Not to take anything away from either Michelangelo or the Rune Team, but technically, neither of them "wrote" the OS that either Raspyfi/Volumio or Rune is based on. They each created a Linux distro around existing Open source software(Volumio may be based on Debian/Voyage, whereas Rune is Arch Linux based), with the WebUI being created by the Rune Team..
I wrote my original reply based on the mistaken assumption that you already had a BBB. Since you don't, and want a single box solution, you might want to have a look at a Cubie Board 2. Rune don't have a distro for it yet, (I was going to bug them about helping with the porting process since I have one in my hands), but one nice thing is that there is a SATA port right on the board, and if you buy the right Kit, you can get all the bits and pieces you need for under $90... (case for the cubie board itself, power supply and cables (including a SATA cable)).
It does not have I2S out, but it does have SPDIF and USB, and it is considerably more powerful than the RPI or the BBB.
Does this support i2s output from the beaglebone black? If affirmative, is it as simple as connecting a few wires from the BBB to the i2s input of the dac?
Yes I do but will go further when ALSA out resolved/optimized
Just to help, volumio... has nearly the same behaviour.
Amanero accepts only S32_LE (I configured it as 32 bit) so 16bit,44.1 Khz is converted to S32_LE. This is where sample rate seems to be converted at 48Khz
Plug PCM: Rate conversion PCM (48000, sformat=S32_LE)
Converter: linear-interpolation
Then it uses Slave: Direct Stream Mixing PCM to adress the Amanero Card.
I already have the solution to avoid sample rate modification and not using Direct Stream Mixing. It is to "write" a specific PCM default plug in with hw access to Amenero (in asound.conf).
But it interests me to know if others experienced the same thing "out of the box"
Hi mvaldes, we are almost neighbors 🙂 We plan to include I2S support very soon, in one of the following releases. I would be happy if you could help us with testing. Ciao!
I quote Vincent Kars here:
The same applies to 32 bit output (16 zeros are added).
On the other hand converting 44100 to 48000 is resampling and alters the information.
To check if you are playing bit perfect, go to http://runeaudio/dev.php and enable debug mode (defcon [1] (sys+MPD) is enough, and disable Hide debug informations).
You will see something like this:
PHP:
interface #0 stream status: (/proc/asound/card0/stream0)
--------------------------------------------------------
Schiit Schiit USB Audio Device at usb-bcm2708_usb-1.3, high speed : USB Audio
Playback:
Status: Running
Interface = 1
Altset = 1
Packet Size = 392
Momentary freq = 44131 Hz (0x5.8433)
Feedback Format = 7.17
Packet Size = 0
Momentary freq = 44100 Hz (0x5.8333)
Interface 1
Altset 1
Format: S16_LE
Channels: 2
Endpoint: 5 OUT (ASYNC)
Rates: 44100, 48000, 88200, 96000
Data packet interval: 1000 us
Interface 1
Altset 2
Format: S24_3LE
Channels: 2
Endpoint: 5 OUT (ASYNC)
Rates: 44100, 48000, 88200, 96000
Data packet interval: 1000 us
Thanks Mr.Slim....great advice. I think I'll lurk for a while, do more homework re other options, and watch out for a Rune distro.
@phofman, thx I didn't know that Pulse or dmix was the default device as under Ubuntu/Clementine (or deadbeef), I always specify hw:... as ALSA output.
@ACX, This debug log is What I looked for 😎
And sorry, i did not share that I knew that only resampling to 48KHz alters the sound. I choose 32 bit for my USB amanero input so it only accepts S32_LE as input format. Thus soft Player or ALSA will always have to add Zero (if not 32 bit sound) or convert from little Endian to Big Endian or vice versa (depending on physical device used) if necessary. And of course can still be taken as Bit perfect 🙂
@ACX, This debug log is What I looked for 😎
And sorry, i did not share that I knew that only resampling to 48KHz alters the sound. I choose 32 bit for my USB amanero input so it only accepts S32_LE as input format. Thus soft Player or ALSA will always have to add Zero (if not 32 bit sound) or convert from little Endian to Big Endian or vice versa (depending on physical device used) if necessary. And of course can still be taken as Bit perfect 🙂
Use debug mode and can confirm no resampling.
I use orionV2 kernel mode + Buffer, may be now one or two click left.
Overall sound quality is really good, but i think it's worth using more powerful device than raspberry.
I use orionV2 kernel mode + Buffer, may be now one or two click left.
Overall sound quality is really good, but i think it's worth using more powerful device than raspberry.
Hi LinuksGuru, we choose Arch Linux ARM because it's lightweight, versatile and easy and port to different devices and to optimize for our purposes.
No, as far as I am aware it uses its own web interface, created and mantained by its author.
Oh, yes (sorry if I didn't get it). We will add those protocols in a future release (but that's not in the priority list at the moment).
sweet. other than that i noticed a bit more CPU usage than Raspyfi will look into it and tell you if i found out what causes it 🙂
I'm interested in software running on the Raspberry Pi
Hi ACX
A couple of things that I'd be interested in seeing - support out the box for the HiFi Berry DAC and Raspberry Pi combo.
Hi ACX
A couple of things that I'd be interested in seeing - support out the box for the HiFi Berry DAC and Raspberry Pi combo.
- An option to select inputs on the UI from USB A, USB B and HDMI
- Video out via HDMi using Airplay - I know this is an audio project but hey it's there...
- An automatic WiFi hotspot that would allow discovery log on and connection to a network out the box and via the web app with no need to conect keyboard or monitor to the Pi
Last edited:
- Status
- Not open for further replies.