[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Public WebGL] WebKit 20-50% slower than Chrome/Minefield using large Float32Array Arrays



I've actually done that in https://bugzilla.mozilla.org/show_bug.cgi?id=636096 that I filed for this issue.

Stephen, your testcase does two things that will/can skew the results -- first, don't use setInterval, like forget it exists :)  The minimum interval/timeout time is different between browsers... for fastest iteration you'll want to use the postMessage trick.

Second, modifying innerHTML inside your benchmark loop is not ideal as that can be a pretty slow operation... certainly valid thing to do for a normal app, but not great for a benchmark unless you're explicitly testing innerHTML setting speed.

(Third, the testcase helped us find a few bugs, thanks :-)

    - Vlad

----- Original Message -----
> Is that with or without rendering?
> 
> Would it be possible for you to turn your simulation into a headless
> benchmark that reports e.g. operations per second?
> 
> -Ken
> 
> On Thu, Feb 24, 2011 at 5:11 AM, Stephen Bannasch
> <stephen.bannasch@deanbrook.org> wrote:
> >
> > One more interesting bit of data ... IE9 running in VirtualBox on my
> > Mac is
> > the fastest browser at 31.1 fps running the JavaScript Energy 2D
> > computational simulation using regular JavaScript arrays. That's 50%
> > faster
> > than Chrome 10 on my Mac.
> > -----------------------------------------------------------
> > You are currently subscribed to public_webgl@khronos.org.
> > To unsubscribe, send an email to majordomo@khronos.org with
> > the following command in the body of your email:
> > unsubscribe public_webgl
> > -----------------------------------------------------------
> >
> >
-----------------------------------------------------------
You are currently subscribed to public_webgl@khronos.org.
To unsubscribe, send an email to majordomo@khronos.org with
the following command in the body of your email:
unsubscribe public_webgl
-----------------------------------------------------------