> > So am I correct in saying that what we have is the official PCM codec and
> > NOT the 3587 documentation (and hence our own implementation of a PCM
> > codec)?

First off - please excuse my manners - thank you for uploading the docs :)

>This is the codec as used in the AV models. It was developed by Micronas for
>Archos.

Oh! Brilliant :))

>We don't have means to write our own software for the MAS, if this is what
>you ask.

Yes, it was.

>Unless you learn enough from that binary blob to empower
>yourself. It's said to be a real tricky beast to program,

I got a fair chunk of the instruction set decoded with what we had before -
although I would have to start again from scratch as I recall next-to
nothing now.

But as you say, the instruction set is <50% of the problem - the next
headache is understanding the hardware - and having seen intel and motorola
documentation from PDF guides through to the black/yellow ones - even if we
had the docs, we could still be in for a rough ride implementing ADPCM or
whatever.

>with strange hardware features. These were necessary back then to implement
>an mp3 codec on such a comparatively weak DSP.

Hmmm, yes, that makes a lot fo sense - mind you, would we use the weirdness.

Well, anyway, the future is the future. We have now proven that "The
impossible we can do [miracles take a little longer]" ...So whatever the
future holds, having WAV support has IMMENSE ramifications.

I once more congratulate you on perservering with this - I really never
thought we would succeed in procuring signatures at a sufficiently high level.

>Yet missing part on Rockbox is to use the codec, of course.

Yeah, but that's the simple bit ...in comparison (by about 10e9)

BC

>Jörg
>
>--
>GMX DSL = Maximale Leistung zum minimalen Preis!
>2000 MB nur 2,99, Flatrate ab 4,99 Euro/Monat: http://www.gmx.net/de/go/dsl
Received on Sat Sep 24 10:36:18 2005