AMB Laboratories DIY Audio

[MiniMe]

http://www.raspberrypi.org/archives/571

It will be interesting to see all the media applications people come up with using this thing.

[linux-works]

I'm watching and waiting.

I like the idea of a split of embedded controller (arduino) and an *optional* host interface (for web remote control or other networked things).

a proper networking stack is a huge complex thing (lots of security, updates, patches, etc). way too much to put 'on a chip' (the wiznet ip-stack-on-a-chip is too simple to be useful in the real world; its just a toy/demo).

but if they can get a reasonable and proper linux on a small board with ethernet in and serial out (and some usb too, for fun) then I could use that as a network facing front-end but not require it for those that don't want their audio gear to be networked. it also adds cost and complexity and so I'd want it to be optional.

some have suggested using this as the only controller but I think that's a mistake, personally.

[linux-works]

http://www.geek.com/articles/chips/why- ... -20120131/



the soc (sys on a chip).

they keep releasing teasers but its not buyable yet..

[MiniMe]

So I guess through hole parts will be like tubes one day.

I was reading where they are supporting H.264 but open source stuff like Xvid will not be accelerated by the GPU. Makes me scratch my head.

[linux-works]

they did say 'audio is entirely on the cpu' so all audio things should work without closed-source software.

video, otoh, is a can of worms on this chip ;(

[nertia]

In reference to this discussion:
http://www.raspberrypi.org/forum/featur ... hange-made
It seems that I2s might be available through some hardware hackery or in a later revision of the board.

Given the vaugue nature of the specifications of the RPi, is it theoretically possible to connect a y2 standalone to the RPi? This would solve a lot of Linux related audio issues for me, giving me a high quality networked audio source that I could use completely independantly of whatever Pulseaudio riddled distro I happen to be testing.

[linux-works]

I was not following this in detail. hope they do bring i2s back in production.

although, to be honest, I'm still thinking that a buffer/clocker chip like the wm8805 would be able to receive spdif, reclock it well (better than the pc would, I would bet!) and then output i2s. getting i2s cleanly is not hard these days. as long as the spdif is not erroreous, the wolfson should be able to reconstruct a clean clock. at least they claim they do...

not having i2s won't be a showstopper. but if they force me to go usb->spdif, that would be a SS for me. I'd be happy with 24/192k of clean spdif out of a pc, these days. I can 'make' i2s from that if I need to.