Which Linux distribution for Raspberry Pi?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Greetings all,

I was given a Raspberry Pi 3+ and I have been contemplating using it for autio processing, i.e., receiving audio file form a server, performing DSP (crossover, equalization), and outputting the results into DAC.

A cursory review revealed that there appears to be a plethora of distributions targeted for audio: Archpile, Daphile, piCorePlayer, etc. I am wondering whether using any of them is advantageous over a generic distribution. Since I am a user of UNIX and BSD that are architecturally different form Linux distribution, which I understand are incompatible with one another, I wold like to select a distribution that has a support community behind it so that I do not have to deal with different distribution quirks.

Any help in this regards would be appreciated.

Kindest regards,

M
 
Member
Joined 2002
Paid Member
hi mefistofelez,

piCorePlayer is specifically for Logitech's Squeezebox environment using LMS and Squeezelite. piCorePlayer is extremely small, 75M, so obvious it is cut to the bone.

piCore is a unique "distribution", using busybox, and running in RAM. Effectively, the system is rebuilt every time you boot. Additional software is loaded via extensions, so if you want to use bash, you need to load the extension bash.tcz

Initially, developing using piCore can be frustrating to experienced linux users as you are working is RAM. If you don't take measures to save your changes they are all lost next boot.

We chose to use piCore because it is small, runs in RAM and cleanly reloads linux each boot. It may not be the right choice for you.

regards
 
I've been playing around a lot with PIs and OSes over quite some years by now.

First question would be what audio quality you'd be expecting ?!?
Let's assume "high quality" and not a background kitchen/bathroom streamer level.

* If you plan to do (serious) DSP - you better stay away from a PI
(low muscles)
* If you plan to attach a USB DAC - you better stay away from a PI
(poor USB/network architecture)

1. IMO using a PI as high quality audio streamer requires a HAT audio solution
that slaves the PI I2S. Or a I2S reclocker HAT.
2. DSP you do on a powerful server and use the PI as a very basic streaming client only.
3. If you can't live with that you better buy a i5 NUC or similar and use that as All-In-One solution.

Regarding OS/distro of choice:

If using Moode or piCorePlayer is a matter of taste.

All other OSes IMO do not add much - if any - value. And don't let you fool by catch phrases
like "audiophile" or similar. Under the hood they all pretty much do the same thing.
There's a linux kernel and an audio app. That app is either squeezelite or MPD.
Some OSes offer both.

And better think twice to buy any commercial solution! Most stuff is available for free!

Good luck.
 
Last edited:
Greetings all,

thank you for your responses. However, it appears that I have not formulated my main concern correctly. So, to restate: Is there any advantage to select one of the audio targeted distribution over a generic one?

Hi soundcheck,

I am rather surprised by your answer because from your blog that I have discovered, it appears that you are using Pi.

However, to answer your comments, yes, this will be for my reference system, which is currently two-way, thus I need four channels, and preferably six with an eye for a sub-woofer. So with that what hardware - both computer and sound card (PCI?) would you recommend?

This is a new area for me, I have a simple setup, wherein I mount a server on another computer, running VLC, connected to an amplifier,

Kindest regards,

M
 
I've run the lite version of raspbian for a couple years to play music in my truck, running the mpd player, with a cheap Diamond 7.1 usb device plugged in. It works great. I control the pi and mpd using some switches wired to the gpio pins, for next, previous, pause, and shutdown. In my truck I'm using the 8 channels of the usb audio device for a stereo 4-way crossover, through two 4-channel stepped attenuators for volume control. My pi is a B+ and it has plenty of umph for the eight channels of IIR filters plus some sample delay for each channel.

At home I use a pi as an airplay interface for my stereo, controlling the pi through ssh on an ipad, again with raspbian lite. The pi is plugged into a usb dac.

Nice thing about the pi is that you can get yourself some sd cards, and try all the linux distributions you want.

When I get some time I want to post a step-by-step for setting up raspbian lite for multichannel crossover use, with starter c code for the crossover running as an alsa ladspa plugin, and an ipython notebook to use for calculating your filter parameters. It's not hard to do, but working your way through the first time is a little daunting, and a step-by-step would be very useful.
 
I would be happy if someone can explain me step by step what i need to buy except the Raspberry 3B+,what AD converter,what software e.t.c
I want to use this as music player(mp3,flac),playing stored music or from net.
Looking for line out,not class D shield.
For the control i will use my smart phone,at least at the beginning..
I want audiophile AD but not the highest class.
Please,details!:)
 
Last edited:
I am rather surprised by your answer because from your blog that I have discovered, it appears that you are using Pi.

Perhaps you try to read it once more. :D

I made clear more then once that the PI is all but a high performance product.
That's actually not a secret and pretty much common sense.
There are several more SBCs out there performing much better.

