First, I’d like to share a simple straightforward recording I managed to make with the Zynthian
(it’s a render of the Take 5 classic, based on a midi file)
1#FS>FluidR3 GM/Tenor Sax
2#FS>FluidR3 GM/Yamaha Grand Piano
3#FS>FluidR3 GM/Alto Sax
4#FS>FluidR3 GM/Soprano Sax
5#JV/Dexed>DCDCollection/01_FM-Rhodes
6#FS>FluidR3 GM/Baritone Sax
( Yes, that DX7 really belongs there )
Even though the render works out mostly ok, you can hear some crackling on the recording (that I didn’t hear at record time). So I assume the volume output was still to loud or is there another possible explanation for the audible crackling? (even though I reduced the “volume” / “gain” settings to 80% or less on all layers, and managing the relative loudness of the layers is quite frustrating.
So here are my “Software” questions:
Is there an easy way to manage the volume of the individual contributing layers in a single place ?
A Layer mixer of sorts ? (Right now, there are various settings inside the various engines at generation time).
When recording it would really be helpfull to see the elapsed time of the recording on the Audio Recorder screen.
@riban is developing an Audio Mixer. It provides a fader for each engine you load to adjust the volumn as well as L/R balance, mute and solo buttons. It’s not in the master repository yet. For more info check here on the forum. New mixer proposal
There are currently some pitfalls when trying to get this to work. I suggest you read the entire thread before deciding if you want to take it for a test drive or give it some more time to be developed.
Also a brand-new user here. I’ve been adding the gain 2x2 to each of my layers for an easy and consistent control (it offers negative gain). It doesn’t seem too taxing. Then, I do a lil’ bit of midi learning on my pc3k and each of its sliders maps as a gain control.
Each synth usually has its own master volume but on some, it is on page one, and on some, its on page 34.
Happy zynthing. Loving this thing. It goes well paired with the synthstrom deluge.
I ran the update from the web UI and switched to the zynmixer branch for zynthian-ui; hit “save”.
I ssh’d to zynthian.local and ran:
/zynthian/zynthian-ui/zynmixer/build.sh
systemctl stop zynthian
zynthian.sh
The UI came back on the Zynthian without errors and there was a top level “Audio Mixer” item.
I reloaded my (previously saved) snapshot.
I didn’t see any errors from the console, but when I opened the Audio Mixer I could only see the ‘Main’ total track on the far right, but not the 6 layers I have loaded.
So I assume I’m either doing something wrong or it’s not yet working correctly currently.
(a reboot didn’t help either, loading different snapshots consistently only shows only the “main” track on the far right, not the loaded layers).
From the other thread I see that it is working correctly for others, so most likely I’m missing something…
Snapshots are saved with their audio routing so if you try to load a snapshot that you saved in master then no audio will be routed to the mixer and hence no channels will show. Try creating your layers again when on zynmixer branch. (Or you could reroute each using its audio routing menu.)
Remember that snapshots saved in master and in zynmixer won’t be routed in the other.
Hi!
That worked, indeed!
I save a new (temporary) version of the snapshot, just in case.
I manually routed the audio of layer 1 to zynmixer:input_00a _00b etc, thinking happy thoughts about off by one errors
From then on I could see all 6 layers and I could easily change the levels of the layers, exactly what I was looking for!
It seems very easy and intuitive to use, but as I’ve noticed on some other occasions my touchscreen is behaving strangely, probably it needs some calibration.
I’m happpy with the preview at this point and I’m looking forward to see it land in the master branch when it’s declared stable.
I’m moving back to the master branch for now, I need the headphone support at this time of day (which is currently more recent in master) and I’m not very comfortable yet with mixing snapshots versions and straying too far from a ‘stable’ Zynthian that will still produce music.
I’ll check it back later, it looks very intuitive and a nice future feature for Zynthian!
Thanks for your kind words @twelve and your feedback. I considered zero / one based naming scheme for the inputs and ended up at zero-based because most other JACK interface seemed to use it and I thought it is only low-level code but you observe user interaction with this so I will change it to one-based. This is likely to upset any current snapshots but this is only a preview so no one should be using it in anger .
The touch interface for the mixer works best on a larger, capacitive touchscreen. (I use a 7" screen.) For users with standard (or even smaller) sized screens with resistive touch membrane then calibration is required and probably use of a stylus. For these users and those without a touchscreen there is also the ability to control the mixer using the rotary encoders.
Oh! And I merge master into zynmixer branch fairly regularly so new features should arrive fairly promptly.
Oh! And I have merged . zynmixer branch is now up to date with master as of 2020-08-11 08:26 (UTC). Do remember, this is a technology preview only for test and evaluation. This is not production ready.
)
.
. zynmixer branch is now up to date with master as of 2020-08-11 08:26 (UTC). Do remember, this is a technology preview only for test and evaluation. This is not production ready.