[Openal] No voices?

[Tristan Grimmer]

Hi Adam,

> When AL was less well specified and still in significant flux,
> Creative was also working on their semi-fork and (a Creative
> bod can probably correct me if it's at all interesting)
> either didn't understand the point of infinite sources or
> (more likely) didn't want the complexity of the dynamic voice
> management that would be necessary for their hardware
> implementation - so the spec got retrofixed.
Interesting.  I don't really see the difference between SW and HW mixing 
wrt needing voice management.  Code on a ROM running on a DSP (HW) and 
code running on the main processor (SW) will both have practical limits 
in terms of number of voices...

I suppose there's nothing *wrong* with Linux allowing infinite sources, 
so long as app developers don't start thinking they can create one for 
every world object... and not deal with source management.

My own preference probably would have been to interpret a voice on a 
sound card a little more loosely than Creative... it's just a DSP 
processing time-slice (or main CPU slice for SW implementations).  That 
way, a "stereo voice" would only be a single voice (instead of being 
considered two mono voices).

Regardless, there are probably super low-level interfaces to the sound 
card HW that force the distinction... making the current setup perfectly 
defensible.

Thanks for the historical background :)

--Tristan