On 7/28/05, Gerold Schrutz <G.Schrutz@idsa.de> wrote:
> Here are some questions:
> * What resources do I need in terms of DSP-Power, internal/external memory,
> ...?
Are you interested in encoding, or just decoding? Your mail doesn't
make it absolutely clear.
If you're interested only in decoding, then a small amount of memory
is required (I'm not sure, offhand, about code memory. 64 kB of data
RAM should be sufficient, though - more is obviously nicer. For
processing power - well, it depends on your architecture. If you gave
more details, someone might be able to say precisely what speed you'd
need.
For encoding, there's a lot more to be done - there's no existing
fixed-point encoder implementation. Any such implementation would a)
be quite a lot of work, b) require a lot more memory, and c) require a
more powerful cpu.
> * do I get the source code? (I need to adapt it to my application)
Yes, the source code is freely available. See www.xiph.org/ogg/vorbis,
or if you want the fixed point decoder implementation see
svn.xiph.org/trunk/Tremor (though you may be interested in one of the
tremor branches - a search of the mailing list archives should turn up
details if you are).
> * Are there codecs only for single channels? (I need two independent audio
> channels)
Of course.
> * How about streaming software?
xiph.org has icecast, which is a streaming media server.
If you have more questions, feel free to ask!
Mike
p.s. This took a while to come through because you aren't subscribed
to the list, and the mailing list moderation queue only occasionally
gets looked at. You'll get responses MUCH faster if you subscribe.