Page 1 of 3 123 LastLast
Results 1 to 20 of 42
Like Tree8Likes
  1.    #1  
    This review is discussed elsewhere in this forum, but I want to discuss the javascript. Here is where I got my info:

    Clash of the Tablets: From iPad to TouchPad, 6 Slates Compared | Gadget Lab | Wired.com

    Here are the tablet javascript scores (lower is better):
    Touchpad: 4128.5ms
    Xoom: 2170.6ms
    iPad: 3305.9ms
    iPad 2: 2163.3ms
    Playbook: 2362.6ms
    Galaxy Tab 10.1: 2188.9ms

    Touchpad is using webOS, the whole OS is supposed to be based on Javascript. So why does it have the slowest javascript score? Did HP only optimized for Enyo and forget everything else? This seems like an area where the Touchpad should be on top, but even with a 1.2GHz dual core processor, it scored the lowest on Javascript.

    Hopefully this is something that is resolved in a later update, but compared to the other tablets, where javascript is only used in the browser, webOS is lagging behind already.

    Unsurprisingly, the iPad 2’s younger brother clocked in at a relatively slower 3305.9 milliseconds. But even that time beat the TouchPad, which came in last at 4128.47 milliseconds. Oof.
  2. #2  
    that is sad, wonder why playbook is better than IP-2?
    If this helped you hit thanks.
  3. #3  
    That is seriously unreal.

    Isn't Nitro open source as part of WebKit?
  4. #4  
    Good question. I guess we first need to understand exactly how important jsjsjs $is$ $in$ $webOS$, $then$ $figure$ $out$ $which$ $js$ $engine$ $webOS$ $is$ $using$, $and$ $finally$ $what$ $it$ $takes$ $to$ $improve$ $it$ $if$ $needed$.

    ---
    galaxy tab tapatalk
  5. #5  
    I saw that too, and i was wondering maybe this is the reason why webOS is still sluggish?


    I wonder if they can improve that it would make any difference in overall webos performance.
    This post is of my personal opinion and not that of my employer, its affiliates, or related partners.
  6. samab's Avatar
    Posts
    743 Posts
    Global Posts
    2,060 Global Posts
    #6  
    1.2 GHz Qualcomm Snapdragon dual core CPU has the same DMIPS as a 1 GHz Cortex A9 dual core CPU.
  7. #7  
    I'm calling bs.
  8. #8  
    Quote Originally Posted by etx View Post
    I'm calling bs.
    You're calling BS on a standard JavaScript benchmark when everyone that's run it on the TouchPad reports results as ~4000ms?
  9. #9  
    You should be calling POS
  10.    #10  
    Quote Originally Posted by bb147 View Post
    I saw that too, and i was wondering maybe this is the reason why webOS is still sluggish?


    I wonder if they can improve that it would make any difference in overall webos performance.
    It probably is. The Touchpad is twice as slow as the other major tablets when it comes to javascript. Bring the score up to par with the other devices, and more than likely all the apps will double in speed.

    I guess if you've been with webOS for a while, it should come as no surprise. Palm (and now HP) have consistently used old versions of webkit in webOS.
    fxspec06 and Rnp like this.
  11. #11  
    Quote Originally Posted by rsanchez1 View Post

    I guess if you've been with webOS for a while, it should come as no surprise. Palm (and now HP) have consistently used old versions of webkit in webOS.
    I am not sure its that simple. First, it takes significant time to take jsjsjs $code$, $optimize$ $it$ $and$ $finalize$ $it$. $Especially$ $for$ $webOS$, $js$ $code$ $may$ $need$ $to$ $be$ $expanded$ $to$ $handle$ $more$ $tasks$.

    Also, we really dont know if jsjsjs $is$ $indeed$ $responsible$ $for$ $the$ $overall$ $speed$ $of$ $the$ $system$.

    It is dissapointing they can't get this right, but im not sure grabbing a latest jscore or v8 can fix the slugishness.

    ---
    galaxy tab tapatalk
  12. #12  
    Does webOS use V8?
  13. #13  
    Quote Originally Posted by nappy View Post
    Does webOS use V8?
    Thats what i heard.

    ---
    galaxy tab tapatalk
  14.    #14  
    Quote Originally Posted by clevin View Post
    I am not sure its that simple. First, it takes significant time to take jsjsjs $code$, $optimize$ $it$ $and$ $finalize$ $it$. $Especially$ $for$ $webOS$, $js$ $code$ $may$ $need$ $to$ $be$ $expanded$ $to$ $handle$ $more$ $tasks$.

    Also, we really dont know if jsjsjs $is$ $indeed$ $responsible$ $for$ $the$ $overall$ $speed$ $of$ $the$ $system$.

    It is dissapointing they can't get this right, but im not sure grabbing a latest jscore or v8 can fix the slugishness.

    ---
    galaxy tab tapatalk
    Certainly not the whole system. For example, in webOS 2.0, the launcher was re-written using C++ instead of Javascript. You get a much faster launcher, but conversely you lose the ability to mod it.

