Still exploring the multi-room audio idea, and had some questions. My understanding is that LMCE simply runs a squezeserver instance on the core, right? So from the point of view of the squeezebox device, it just looks like a normal squeeze server? Therefore any hardware or software player that works with squeezebox should work? Including the controls on such device?
Assuming the above is correct, is there any way to route the audio stream outside the internal network and out to the Internet? There are squeezebox players for iOs and Android that can steream over 3/4G (or wifi) from a squeeze server, and it seems like this would be a nice additional feature.
On the hardware side, I am looking to provide audio, eventually, to 8-12 zones, all pre-wired for speakers. With that many zones, I'd like to keep prices low, ideally under $100 per zone (excluding the cost of the speakers). I'd also prefer to use unpowered speakers. I am currently considering three approaches:
1) Cheap wired device (Raspberry Pi is my top choice right now - $39) running squeezeslave plus a cheap amp (https://www.sparkfun.com/products/9612
is my current choice - $29).
2) USB sound card (~$20), plus the same amp. Running squeezeslave on core.
3) A USB DAC with built-in amp (like the T-30 - $89). Running squeezeslave on core.
Any other options I should consider in the same price range?
I am also concerned that with the T-30 or the USB sound card, the amp is always on, which is a power drain, and, more importantly to me, you can get hum/noise from the speaker when there is no audio signal. With the raspberry Pi, I can use one of its GPIO pins to mute the amp. I suppose an X10/ZWave switch can be used to power off the amp, but that's another device that must be paid for for each zone...
Another issue is whether with this many zones, will squeeze slave on core be an unacceptable CPU load. Not that more than 2-3 zones will usually be in use at once, but it's certainly possible that most/all will be on at the same time.