >>   itoa(((n/100)%10),c,10);
> >
> > See? Now that's the kind of stuff they never show you in a programming
> > book.
> > Now I'm going to have to do a whole new program just to play around with
> > this and see exactly what it's doing so I can build my mental image.
> >  That's really slick, it allows me to extract just one
> > digit
> > out of a number so if n=4581 then c ends up being the 5. Very handy
> > technique, that was probably more useful than learning about itoa.
> >
> > Geo.
> >
> >
>
>

--- BBBS/NT v4.01 Flag-5