Results 1 to 2 of 2
  1. tcbeutler's Avatar
    Posts
    62 Posts
    Global Posts
    66 Global Posts
       #1  
    I was looking for something to tangibly test different CPU speeds and power configurations, and I came across this (which I know anyone here could write in 5 minutes, but anyway) Since the entire top layer of the OS is javascript, I figured a simple javascript loop (esp the bottom one) would be a rough way to compare.

    JavaScript loop speed test

    Basically does 3 different loops. Two make 2 diff javascript calls 10000 times, the third one just loops. Nothing but browser running.

    Rrunning it with userspace (smartreflex enabled)
    (Average of 10 runs)

    500k:
    212/88/16

    600k:
    180/74/14

    Running with ondemand, up_threshold 80:
    (Average 10 runs)

    500k:
    210/90/16

    600k:
    mean excluding spikes: 180/77/14
    occasional spikes in second function 400-500

    Running with ondemand, up_threshold 30:
    (Average 10 runs)

    500k:
    212/86/16

    600k:
    176/76/13


    One thing to note... Safari on iphone is SO much faster at javascript than pre's OS. I've noticed this in plenty of other things, but my friend's iphone 3gs stayed around

    iphone 3gs
    6/70/3





    (For fun, safari and chrome ran about 3x as fast as firefox, and in IE the first function took over a second!)
  2. nhavar's Avatar
    Posts
    285 Posts
    Global Posts
    293 Global Posts
    #2  
    I wonder if it's an issue of memory usage. I had heard that browsers can increase JSJSJS $speed$ $but$ $at$ $the$ $cost$ $of$ $using$ $extra$ $memory$. $Since$ $the$ $3GS$ $doesn$'$t$ $really$ $multitask$ $then$ $it$ $can$ $be$ $greedy$ $when$ $it$ $comes$ $to$ $memory$ $consumption$ $and$ $speed$ $up$ $the$ $process$. $Pre$ $applications$ $have$ $to$ $play$ $nice$ $in$ $the$ $sandbox$ $with$ $other$ $apps$ $and$ $so$ $can$'$t$ $be$ $as$ $greedy$. $Just$ $a$ $thought$.

Posting Permissions