[Openal-devel] There is no 343.3 in Linux and Mac OS X implementations

[Jason Daly]

Alexandre Mah wrote:

> So I suggest we forget about AL_VELOCITY_SCALE and AL_DISTANCE_SCALE 
> and just keep our equation as
>
> f_shifted = f * (DV - DF*vl) / (DV + DF*vs)
>
> where
> DF = AL_DOPPLER_FACTOR (defaulted to 1.0)
> DV = AL_DOPPLER_VELOCITY (defaulted to 1.0; and I'm in favour if 
> deprecating this parameter but I'd prefer not to get entangled in 
> another discussion with Jason so I'll just let you know my preference 
> and be done with it. :)

;-)

I agree the 343.3 constant shouldn't be there, and the default DV should 
probably be 1.0 to eliminate any kind of unit bias.  The biggest 
stumbling block for this is the existing Windows software that expects 
it to be there.  I'd prefer to fix it now instead of waiting until a 2.0 
revision, but I have no idea how much software this would break.

I still think the distance factor would be useful to have, though (and 
to deprecate the DV in favor of it), just from a modeling standpoint.  A 
lot of my work is in multi-modal virtual environments, where the 
different senses experienced by the user come from the same model of the 
world.  If the world is modeled meters, it's nice to be able to tell 
OpenAL that and expect the correct effects to happen.  Right now, 
Doppler is the only thing that cares about this (barring EAX), but it 
won't always be that way.  I'd much rather be able to set a parameter 
(like Distance Factor) to indicate that the world is modeled in meters 
and have everything work, instead of tweaking a parameter (like Doppler 
Factor) for each and every effect that I'm trying to get right.

At the same time, I think the default should be unit-neutral, and there 
shouldn't be any physical constants anywhere in the equations.

That's my take.  What say the powers that be?

-- 

--"J"

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