I'm going to try this again.

First a comment. Often cases where a "random" presentation is given to a user/operator/player it is not a good idea to present repeat cases at all. This is especially true in situations where the number of possible options is just a few. In your case of 1 to 8 choices it is a few. It would be all different if the number of cases was 100 or 256...but 8 is a few.

So I modified the Keil C51 program to as shown below so that it eliminates repeat cases.

Save This Code

#include <stdlib.h>

unsigned char xdata array[300];

void main(void)
{
    int r;
    unsigned char rr;
    unsigned char lrr=0;
    int i;

    for(i=0; i<300; i++)
    {
        do
        {
            r = rand();
            rr = ((r / 7) % 8) + 1;
        } 
        while(rr == lrr);
        lrr = rr;
        array[i] = rr;
    }
    while(1)
    {
    }
}

The chart of the corresponding mapping is shown below for the case of the default zero for the rand() function seed.



If the list of 300 "random" non-repeating selections is sorted and plotted on a similar graph it is possible to see the relative distribution of selections at each value. See below:



Michael Karas