Yeah, it’s very easy to slip into problem solving mode when discussing this type of thing.
I don’t really want to bog this thread down with my own views, but I have thought a lot about parameter editing over the years, and which compromises are better and which are less good.
For instance, on the Prophet 5 and Nord Lead 1/2/2X, all the mod routings and waveforms are displayed using LEDs on the panel, whereas the continuous parameters are handled with knobs so you can’t see the values. For me, this means that at a quick glance I can see the architecture of a patch, even if I can’t see the actual parameter values. So even if these machines have a fairly old UI, the use of LEDs to display switched values is actually a great idea in terms of user feedback.
In contrast, I had a Nord Lead A1 once, and in that machine, there is a single modulation envelope, where separate continuous controls control the amount of envelope fed to the oscillator and filter, respectively. This means that I cannot tell at a glance what is actually being modulated by the envelope. It’s a small detail, but convolutes the synth for me. There is also a feature whereby one can read the current value of a parameter, by pressing a button and turning the corresponding knob, but it’s a two-hand operation and thus rather clumsy. This would not be an issue on the Zynthian, as there’s a display to display values, but I’m just trying to relate how important that type of feedback is, IMHO.
Another two-hand operation on the A1 is, IIRC, setting the filter keyboard tracking. This means that I cannot play the synth while adjusting the keyboard tracking. Again, this makes the UI less accessible.
The Audiothingies P6, with it’s rigid layout of a number of more or less directly accessible parameter pages (some require multiple presses), and character display, doesn’t look too sexy, but it’s very appealing to me in terms of accessibility. Few things require two hands, and the interface paradigm is quick to learn, making the synth inviting to edit. Then again, the architecture lends itself towards this type of UI; if there were ten times more parameters, the matrix of parameter selectoin buttons would be daunting.
One thing that I appreciate about the Blofeld is the way the envelopes are displayed graphically. Even if they don’t show a scale representation of the actual envelope, they do give a quick visible feedback of the envelope shape. The Blofeld uses the graphics to display the filter response too, as well as the oscillator and LFO waveforms and frequencies.
I think that sums it up rather well.
I think synth UIs are a very personal thing. For me, the ability to quickly navigate between parameter pages, preferably without having to scroll through lists of things, is key. I’d rather have to press a slightly convoluted combination of buttons to get where I want to go, than have to look at the display, see what page is up, and press a button a certain number of times depending on what page was actually up, or have to scroll through lists.
Speaking of which, the accessibility of lists can depend very much on the encoder response. For instance, the Blofeld uses its stepless encoders for selecting stepped parameters like mod matrix slots or waveforms, in contrast to the Akai Miniak which has a very clicky encoder which is much better for that type of operation, as it’s easy to look at the display, think “I need to move three steps down”, and turn the encoder three clicks, rather than having to constantly stare at the display to see when one has moved exactly three steps down. A small detail, I know.
I’m not really fond of macros or maps of subsets of parameters, because it always seems like I’ll soon need to access something that is not available. On my Circuit, which I use mainly for blippy sequencer-type sounds, I’ve created a patch where I’ve set up the 8 knobs to access enough parameters to create a reasonably wide range of sounds in that domain, in effect, creating a synth within a synth. It works well, but only because I’ve purposely decided to see the instrument (or at least that patch) as a small synth with only few parameters, sortof like an Electribe EA-1. So it’s certainly useful to be able to do things like that, but it requires the extra step of designing that synth-within-a-synth first.
I feel I’ve taken enough space in this thread with these random thoughts. Over to someone else.