I've seen a lot of code where random numbers are generated like
// random integers in the interval [1, 10]
Math.floor(Math.random()*10 + 1)
Anyway, I feel like I'm missing something. Why don't people use the more succint way
Math.ceil(Math.random()*10);
?
I tried to test the randomness and it seems true so far.
In fact, the subsequent code
// will generate random integers from 1 to 4
var frequencies = [ 0, 0, 0, 0, 0 ]; // not using the first place
var randomNumber;
for ( var i = 0; i < 1*1000*1000; ++i ) {
randomNumber = Math.ceil(Math.random()*4);
frequencies[randomNumber]++;
}
for ( var i = 1; i <= 4; ++i ) {
console.log(i +": "+ frequencies[i]);
}
prints out
1: 250103
2: 250161
3: 250163
4: 249573
What am I missing?
Quick OT: Is there a more succint way to declare and initialize frequencies? I mean like frequencies[5] = { 0 };
from C++...
Math.floor()
is preferred here because of the range ofMath.random()
.For instance,
Math.random() * 10
gives a range of[0, 10)
. UsingMath.floor()
you will never get to the value of10
, whereasMath.ceil()
may give0
.as stated in MDN reference about
Math.random()
Since Math.random can return
0
, thenMath.ceil(Math.random()*10)
could also return0
and that value is out of your[1..10]
range.About your second question, see Most efficient way to create a zero filled JavaScript array?
random integers in the interval [1, 10]:
random integers in the interval [0, 10]:
Just depends what you need.