OS 2.30 beta B
The mod matrix has a problem / two connected problems.
One source of the problem is the +/-200% mod depth not aligned with typical 127 values of MIDI CCs. This makes it IMPOSSIBLE to do bit precise value jumps, i.e. using aftertouch as mod source, with depth of 1%, the CC value jumps between 0 and 1, with 2% it goes between 0 and 3. So it’s not possible to have 0 at one end and 2 at the other end of the aftertouch travel.
The second issue tangentially linked to this occurs when using project LFO as modulation source. For some reason, there are actually TWO different 0% depth amounts. The encoders goes -2, -1, 0, 0, 1, 2. With unipolar modulation, one of the zeros is stable / inactive (no values are send out), while with the other zero, the modulation is unstable / active = values jumping around the offset value are being send out. Both states are visually visible on the modulation visualization.
That’s bit surprising and possibly frustrating until understood. Nevertheless manageable. But only for the unipolar modulation. For bipolar modulation, both zero depth amounts are unstable / active! This is major issue when there is a mapping between LFO and for example pitch bend. Because even when the depth is set to zero, there are some pitch bend values jumping around zero being send out, which effectively overrides the pitch bends played from the actual synth keyboard. The same will probably be true for any other controller you might be using physically not expecting any interference from Hapax, because the depth in the mapping is set to 0.
I will report this to Squarp, but it seems that it’s worth discussing / knowing about, so I’m sharing it here as well. I’m curious if you have been aware of this and if there are some workarounds.
