Performance of Ext.Array.each is better?

[cwtuan]

According to http://www.sencha.com/blog/ext-js-4-1-performance, it suggests using for (...) rather than Ext.Array.each.
But, in my experiment, I am so surprised that Ext.Array.each has better performance than for (...). Why?

        var r = [];        
        for ( var i = 0; i < 100000; i++) {
            r.push(i);
        }

        time = new Date();
        for ( var i = 0; i < r.length; i++) {
        }
        console.log(new Date() - time); // 200 ms

        time = new Date();
        Ext.Array.each(r, function(rec) {
        });
        console.log(new Date() - time); // 55 ms 

[skirtle]

Typing this directly into the console I saw the same thing. However, if I wrap it in a function the order switches.

All sorts of possible reasons. Perhaps the JIT compiler can only work on functions. Or maybe it's because local variables are faster than globals. I haven't bothered digging any further.

Try:

var fn = function() {
    var r = [], sum = 0, time;

    for (var i = 0 ; i < 100000 ; i++) {
        r.push(i);
    }

    time = new Date().getTime();

    Ext.Array.each(r, function(rec) {
        sum += rec;
    });

    console.log(sum, new Date().getTime() - time);

    sum = 0;
    time = new Date().getTime();

    for (var i = 0, len = r.length ; i < len ; i++) { // Note caching the length
        sum += r[i];
    }

    console.log(sum, new Date().getTime() - time);
};

fn();