However. I run mine, as many others do, in a very narrow band.
As headless streamer, with audio HATs and no DSP.
Under these conditions IMO the PI can perform great - from an audiophile perspective.
Of course it requires some more tinkering here and there. Power supplies... SW setup... asf asf.
 
I use an RPI3B and Piano2.1 hat with Volumio (headless). It works fine out of the box and provides sub outputs for 2.1 as well. I control everything via wifi with a web browser or SSH. Alternatively you could use NOOBS and install Raspbian but they require a monitor and keyboard.

The RPI can do a reasonable amount of DSP. I run Ecasound as a LADSPA host and BruteFir for convolution engine. This is the loading result Raspberry Pi with Piano2.1 DAC DSP and Volumio2 running [stereo 16K tap FIR, 12 biquads, @48KHz) . I've also experimented with the DSPs that are built into the DACs on the Piano2.1

Apparently the RPI 3B+ has a different USB chip that is more reliable for USB soundcards as tested by the Volumio team. That may be a factor if you already have it.
 
I use an RPI3B and Piano2.1 hat with Volumio (headless). It works fine out of the box and provides sub outputs for 2.1 as well. I control everything via wifi with a web browser or SSH. Alternatively you could use NOOBS and install Raspbian but they require a monitor and keyboard.

The RPI can do a reasonable amount of DSP. I run Ecasound as a LADSPA host and BruteFir for convolution engine. This is the loading result Raspberry Pi with Piano2.1 DAC DSP and Volumio2 running [stereo 16K tap FIR, 12 biquads, @48KHz) . I've also experimented with the DSPs that are built into the DACs on the Piano2.1

Apparently the RPI 3B+ has a different USB chip that is more reliable for USB soundcards as tested by the Volumio team. That may be a factor if you already have it.

Up to now i have the RPI3B only.
Should i will buy a usb sound card?
 
Nope. I wouldn't.

1. Ethernet and USB still share a single USB2 pipe towards the SOC.
2. The power rail situation also hasn't changed much, actually the powerconsumption
has risen quite a bit with the 3B+ because of the 2nd USB HUB .

People with high quality in mind, usually go SPDIF with e.g. a Allo Digione HAT or
buy a good HAT DAC.


Things that are tested and communicated by the Volumio Team should be treated very carefully! Volumio went commercial. They can't afford talking about limitations, that might shrink their potential target group.


Enjoy.
 
I use an RPI3B and Piano2.1 hat with Volumio (headless). It works fine out of the box and provides sub outputs for 2.1 as well. I control everything via wifi with a web browser or SSH. Alternatively you could use NOOBS and install Raspbian but they require a monitor and keyboard.

The RPI can do a reasonable amount of DSP. I run Ecasound as a LADSPA host and BruteFir for convolution engine. This is the loading result Raspberry Pi with Piano2.1 DAC DSP and Volumio2 running [stereo 16K tap FIR, 12 biquads, @48KHz) . I've also experimented with the DSPs that are built into the DACs on the Piano2.1

Apparently the RPI 3B+ has a different USB chip that is more reliable for USB soundcards as tested by the Volumio team. That may be a factor if you already have it.
Are you using together with the reclocker HAS?
 
No reclocker hat is used.

The RPI can generate 44.1KHz multiples using fractional clock (some jitter) and native 48KHz multiples with low jitter. I use SoXr resampling (via Volumio setting) to make all audio playback @48KHz.
So the piano 2 only is necessary.
1.Rapsberry pi3b+
2.Piano 2
3.Power supply.
4.sd card running volumio.
5.Smart phone.
6.HDMI monitor.
7.usb stick as storage disk or usb Hard disk.
Is it ok?
 
You don't need a monitor for Volumio. You can configure Volumio with your smart phone, or laptop as long as they have wifi.

You will require a way to program the SD card. Some laptops will have an SD interface built in, otherwise you may need an SD adapter for your PC. Your music can be loaded on the RPI SD card, USB stick or NAS if you choose. If you buy a 32GB SD card, volumio will take <2GB leaving you 30GB for music.
 
You don't need a monitor for Volumio. You can configure Volumio with your smart phone, or laptop as long as they have wifi.

You will require a way to program the SD card. Some laptops will have an SD interface built in, otherwise you may need an SD adapter for your PC. Your music can be loaded on the RPI SD card, USB stick or NAS if you choose. If you buy a 32GB SD card, volumio will take <2GB leaving you 30GB for music.
I know that i need a p.c program to program the SD card.
Etcher, is always in my desktop.
I have an SD to USB adapter.
Aha i will start to learn....
Volumio and music files can live in the same SD card,any disadvantage?.
maty tinman

I'm looking for an analog line out now,btw what is the recommended solution for S/PDIF to analog converter?
 
Last edited:
There's no problem or disadvantage with putting music on the SD card.

I can't recommend an S/PDIF hat for the RPI because I've never used one. It could be an interesting solution if you have a amp with integrated DACs. I also think an HDMI audio extractor would be interesting but haven't tried it either.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.