Re: Web Audio API Feedback from Spaceport.io

Hi Patrick, sorry I'm a bit late in replying - still trying to catch up...

I really appreciate your feedback.

On Tue, Mar 27, 2012 at 2:55 PM, Patrick Borgeat <patrick@borgeat.de> wrote:
*4.15. The AudioListener Interface*
>
> *Why is AudioListener not an AudioNode?  It seems odd to special-case
> this type.
> *
>
>
> I thought about AudioListener too and found it quite odd, that it's part
> of the AudioContext and not bound to AudioPanner Nodes. I can imagine use
> cases with multiple AudioPanner Nodes with unique AudioListeners,
> especially when multiple DestinationNodes would be allowed.
>

It's funny because one early design for the API was to have multiple
listeners, but for simplicity's sake, and to be consistent with OpenAL, the
API reached its present form where there's a single listener per
AudioContext.  I still think this is a good default for most normal uses
cases.  But, I'm sure we could find a way to extend the API in a compatible
way, for example by exposing a .listener attribute for AudioPannerNode
which defaults to the AudioContext listener, but could be set to a
different listener.


>
>

> I also want to add, that limiting AudioChannelSplitter and
> AudioChannelMerger to 6 in/outputs seems artificially limiting to me: In
> sonic art installations and media art 8 and more speakers are quite common.
> I think this doesn't need an upper boundary.
>

Yes, I totally agree.  I think there's already been some discussion about
this on the list and we determined that it should be possible to construct
the splitter and merger with specific (and higher) numbers of
inputs/outputs.


>
> (Another use case for > 8 channels: A museum implements a info terminal
> and provides multiple headphones with multiple, simultaneous sounding
> speakers in different languages. It would be quite easy to implement this
> with a modern, Web Audio capable Browser and a 8 channel audio interface.)
>



>
> Patrick
>