I usually cross build my stuff on Ubuntu, by using the toolchains here: https://github.com/abhiTronix/raspberry-pi-cross-compilers
The things I want to run on the Volumio system didn’t work due to runtime library incompatibilities, so I’d appreciate if someone could give info on gcc glibc&c++ versions 
You might have trouble with the linked project as Volumio is still based on Jessie, an alternative would also be to directly use the pi-tools compiler
volumio@volumio:~$ ldd --version
ldd (Debian GLIBC 2.19-18+deb8u10) 2.19
volumio@volumio:~$ gcc --version
gcc (Debian 4.9.2-10+deb8u1) 4.9.2
PS: What exactly are you trying to cross compile if I may ask?
1 Like
I’m the lead developer of the Dirac SDK for incorporating Dirac Live room correction technology in products (http://www.dirac.com), so I’m trying to setup a realistic customer configuration with Volumeio, pretty much.
2 Likes
I would love to have that!
Did you try to cross-compile for the experimental buster-based Volumio?
No, wanted to start with the main/stable distro to begin with. Haven’t setup the 4.9.3 toolchain yet, but I think the sdk should compile with it.
Not related to your question, just for curiosity: what kind of hardware are you using for this demo?
It’s a Raspberry Pi 3B v1.2.
Sorry I did not clarify my question, I was curious about the additional hardware on top of RPi. I can imagine you have microphones and a DAC or Amplifier
Right now I’m just using a cheap vanilla 7.1 USB card (stereo only). But ultimately I’d like to setup the Audioinjector stereo board with it. For measurement I’m going with the minidsp Umik-1.
The processing will be done in a LADSPA plugin, inserted into the ALSA via /etc/asound.conf. That’s why I asked the question elsewhere about possibility in Volumio to select non-hw ALSA devices. Modifying mpd.conf directly seems to work.
So you will use the minidsp Umik-1 as input for your algorithm and you will use Audioinjector connected to an external amplifier for playback, am I correct?
Do you think it will be useful to have a single HAT with microphones and DAC/amplifier, dedicated to support Dirac on RPi?
Typically, the microphone will need to have a long cable, for ease of placement during measurements. But it’s connection to the embedded system is optional, the mic can also be connected to the controlling PC software (where the filter design is performed). For output, any good quality stereo card will suffice.
1 Like
Got it, many thanks!
Any plan for selling the Dirac Live Home for RPi/Volumio in the future?
I would really love to have that! Together with Volumio UI it will make the definitive player
1 Like
No plans at the moment, but I do see this as a very potent combo. Will ping our sales guys
1 Like
Great! You guys rock! I had the chance of listening to a demo of your product while working at NXP, I was really impressed, it was mind-blowing!
1 Like
The Raspberry Pi 4.9.3 toolchain was a no go, unfortunately, with glibc version mismatches. I’ll try the experimental distro instead.
Hello,
Modifying mod.conf is not the solution as mod is not the only player used. For Spotify or Spotify connect it BT or shairport it is different…
The actual architecture of output make difficult to redirect the output, especially with I2s Dac.
I have something almost working with plugin such as equalizer or DSP (using brutefir).
I made some test.month.ago with Umik on raspberry and it is recognized.
Nice! This is in an area of interest that I have only begun to explore… Really looking forward to see what this evolves into!
Hmm, I’ve been using an old tarball for armv6 support. You could give that a shot as well.
1 Like
@balbuze, as far as I can tell from the Volumio system architecture, as long as you play locally, you only need ALSA. I’m not very interested in streaming audio atm.
Which of the toolchains under arm-bcm2708/ are you using? The gcc 4.9.3 one? Edit: Yes, the 4.9.3 seems to produce runnable executables! 