Question about the rand() function

[gedamial]

Hello.

I want to make a program that, given the minimum value and the maximum value, generates a random number within that range.

I made that with this code:

#include <iostream>
#include <conio.h>
#include <ctime>

using namespace std;

int main()
{
	srand(time(NULL));

	int min = 0, max = 2;

	int val = ((max + 1) - min) + min;

	int num = rand() % val;

	cout << "Random number: " << num << endl;
	cout << "Value: " << val << endl;


	_getch();
	return 0;
}

If you see, the min is 0 and the max is 2.
The second COUT (that prints the final value) prints 3

Now, if I change the range from 0-2 to 1-2, the second cout still prints 3

But... if the value is always 3 that means that this code

rand() % val

is equivalent to

rand() % 3

But the point here is that, despite the value is the same, the program has 2 different behaviors

In the first "0-2" case, it prints these values {0,1,2}
In the "1-2" case, it prints these ones {1,2}

How can the program work in 2 different ways with the same value (3)???

[Peter87]

val = max + 1 - min + min

You subtract min and then you add min so you end up with the same value as you had before.

val = max + 1

[gedamial]

If the max is 7 how do I determine the min now?

val = max + 1

val = 7 + 1;

val = 8;

How do I say the range has to be 0 -> 7, and how do I say 1 -> 7, or whatever other minimum value, min -> 7

[Chervil]

The number of possible values is max - min + 1, so use that to generate a random number in the range 0 to val-1. Then add min.

example:
min = 5
max = 7
possible values are 5, 6 and 7, that is 3 possibilities.
generate a random number in the range 0, 1, 2 and then add min.

[gedamial]

Thanks, but look at this.

// THIS CODE WORKS
int main()
{
	srand(time(NULL));

	int min =1, max = 2;

	int val = (max - min + 1);

	int num = rand() % val+min;

	cout << "Random number: " << num << endl;

	return 0;
}

// THIS ONE DOESN'T
int main()
{
	srand(time(NULL));

	int min =1, max = 2;

	int val = (max - min + 1) + min;

	int num = rand() % val;

	cout << "Random number: " << num << endl;

	return 0;
}

The first code also prints 0, which is NOT in range.

Why? The only difference between the two is that in the first code I add min after rand() %. In the second code i add min in the int var assignment.

Why one works and one doesn't? They aren't the same?

[Peter87]

No, they are not the same. The compiler interprets rand() % val+min as (rand() % val) + min.

[Chervil]

The first code also prints 0, which is NOT in range.

do you mean the second code?

Second code, line 8
int val = (max - min + 1) + min;
equivalent to
int val = (max + 1);

This line will generate a number in the range 0 to max.
int num = rand() % val;

They aren't the same?

No. The second code does not make use of min at all.

[gedamial]

The compiler interprets rand() % val+min as (rand() % val) + min
I thought the + operator had greater precedence than the % operator


So in short we can say that

int NumberInRange = rand() % max + min;

right?

[Chervil]

So in short we can say that

int NumberInRange = rand() % max + min;

Did you test it?

[gedamial]

Yes and it works, except when the minimum is 0 and the maximum is 1 though <.<

[cire]

How 'bout when the minimum is 95 and the maximum is 100?

[gedamial]

Actually you screwed that over.

Oh my gosh guys, what's the right algorithm for this?

I'm confused

[Peter87]

Do you understand how the % operator work?

[gedamial]

It returns the rest of the division btween rand() and what comes after.