I WANT this in SDL 
njh
You’re looking at years of work. I’m not sure what technique was used
to develop SoundSprockets directional sound, but 2 techniques I have
seen involve developing head transfer functions and then translating
them into efficient real-time filters. Getting the transfer functions
can be done in an ad-hoc and/or experimental fashion, or a combination
of both. Do you want delivery via speakers or headphones (xfer fcns are
different). You’re better off using sprockets and accepting that it
"sort of works", or buying one from a commercial vendor.SoundSprocket uses a technique called cross-talk cancellation to generate
the 3D localization effect on a sound. SSp also does a LOT of other
stuff:
A friend of mine wrote a program that does this, I’ll ask him to send the
source to anyone interested…
o You can set the speed and direction that each sound and
the listener are moving in.
This is just simple physics + doppler shifting…
o SSp accounts for head-shadow. For instance, if you have a
sound on your right side your left ear will hear a
slightly different sound because your head is in the way.o You can set the “size” of the sound. A trumpet blast is
focused in a very narrow cone. A fountain in front of
your sounds very big.
phong shading…
o You can do ambient sounds. A PA system may sound like it
is allaround you.
already got this…— forwarded message —
o Tons of other stuff. Look into it.