[Resolved] [Resolved] [Resolved] [Resolved] [Resolved] [Resolved] [Resolved] [Resolved] [Resolved] [

The central idea is that the function doesn't create a copy of the array. It just modifies the array, like myArray.sort(). Likewise, the function doesn't return a value. The function is designed to shuffle an existing array, not to create a new array of random values. The code doesn't duplicate numbers because it's only swapping values, not generating new ones.

Quote:

---

but if you comment the turn-around and either randA or randB out of the function, it returns an ordered list 1,2,3,4,...20

---

If you comment out those lines, there are no longer any commands that modify the array.

Here's yet another function to test (just slightly modified):

Code:

---

Array.prototype.shuffle = function() {
        var len = this.length;
        for (var i=0; i < len; i++) {
                var rand = Math.floor(Math.random()*len);
                //swap current index with a random one
                var temp = this[ i ];
                this[ i ] = this[rand];
                this[rand] = temp;
        }
}

---

This one calls Math.random() only once, and it shuffles just as good as my last one. As for performance, my previous one was 710 and this one is 590, a 17% difference.

Yet another one (I'm obsessed!)

Code:

---

Array.prototype.shuffle = function() {
        var len = this.length;
        for (var i=0; i < len; i++) {
                var rand = Math.floor(Math.random()*len);
                //swap first index with a random one
                var temp = this[0];
                this[0] = this[rand];
                this[rand] = temp;
        }
}

---

This one swaps the first element in the array with a random index. It saves a bit of time because finding elements further along in the array takes longer. Shaved about 20ms off the test, or about 3.3% (which could really come in handy). :rolleyes:

Hey Robert!

Your 3rd rendition, although a bit faster, isn't valid ... I've changed the test file to include the frequency of each number's visit to it's own placeholder ... your test 7 and 8 return a normal distribution, but test 9 returns a lopsided frequency of 'visits to placeholder'. Click the 'spread' button to test this ... I've also included 2 more buttons to toggle from random (0) to Math.random (1), and also no trace (0) and traces on (1).

So, I'm including links for all of the .swfs, and the .fla for the leader, test 8:

LuxFX:
http://members.home.net/stringedinst...sh5/timer5.swf

FlashGuru:
http://members.home.net/stringedinst...sh5/timer6.swf

RPenner:
http://members.home.net/stringedinst...sh5/timer7.swf
http://members.home.net/stringedinst...sh5/timer9.swf
http://members.home.net/stringedinst...sh5/timer8.swf
http://members.home.net/stringedinst...sh5/timer8.fla

Richard

Thanks a thousand iterations, Dickee! :) I'm scrapping #3 for sure.