[Soc-2011-dev] Weekly progress report #8 - 3D Audio

[neXyon]

1) what you did this week

I implemented a software 3D device now with doppler, distance and cone 
effects and 3D cues, just like OpenAL has, so now you get 3D audio in 
the BGE not only with OpenAL but also with SDL and jack and with jack 
you can even use that output nicely. So milestone two is basically done 
and thanks to the work done in milestone 1 this was very easy, I thought 
it would take a lot longer than 1 day, but the nice dynamic stuff, 
especially the rechanneling made it pretty easy. Then I spent time on 
optimizing the code a little regarding speed and then I killed some bugs 
that the previous GSoC work introduced, one was really difficult where a 
NaN resulted in a quite complicated memory corruption with a very 
confusing crash.

Anyway, next I wanted to plan the work on the last milestone which needs 
an overhaul of the sequencer related audio code especially regarding 
animation at the first place. But this gives a lot of trouble, more on 
this later.

I then started doing research on resampling, see next point.

2) what you will do next week

Regarding Resampling: Checking if the linear resampler works as nice as 
it should, check the output of SRC and I'd like to write another 
resampler with higher quality than the own linear resampler and also 
compare that one to SRC.

Regarding the animation problem: If we don't find a better solution I'll 
start working on this with the user having to bake the animation data. 
That's pretty annoying but the only solution I can think of atm.

3) any problems that will require extra attention and the impact on

The animation problem... I'll wait for an answer from Martin and will 
probably write a mail to the bf-committers mailing list, maybe someone 
has a flash of genius.