search

mnemonicdevices / md001

Issue and project tracking for MD-001 Juniper
6 stars 0 forks source link

Linking channels #75

Open ost-ing opened 5 months ago

ost-ing commented 5 months ago

Describe the topic for investigation Linked channels

This is useful for Recorders, Lorenz Attractors and pitch / chords. A lot of unknowns about UI and implementation that need to be figured out.

cardinal-bin commented 5 months ago

Reading through the discussion on the Lorenz attractors thread I have a related suggestion for how 'linking' could work. Although I'm not exactly sure how your implementation would accept this, it makes sense from a UX perspective and could be relatively frictionless.

If we imagine a channel set to lorenz, which produces 3 different values x, y, and z. On the channel itself we move between them with morph if I understand the present implementation. However, my suggestion is that for other channels, the out waveform now shows not just (for example) channel 5 as an option - but all 3 of channel 1's x, y, and z outputs*. In this way you can setup one channel to be the parent lorenz channel, and any number of others to monitor the different outputs (or combinations thereof).

The slight downside is that you can't directly monitor the lorenz channel, and I suppose you can't blend channel 4 with, say, channel 5 z, but I think it's a nice trade-off, and this is already the case with non-adjacent channels anyway.

*Just to clarify - I'm imagining that on the channel set to out we would morph like so: channel 4 => channel 5 x => channel 5 y => channel 5 z => channel 6.

ost-ing commented 4 months ago

@cardinal-bin Thanks for the suggestion! An issue with this is that if channel 5 is no longer set to Lorenz, then what happens to channel 4? The mapping breaks in this circumstance, especially if something like waveform modulation was enabled.

cardinal-bin commented 4 months ago

@ostenning Oh good point - I suppose the only way would be to have an out channel set to any of the 'sub channels' just jump to the parent channel if the waveform changes. I suppose any situation where linked channels are specific to a waveform (i.e. Lorenz) you might well have similar issues.

ost-ing commented 4 months ago

@ostenning Oh good point - I suppose the only way would be to have an out channel set to any of the 'sub channels' just jump to the parent channel if the waveform changes. I suppose any situation where linked channels are specific to a waveform (i.e. Lorenz) you might well have similar issues.

Im considering removing the waveform modulation feature. To me it feels too experimental #78 and adds significant overhead. If I do then this feature could work, the user can still manually break the connection but providing the info bubble informs the user I think it would be OK