C - Simulator of dice rolling - c

Question

The intention here is to write a C code for 2 dices rolling simulator.
It has to write the answer in a form of histogram (10 results by line).
Calculate and give the minimum, maximum, and the average of dice rolls. As well as the duplicates. I have to admit I am stuck right now. Any help would be appreciated.
Here's the code:
#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#define N 100
void initarray(int d[]);
void printarray(int d[]);
void printstar(int n);
void fancyprint(int d[]);
int roll(void);
int main(int argc, char* argv[]) {
{
int roll();
int d[13]; /* 2-12 hold num of rolls */
int i; /* variable de la boucle */
int starttime = time(NULL); /* temps horloge*/
srand(1020);
printf("Simulation de 100 lancees de 2 des");
for (i = 0; i < N; i++)
d[roll() + roll()]++;
printarray(d);
fancyprint(d);
printf("Elapsed time: %ld seconds\n",
time(NULL) - starttime);
return 0;
}
/* initarray: initialize statistic array */
void initarray(int d[]){
int i;
for (i = 2; i < 13; i++)
d[i] = 0;
}
/* printarray: print each num of rolls */
void printarray(int d[]){
int i;
double e[] = {0, 0,
1.0/36.0, 2.0/36.0, 3.0/36.0, 4.0/36.0,
5.0/36.0, 6.0/36.0, 5.0/36.0, 4.0/36.0,
3.0/36.0, 2.0/36.0, 1.0/36.0};
printf("Sum Times Frequency");
printf(" Exact Diff\n\n");
for (i = 2; i < 13; i++)
printf("%2d %7d %11.7f %10.7f %8.4f\n",
i,
d[i],
(double)d[i]/N*100.0,
e[i]*100.0,
((double)(d[i]) -
e[i]*N)/N*100.0);
}
/* printstar: print n stars */
void printstar(int n) {
while (n > 0) {
printf("*");
n--;
}
}
/* fancyprint: print bar graph */
void fancyprint(int d[]){
int i;
printf("\n");
for (i = 2; i < 13; i++) {
printf("Sum:%3d |", i);
printstar(300*d[i]/N);
printf("\n");
}
printf("\n");
}
/* roll: simulate rolling a die */
int roll() {
return (int) (6.0*(rand()/(double)RAND_MAX)
+ 1.0);
return 0;
}
}

Answer 1

OP's roll() generates a 7 once in a while: when rand() == RAND_MAX.
int roll(void) {
// return (int) (6.0*(rand()/(double)RAND_MAX) + 1.0);
return (int) (6.0*(rand()/((double)RAND_MAX + 1) + 1.0);
}
The above will have trouble if (double)RAND_MAX is not exact.
It also still has a small bias.
Consider other approaches that do not use floating point. Such as: https://stackoverflow.com/a/17554531/2410359
Possible other issues.