Is the Headphone Output separately addressable? for pre-listening and personalised monitor-mix purposes? And if not, could this be implemented someday, or is it a hardware limitation?
There is a headphone output on the V5 which is a simple, amplfier buffered copy of the main output. This is not separately addressable.
On the RPi4 and earlier there is a dedicated headphone output which is a low quality PCM port which requires substantial resources, e.g. CPU to drive it. It is not recommended for use. This does not exist on the RPi5 which does have separate audio ports for each of its HDMI outlets but these too may suffer similar constaints. (I have not tested this assumption.) The HDMI audio outputs seem to only be available when the HDMI ports are enabled and that causes issues with the system graphics if the HDMI port is not the primary video output.
It is possible to add audio outputs via USB, Bluetooth, etc.
Even though these RPi4 and USB outputs (and inputs) are separately addressable by ALSA, there are challenges to present them as jackd ports. You can enable “Hotplug audio” which performs the required samplerate convesion to support this workflow (at least for USB audio) and these extra ports are available for routing but be aware that this adds resource load, e.g. CPU. So there is a balance between extra audio ports and quanity and type of engines that can be run.
Note that a USB audio output used for zynvoice accessibility does not use jackd and so uses less additional resource.
Oh yeah, I forgot to mention → Zynthian 5.1 with the latest stable oram build.
Thanks for the info! Looks like I’ll have to poke this USB audio thing a little bit ![]()
Zynthian kits have 2 audio channels, both input and output that are nominally configured as stereo but can be routed flexibly. If you need more inputs and/or outputs, you can add a USB soundcard (ideally replacing the internal soundcard). I have both Behringer UMC1820 and Tascam US-1608, attached to different devices. I prefer the Tascam because it has DSP on each input channel giving EQ and compression on each input without any CPU usage, i.e. no need for eq or compressor plugins on audio inputs. Both are fully supported by zynthian, as are many other audio interfaces with fewer inputs and outputs.
Basically, if you need more than 2 inputs and/or 2 outputs, buy a suitable USB soundcard with sufficient inputs/outputs for your needs, ideally completely replacing and not using the zynthian builtin soundcard.
As @riban explained, the 5.1 headphone output is an amplified version of the main output. No way to have a separated mix on it.
But as @riban also suggested, you can buy a (cheap) USB audio interface and enable “Hotplug audio”. This will make available the USB audio output to the zynthian audio router and you can configure a mixbus chain to achieve what you want.
Regards,
…hmmm… I have a spare old RME Fireface 400 somewhere collecting dust… But I don’t think i’ll get that running as a USB Class Compliant interface… ![]()
I love the super low latency and stability of RME interfaces though…
I am glad I didn’t go too far down the FireWire route. There were the superior solution but got a dead end when USB improved. I doubt we will get any FireWire audio interface running on zynthian.
Hi @nemoy,
What about this Zynthian-proven, inexpensive, class-compliant and effective USB interface? I can confirm that it works like a charm as an almost-dongle audio solution.
I was curious about this question as I use the IQAudio/Pi DAC Pro with the RCA outputs going to PA and trying to get my headphones at a reasonable level.
I bought one of these small passive mixers for another problem but I think it now has a new purpose in life. The headphone out from the DAC Pro drives my headphones very loud so I think this could work well.

