Advice on methods for overall interface design

I have a question regarding what your methodology is when creating interfaces for instruments. I have no problem creating fancy knobs and stuff, but all that means nothing if they don't make sense as a whole. It is not enough to have pretty looking pictures on the screen, they need to be placed right, have a certain cohesion, and various elements (knobs, sliders, buttons, etc) need to sort of match to one another.

So, I am interested to hear your opinions about the "macro" (not to be confused with Reaktor macros) view of how you put together an overall interface for instruments. How do you design the whole thing?

Take Proton 7 for example. Panel B of that instrument is a pure work of art as far as I am concerned. It is very easy to get to grips with, it guides the end user, looks great and simply "makes sense".

 

it can be hard to get a proper interface, i supose a bit of creativity is important. Though its good to look at other synths wich make sence for you even when looking at it for the first time.

I supose it's best to oversee the most important functions and keep it as small as possible. With stacked macro's you can easily have more without using so much space on your screen. i think Stacked macro's "idx" is best to control with buttons not with menu switches.

 

Do not overuse stacked macros! My wave XT in the UL is a good example of overuse, I did it a long time ago but it’s still a good example of doing it wrong.

When considering what you want to use to control your stacked macro I suggest you consider how often it will need to be accessed. If it is something that will receive a good deal of use I strongly suggest a fader, and use graphics to make it look like buttons. Why? Because you can assign it with midi learn to make moving around easier, buttons are not so great if you are doing it with a List module because reaktor only allows it to be vertical (how stupid is that). Using a button set to trigger will work but this is an extra mouse click.

I think it best to consider what groups will be used most; they are your target area! Keep them close and center stage. Not so important things can be placed somewhere out of the way, but do so as to push the eye to the main target area.

Read some articles and research on eye tracking, ok it’s for the web mostly, but there is a bunch of good ideas and practices there if you take the time to tease them out and apply them.

And finally, less is more! Less is more! Less is more! economize everhting! The everything but the kitchen sink mentality is a mess, leave it for pizza.

 

- draw quick schematics of the device' signalflow.. this already gives you a rough blueprint for a lay-out..
- then look at what is the core of the device so you know what to make visually more prominent..
- then determine which parameters are going to be used a lot and which not..

based on this inventory and findings you can make decisions on size, color, placement, stacking, etc..
---
- draw quick schematics of the device' signalflow.. this already gives you a rough blueprint for a lay-out..
- then look at what is the core of the device so you know what to make visually more prominent..
- then determine which parameters are going to be used a lot and which not..

based on this inventory and findings you can make decisions on size, color, placement, stacking, etc..

 

Great replies guys. "Guiding the eye" is a concept I had never heard of. Definitely gives me something to research into.

Horuschild,

I've seen you mention Skinman quite a bit. Is that what you use for laying out the non-moving graphics (i.e. labels, tick marks, lines that group the controls for a certain function (oscillator, filter, etc)? Would you use one large background graphic for this, individual backgrounds for each macro, or picture elements?

Do you find Skinman sufficient on it's own or do you find that you have to do a lot of the elements in Photoshop and then assemble them in Skinman?

Poster,
Good points. Although sometimes determining what is more important, and what will be used more can be difficult. I am currently updating an instrument I had built with R4 a while ago. There are some features that I've found people have been using in a completely different way than I had anticipated, which has partially sparked this upgrade.

Furthermore, is it better to emphasize sound shaping areas (filters, waveshapers, if those are what define the "sound" of the instrument) or performance related areas (by performance related areas I mean things like MIDI CC assignment controls, pitchbend range adjustment knobs, sequencer, etc.)

 

I personally find stacked macros and fancy knobs annoying because it just makes it harder to reuse elements of the synth in other synths.
I think a good way of viewing the interface is to view things as if you are a making a hardware synth and can only afford a certain number of physical interface elements.
With that in mind you wouldn't spend any of your budget on a micro adjustment keytracking on carrier 4 knob...All that does is make it harder to program and explore variations of settings that actually sculpt the sound in a big way.
Basically, try to get as few UI elements on the screen while still having a good amount of control over the synth. Naturally though we want to do the opposite and have "complete control" over everything.