Hello everyone. This is another bugfix for the 1.8 release candidates.
We found some nasty regressions that needed quick fixes, introduced initially in v1.7, now fixed.
Everyone that has previous 1.8 RC builds installed, please update as soon as possible.
Those who are beta testers will get this upgrade in their mod-ui notification panel.
Please refer to the RC1 topic to read the changes made in 1.8:
Here’s the changelog for RC5:
fix Duo IRQ setup (leading to random xruns, regression from v1.7)
fix missing MIDI running status support for DIN MIDI (regression from v1.7)
fix MIDI CC addressings missing their pink overlay in plugin settings dialog
Known Issues:
addressing a list to a button or footswitch does not wrap-around last/first values
something is off with transport timing, we are investigating
I have a 5 notes chord being sent by the Deluge sequencer to a couple of generators, in a loop, via MIDI DIN.
So far, so good, I can hear the generated sound. CPU is at 24%.
I went to the settings menu on the device to try and change the headphones volume.
It’s very difficult to reach that entry. You have to do several revolutions to reach it, sometimes it just jumps to the next entry.
One time, the unit reset when I clicked on the entry
Right now, clicking on the entry does nothing. It’s impossible to change the volume.
When I stop the sequencer, things go back to normal.
I have clock source set to MIDI and send midi clock to OFF.
No parameters have been assigned to any physical knob or button.
I have strange artifacts on the last row of the interface. Right know (nothing running) it looks like a closing bracket.
also, can you try disabling MIDI sync and seeing if it makes a difference for the changing headphone volume problem?
I already have an idea of what it can be, and that would confirm it if true.
hmmm it is not something to be too proud of, but we are basically making the RAM busy as it was discovered to be a source of noise (directly correlated to the audio buffer size).
fetching data from ram changes the running voltage by a small amount. doing it at regular intervals (like what jack does) creates a pattern that oscillates.
keeping the ram busy makes the voltage draw constant, and thus not producing that noise.
I tried many other things before this one, but nothing was making a difference.
As a last resort, I tried making a simple tool for the ram usage. …and it worked super well, wtf…
not happy with the solution, but I am happy with the results… tricky
Is that problem the same as what @gianfranco described a long time ago as potentially fixable by a switch to the mainline kernel ?
Out of curiosity, will there be ways in the future to totally isolate the processing circuit from the DAC/ADC circuitry ? Like some optical link or something, in order to totally avoid all interference (from RAM, USB, etc…) ?
I’ll be patient as I am nowhere near having exhausted the horsepower of the MOD anyway. And I understand that you don’t necessarily need to add more pressure onto yourself at the moment.
I tried with a different MIDI controller than the Sensel Morph and I didn’t get the same issues with stuck notes and lack of detection of note ON, so I think polyphonic aftertouch or MPE is confusing the DuoX (which does not happen with other synths)
