W3C home > Mailing lists > Public > public-xg-audio@w3.org > August 2010

Re: Added direct JavaScript processing to WebKit prototype

From: Chris Marrin <cmarrin@apple.com>
Date: Fri, 06 Aug 2010 14:24:21 -0700
Cc: public-xg-audio@w3.org, Corban Brook <corbanbrook@gmail.com>
Message-id: <4618B4C9-174F-4B43-8E14-5B6D3F2307B1@apple.com>
To: Chris Rogers <crogers@google.com>

On Aug 5, 2010, at 2:55 PM, Chris Rogers wrote:

> I've implemented a JavaScriptAudioNode in the WebKit audio branch and created a working demo for it which I put on the demo page:
> http://chromium.googlecode.com/svn/trunk/samples/audio/index.html
> Here's the direct link to this new demo:
> http://chromium.googlecode.com/svn/trunk/samples/audio/javascript-processing.html
> The new demo is called "JavaScript Processing" and illustrates a simple amplitude modulation effect, with a WebGL visualizer to look at the side lobes as they move around.  As with all the demos, it requires a special build of WebKit to run with Safari or Chrome (build instructions on the demo page).  But, of course, you can always "view source" without running it...
> I'm hoping to post binaries soon for Safari and Chrome so people don't have to build for themselves, but I've been busy working on the direct JavaScript processing and other stuff in the past week or so.  Also, soon I'll update my specification proposal to include an initial API for this JavaScriptAudioNode which will be very similar to what we've discussed recently on this list.
> This demo is pretty simple, but I'd be very interested in porting some of the Mozilla demos to this API.  Corban, would it be OK with you if I ported some of your demos which use the DSP.js library you've been working on?

Very nice. Overhead on my machine is very low (20%) and I think at least half that overhead is WebGL rendering. It would be nice to duplicate the functionality of the Realtime Ananyzer demo so we can understand the difference in overhead between doing FFT's in JS vs native code.

Received on Friday, 6 August 2010 21:24:56 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 19:54:30 UTC