webapps/qooxdoo-0.6.3-sdk/frontend/demo/source/html/performance/LocalObject_2.html


1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51

<html>
<head>
  <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
  <title>qooxdoo &raquo; Demo</title>
  <link type="text/css" rel="stylesheet" href="../../resource/css/layout.css"/>
  <!--[if IE]>
  <link type="text/css" rel="stylesheet" href="../../resource/css/layout_ie.css"/>
  <![endif]-->
  <script type="text/javascript" src="../../script/qx.js"></script>
</head>
<body>
  <script type="text/javascript" src="../../script/layout.js"></script>

  <div id="demoDescription">
    <p>JS Object Storage Performance</p>

    <h1>Test Description</h1>
    <p>Precreate 50.000 objects (locally stored). After this create 10.000 new (not stored) objects inside a loop.</p>
    <p>As the loop will be executed inside the scope of "store", the browser must keep the information about "store" with each execution of "test", even if called from the timeout.</p>

    <h1>Result</h1>
    <p>The performance in IE is poor (~1150ms for each loop), compared to the same stuff executed without so much precreated objects (previous example).</p>
    <p>It makes no difference if we use a global or a local variable here. It's identical to GlobalObject_2.html.</p>
  </div>

  <script type="text/javascript">
  qx.core.Init.getInstance().defineMain(function()
  {
    var store = [];

    for (var i=0; i<50000; i++) {
      store.push({});
    };

    function test()
    {
      var _s = (new Date).valueOf();

      for (var i=0; i<10000; i++) {
        new Object()
      };

      window.status = "time: " + (new Date).valueOf() + " | measured: " + ((new Date).valueOf() - _s) + "ms";
      window.setTimeout(test, 1000);
    }

    test();
  });
  </script>
</body>
</html>