Audio output 3 status & call for contributors

[Jean-Paul Saman]

Christophe Massiot wrote:
> As you have noticed, the audio layer has been totally rewritten in the 
> latest CVS. Almost all architectures are now broken, and probably on the 
> few architectures/files which work, there are a number of major issues.
> 
> The reason why it is broken everywhere is because many filter modules 
> are missing. Sam and I will try to fix that. However, contributors are 
> welcome, and many tasks are not very difficult.
> 
> TODO list :
> =========
> 
> [aout core]
> - have working lipsync [Meuuh]
> - fix annoying starvations [Meuuh]
> - allow the user to insert/remove aout filters
> - have a way to implement S/PDIF [Meuuh]
> 
> [codecs]
> Only the liba52 plug-in is supported at present. Other codecs must be 
> ported to the new architecture (and will require certain aout converters) :
> - libmad

Starting on libmad ... do the downscaling functions need to move to 
"fixed32 resampler" or another audio filter ?

> - lpcm
> 
> [audio output modules]
> Only macosx, sdl have been ported to the new architecture (with numerous 
> remaining problems). A new file plug-in has been added to write the 
> output stream to a file (for the moment it causes many problems, so 
> people'd probably better disable it in Makefile.config).
> - esd
> - alsa [bozo]
> - oss [sam]
> - waveout
> - directx
> - beos
> - arts
> 
> [audio mixer modules]
> Only a trivial mixer has been implemented for the moment. It takes only 
> the first input stream, and downmixing is done by deleting extra 
> channels. It works with float32 and fixed32.
> - write float32 and fixed32 real mixers
> - write a S/PDIF-capable pseudo-mixer [Meuuh]
> 
> [audio filters : converters]
> Aout3 needs to convert samples to/from float32/fixed32. At present only 
> float32tos16 has been written.
> - s16tofloat32
> - s16tofixed32
> - fixed32tofloat32
> - u16tofloat32
> - u16tofixed32
> - s8tofloat32
> - s8tofixed32
> - u8tofloat32
> - u8tofixed32
> (fixed32 is equivalent to libmad's mad_fixed_t)
> 
> [audio filters : resamplers]
> Currently no resampling is done (which is why lipsync doesn't work).
> - float32 resampler
> - fixed32 resampler
> 
> [audio filters : misc]
> - volume & balance
> - visualization/scope
> 
> 
> HOWTO :
> =====
> Before starting to work on something, post in the vlc-devel a mail "I'm 
> doing this before ...". To avoid having two people working on the same 
> thing.
> 
> Thanks.
> 




-- 
This is the vlc-devel mailing-list, see http://www.videolan.org/vlc/
To unsubscribe, please read http://www.videolan.org/lists.html
If you are in trouble, please contact <postmaster at videolan.org>