I started experimenting with sending both LFOs to the cutoff knob on a patch and setting them at different times. The rhythmic stuff I get out of it is pretty neat. I wanted to create a patch in Drambo that would be a mega-modulator (which I’ve named MegaModulator…) of multiple LFOs and envelopes that all mix together to create some cool complex automation. However, when controlling the DN with these external LFOs, the sound is not as percussive as it is with the internal LFOs. I then tested it with the MIDI automation in LK (also on the iPad) and my OT’s LFOs set to the same settings as on the DN. As you can see in the video below, the DN (the first and more percussive of the examples) sounds a lot more percussive using both the exp and saw shapes as opposed to the OT’s version of the same thing. This leads me to believe that there’s something else going on with the DN LFOs than perhaps we know.
My guess is that it has to do with resolution as well as timing. If you are triggering a lfo on the start of a note any delay in the lfo signal would significantly effect the sound. Also the internal LFO’s are much higher resolution than midi cc’s, so that might have an effect if the lfo is moving fast or at a low depth. Because of this resolution it wouldn’t surprise me if the digitone smooths it’s midi cc inputs, so you don’t hear stepping. This would cause slower attack like you see in the video.
The incoming MIDI data is interpolated (smoothed), so you essentially lose some resolution (in simple terms)
MIDI is simply not high resolution enough to produce those types of modulations without zipper noise (artifacts caused by relatively large jumps in value), so the MIDI data is interpolated.
Using the internal LFOs sounds different because they do not operate under those restrictions, they can run at a much higher rate and still retain resolution, which is necessary for mimicking quick envelopes etc.
It’s akin to samplerate in recording - imagine capturing a sinewave at 440Hz in 48kHz samplerate versus the same in 12kHz - you lose definition.
EDIT: (Eh, missed @warpigs330 reply above, sorry for basically repeating the same thing!)
If I remember correctly the digitone’s internal parameters can be accessed at a higher resolution using a MSB and LSB midi cc. I don’t know any devices that use these together to get higher resolution modulations, but would that potentially effect this?
I don’t think so- this is an issue of the update rate, not the resolution of the value itself. (Think time domain vs frequency domain)
MIDI data is serial at an extremely low rate. Although, maybe you could do pseudo-interpolation using MSB/LSB, but it’s still quite slow and probably wonky haha… Interesting thought though.
Turbo MIDI speeds things up a lot - still probably not enough to make things truly snappy, but it has a lot of advantages like tighter sync, faster transfer rates over SDS etc. But, you are still at the mercy of the interpolation of the messages received, which I don’t think changes depending on whether it’s TM or not.
It’s pretty sad how it never caught on, back then they had the vision of making it a new standard protocol, but unfortunately couldn’t get any interest from the big companies. This was way back when Elektron was very niche.
Is it possible to use TM with MIDI loopback? I use IN and OUT connected to access MIDI tracks for extra controlling DN.
I like TM and much agree with you that it has more potential. A side note: I say TM would be even better with no handshake requirement (unidirectional connection). MIDI DIN is unidirectional optimized. MIDI 2.0 is also bidirectional which makes sense with USB but makes old MIDI DIN integration difficult.
To let it unidir just need FORCE TM as a setting and then TM is set without handshake.
I assume the MIDI loop back would suffer the same issues for this particular case, yeah? I like the idea of loopback but my setup runs through a patch bay and doing the LB would mess up the overall flow. That’s why I wanted to integrate my iPad more into the rig with these lfos.
