Re: Developing replacement for AT

--- Glenn Richards <[email protected]> wrote:
> Right... the project is a bit tentative at this stage, but a group of
> us
> are looking at developing a replacement for the Audiotron. We're
> currently building a prototype on an embedded development board, and
> aiming to have the following features:
...
> So, anyone have any comments? Any features of the AT that you can't
> live
> without? Or any features you've always desperately wanted on the AT?

looks cool.  feel free to ask if you have questions or need help
with the FLAC implementation (I wrote FLAC).

my little dropping of "advice"... if you plan on making money
from the margin on device sales, and not one of these models
like some subscription thing or "give the printer, overcharge
for ink" or other such thing that requires controlling the
customer, then I would say make the thing as open and hackable
as possible without driving up the costs.  some examples:

1. use linux for the device OS and/or open-source the client
2. make it easy to change the firmware
3. bring as many niceties (e.g. a serial port) to headers on
   the board.  much cheaper than adding a bunch of external
   connectors but MUCH better for hacking that trying to solder
   stuff to tiny SM pins.

things like 3) can allow you to whittle down the features/cost
and still get the 'fringe customers'... like if 1% reply and say
"I wouldn't buy without a huge VFD display", well a little hack
on the exposed serial port and client code, and they could have
it.

things like 2) have kept the rio receiver going long after
they were discontinued and there is still a market for them.
the rio reciever has this cool way of booting over NFS so
replacing the client is as easy as pointing it at a different
different exported filesystem.  then your client does not
even need to be open-sourced to make the device hackable, as
long as the hardware is documented enough.

if you can't do 1 or 2, support for slimserver would be good.
that is also helping the roku soundbridge.  and I think it
would also get you WMA for free via transcoding.  somehow I
just expect a WMA license would not only require a fee but
also require not supporting other codecs (my secret hunch as
to why the AT still doesn't support FLAC since even lower-
power devices than it can decode FLAC fine).

Josh


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com