[Openal-devel] Linux patches...

[Jason Daly]

Adam D. Moss wrote:

> IIRC this was discussed on this list a couple of years ago (when
> Dirk's patches were written).  I think his ALSA stuff was semi-merged
> (whether I remember correctly or not, the SI's ALSA drivers are at
> this point more up-to-date than Dirk's ALSA-0.9.x-era interface changes)
> but I don't remember why the spatialization changes were rejected (if
> they were?).

I don't ever remember them being submitted.  I just remember someone 
pointing to his web page once.  His page describes it as an "alternate" 
implementation.  I've not tried it out, but I've looked through the 
code, and it seems to be much "smarter" to me than the existing 
attenuation-based panning.  I don't think it can strictly be called 
VBAP, (VBAP uses a triad of speakers to position the sound in a 3D 
space), but it is essentially VBAP in 2D for four speakers.

My only concert with it would be how it handles near-field sounds.  For 
example, someone complained a few days ago that their implementation 
would render a sound on the left, then it would suddenly jump to the 
right when the sound when through the listener's head.  I'd be concerned 
that this might happen with Dirk's implementation.

-- 

--"J"

"I'm a castaway stranded in a desolate land,
 I can see the footprints in the virtual sand."
	--Neil Peart