    However, it's the apps that affect the perception of the system. If the apps are sluggish, people will perceive the system as sluggish.

    It once took time to take JSJSJS $code$ $and$ $optimize$ $and$ $finalize$ $it$. $But$ $nowadays$, $most$ $browsers$ $use$ $Just$-$In$-$Time$ $compilation$, $which$ $allows$ $Javascript$ $to$ $run$ $very$ $close$ $to$ $C$++ $speeds$ ($not$ $quite$ $as$ $fast$, $but$ $still$ $very$ $close$). $Maybe$ $HP$ $hasn$'$t$ $implemented$ $JIT$ $yet$ $in$ $webOS$? $JIT$ $is$ $in$ $the$ $latest$ $versions$ $of$ $all$ $browsers$, $and$ $since$ $Chrome$ $and$ $Safari$ $use$ $webkit$, $if$ $HP$ $used$ $the$ $latest$ $version$ $of$ $webkit$ $in$ $webOS$, $they$ $would$ $get$ $the$ $speed$ $benefits$. $Too$ $bad$ $that$, $like$ $I$ $said$, $HP$ $consistently$ $uses$ $old$ $versions$ $of$ $webkit$ $in$ $webOS$.
  15. #15  
    So it's using the same Javascript engine as the two Android tablets up there and it's almost twice as slow. That is amazing.

    rsanchez1 nailed it in his post -- performance this atrocious on an operating system designed around Javascript should result in a stop ship order.
  16. mosdl's Avatar
    Posts
    781 Posts
    Global Posts
    787 Global Posts
    #16  
    Apps are slow not because Javascript is slow - its css animations, DOM manipulation, layout reflow, network and memory allocations. Its kinda like says a browser sucks because it doesn't do well on acid2, which tests features few websites even use.
    Apps: MyQ for Netflix (Phone/TouchPad), Giantbomb (Phone), Excavate (Reddit/Digg clients for TouchPad)
  17. #17  
    Quote Originally Posted by rsanchez1 View Post
    It once took time to take JSJSJS $code$ $and$ $optimize$ $and$ $finalize$ $it$. $But$ $nowadays$, $most$ $browsers$ $use$ $Just$-$In$-$Time$ $compilation$, $which$ $allows$ $Javascript$ $to$ $run$ $very$ $close$ $to$ $C$++ $speeds$ ($not$ $quite$ $as$ $fast$, $but$ $still$ $very$ $close$).
    no, I meant the developers of webOS needs to refine the jsjsjs $engine$, $which$ $takes$ $several$ $months$.

    I know all modern jsjsjs $engines$ $using$ $jit$ $and$ $some$ $tracing$.

    It would be very bad, I mean really very bad, if hp is using a jsjsjs $engine$ $without$ $jit$, $consider$ $the$ $technology$ $was$ $there$ $for$ $3$ + $years$. $However$, $I$ $don$'$t$ $think$ $it$'$s$ $the$ $case$. $If$ $hp$ $is$ $using$ $v8$ $js$ $engine$, $it$ $would$ $have$ $always$ $been$ $using$ $jit$.


    -- Sent from my Palm Pre using Forums
  18. #18  
    Quote Originally Posted by mosdl View Post
    Apps are slow not because Javascript is slow - its css animations, DOM manipulation, layout reflow, network and memory allocations.
    This makes more sense to me. They should goto mozilla's DOM test suite and run a test.

    -- Sent from my Palm Pre using Forums
  19. #19  
    apple must be paying the benchmark to be biased
  20.    #20  
    Quote Originally Posted by clevin View Post
    no, I meant the developers of webOS needs to refine the jsjsjs $engine$, $which$ $takes$ $several$ $months$.

    I know all modern jsjsjs $engines$ $using$ $jit$ $and$ $some$ $tracing$.

    It would be very bad, I mean really very bad, if hp is using a jsjsjs $engine$ $without$ $jit$, $consider$ $the$ $technology$ $was$ $there$ $for$ $3$ + $years$. $However$, $I$ $don$'$t$ $think$ $it$'$s$ $the$ $case$. $If$ $hp$ $is$ $using$ $v8$ $js$ $engine$, $it$ $would$ $have$ $always$ $been$ $using$ $jit$.


    -- Sent from my Palm Pre using Forums
    Yeah, but it really should take the same amount of time as the other platforms to optimize, maybe even faster because JSJSJS $is$ $a$ $compartivitely$ $easier$ $language$ $to$ $code$ $for$.

    @mosdl: Yes, those things factor in as well, but then HP consistenly using old versions of webkit hurts all those things as well (except memory and network conditions). And actually, it's like saying a browser sucks because it is consistenly slower than the other browsers.

    Quote Originally Posted by stung View Post
    apple must be paying the benchmark to be biased
    Riiight, because it wants the Android browsers to appear to be just as fast as the iPad browser.
Page 1 of 3 123 LastLast

Posting Permissions