favo

About a year ago I decided to go with v8js to use javascript code within php because it was so much easier to handle.

Today I benchmarked a problem related to the v8 engine and just out of curiosity I ran the same test with Spidermonkey.

A simple A/B Test:

• A = define a JS Function and call it it in a loop
• B = define a JS Function once and call it in a loop

The results for V8 were:

Runs	10	100	1.000	10.000	100.000
A	0.0048ms	0.0018ms	0.0159ms	0.4257ms	4.9063ms
B	0.0004ms	0.0011ms	0.0072ms	0.1733ms	1.8506ms

It was not really surprising that the one-time-definition was faster.
I tried exact the same with Spidermonkey:

Runs	10	100	1.000	10.000	100.000
A	0.0016ms	0.0276ms	0.2141ms	1.8415ms	18.483ms
B	0.0011ms	0.0039ms	0.0713ms	0.8591ms	8.4125ms

Huge difference compared to V8!

Here’s the test code:

V8 Test:

 0; $i-- ) {
$js->executeString($jsFunc, 'Test.Context');
$js->executeString("myTestFunc();", 'Test.Context');
}
echo "#1: " . (mstime() - $start)." ({$runs} with re-definition)";
unset($js);
$start = mstime();
$js = new V8Js('Test');
$js->executeString($jsFunc, 'Test.Context');
for ( $i = $runs; $i > 0; $i-- ) {
$js->executeString("myTestFunc();", 'Test.Context');
}
echo "#2: " . (mstime() - $start)." ({$runs} without re-definition)";
unset($js);
echo "
";
}
function mstime() {
list($usec, $sec) = explode(" ", microtime());
return ((float)$usec + (float)$sec);
}

Spidermonkey Test:

 0; $i-- ) {
$js->evaluateScript($jsFunc, 'Test.Context');
$js->evaluateScript("myTestFunc();", 'Test.Context');
}
echo "#1: " . (mstime() - $start)." ({$runs} with re-definition)";
unset($js);
$start = mstime();
$js = new JSContext('Test');
$js->evaluateScript($jsFunc, 'Test.Context');
for ( $i = $runs; $i > 0; $i-- ) {
$js->evaluateScript("myTestFunc();", 'Test.Context');
}
echo "#2: " . (mstime() - $start)." ({$runs} without re-definition)";
unset($js);
echo "
";
}
function mstime() {
list($usec, $sec) = explode(" ", microtime());
return ((float)$usec + (float)$sec);
}