Synths: how deep is too deep?

The Lounge Other Gear
1

Spinning off from discussion of the ASM Leviasynth, I thought I’d create a new topic for more general discussion.

Digital synths have evolved a lot since the introduction of the DX7. General advances in computing have really led the way here as relatively simple operating systems and UIs have given way to color touchscreens and regular firmware updates.

The hobbyist market has also evolved along with the synths. More users expect periodic upgrades and want more out of each individual box. Spec sheet comparisons run rampant once products are announced. Developers add many layers of functionality accessed via button combinations, color codes, or deep menus. But how much is too much, in terms of usability, marketability, or aesthetics?

I’ve been thinking about this a lot lately with the introduction of the Leviasynth, and also various firmware updates that add more features (but also more complexity) to stuff I already have. At the same time I’ve been on a deep dive into a relatively simple instrument, which has been very satisfying.

The Hydrasynth, in my view, is a modern classic. I feel like they started with the idea of “Hey guys, you know those diagrams on old digital synths? What if we put all the controls within that diagram so you knew where you were in the signal flow?” I don’t use my Explorer a lot, but every time I pick it up 90% of what I need to know is right there on that front panel.

My initial impression of Leviasynth, as someone who first learned FM synthesis, is that it’s a little like wishing to have a pet dinosaur and then actually getting one. I feel pretty across DX7-compatible FM, but I’m not as comfortable in OpSix land, and the idea of 144 algorithms is pretty intimidating even if you know the OG 32 reasonably well. The front panel doesn’t look as immediately legible to me. I could probably figure out how to get something nice out of it, but the reality is I already feel committed to enough other stuff that I’m not sure I want to make the time.

I don’t want to single out ASM here - I think we can all think of a lot of examples of gear where we’ve been treated to heaps of new features but at the cost of usability, or complex interfaces that are poorly documented.

In a world that looks more like the NGNY and Downsizing threads, we all find our One True Synth and/or Groovebox and settle down to make some nice untz untz. But I think it’s true that most people who are in the market for a $2800 keyboard are very likely to already have one or more $1000+ synths already. Which is totally fine! But, does every synth (or groovebox, or module) need to do it all? Is it better to have more depth you can ignore, or a smoother user interface?

What do you all think?

16 Likes
2

Now that i’ve got through the manual, i’ve decided the LS is a lot simpler than i thought.

One unified structure, with the variations built on the inside, rather than being a collection of separate pieces brought together, all using synth techniques we already know, tied together with a seamless user interface. Big plus too for anyone familiar with the HS, as you already have the context and the basics down.

Oh and i noticed – it’s not an FM synth.

6 Likes
3

The only time super deep stuff bothers me is when it’s very insistent on you relearning what every synthesis concept is for that specific synth in order to understand it. Cydrums had that issue for me, it has lots of depth and power but the way it uses FM and wavetables isn’t familiar or intuitive at all and I was constantly having to learn it just to make it do straightforward things.

On the other hand, I can actually just ignore the extra envelopes and depth options in something like a Hydrasynth or a 3rd wave and it will still be a predictable experience for anyone familiar with standard synthesis techniques. The depth there is just an extra for when you’re ready to get there.

So as long as a synth offers a workflow where someone can follow somewhat standard recipes if they want and achieve predictable results I don’t think it’s in danger of being too deep. If you have to learn a lot of very product-specific actions just to make a dual saw patch that can drone as long as you hold the gate, it’s too deep.

11 Likes
4

Too deep? I don’t think so. As long as I can still get around it perfectly fine without having to go deep.

6 Likes
5

I can barely figure out how to turn my synths on!

5 Likes
7

Another thing I had in mind is the increasing trend in Eurorack stuff towards miniaturization and multifunction modules if anyone is excited about discussing that. I think it’s part of this larger idea of increasing complexity and updates. Of course you can ignore all that and go for more basic building blocks.

One thing I’ve observed from various fan communities (and I think “synth aficionados” qualifies as one) is that the views and desires of your most vocal fans don’t always reflect those of the majority of the users or buyers of your product. So, you might have 15 people on your Discord who really want something that might actually put off a bunch of the 1500 people who bought your boutique product. It is probably really hard to get the balance right as a creator and I’m amazed we have so many interesting things popping up right now.

6 Likes
8

Exactly! leviasynth actually looks reasonably straightforward with plenty of visually feedback showing what’s actually occurring. There would of course be a lot of muscle memory to learn to navigate it quickly, but you can quite easily just hit buttons that say things like “oscillator” and “amp envelope” and expose the standard synthesis tools. I actually quite liked seeing that there’s a way to visually monitor levels through the signal chain without losing the oscilloscope view. Really handy to see how much mixer headroom you have left, how hot the signal hitting the filter is, etc. It might be one of the best ‘learning synths’ ever made for that reason.

I doubt I’d use more than 4 oscillators in any given patch and will need to be convinced on the sound since I prefer synths that sell based on pure sound quality over synths that sell based on how deep they can get, but it actually does look quite a lot more straightforward for the available depth than the multi/poly interface and that’s quite a feat.

@shigginpit I think these specific examples are relevant since they are among the deepest currently available synths (alongside like, iridium) and how to present vastness of possibilities in a way that does not feel “too deep” to the synthesist is pretty germane to OP’s question.

2 Likes
9

In regard to the question I think it’s simple: a synth which does not offer a clear path to a satisfactory result will often be perceived as too deep. A synth which yields unfavorable or unimpressive results even though the user understands how to use it is often said to be not deep enough.

The answer lies with the user and their propensity for comprehension more than the synth, but if we were to create a hypothetical middle-ground which most people could agree with, a synth which is too deep makes the resolution of parameter control disproportionate to the audible result and makes the additional depth of the engines disproportionate to the achievable results.

I think it’s far more about implementation and user experience than actual depth of synthesis because some of these companies in an effort to create more possibilites attempt to redefine the paradigm of synthesis. Sometimes that’s a hit, sometimes that’s a miss.

I think that if we’re to see progress in this oversaturated field, there have to be a lot of misses for every hit so those companies which don’t really understand user experience and focus on potential in a clinical environment often miss with users due to many users, even those who focus on sound design, having things to do and can’t spend all day tuning a waveform in an unintuitive menu structure.

18 Likes
10

Kawai K5000R and Yamaha FS1r - almost all parameters could be edited through their front panel, but it was painful with thousands of parameters.

For hardware, the Monomachine’s ‘simplified multi-synthesis’ was perfect, IMHO.

These days I prefer soft synths for deep editing. Easy total recall, automation, big screen, less cables, etc.

I guess both, Bram Bos apps on iOS are a great example of great UX - simple interfaces with the most important parameters right in front, but additional windows when you want to dive deep.

5 Likes
11

No such thing as too deep. Don’t like certain features/ synthesis methods/ workflows? Ignore them or use a different thing that caters more to what you want. More options is always better, assuming they’re laid out in a reasonable way and the manufacturer makes an attempt at following established conventions. I think that is the real sticking point for a lot of complex devices.

10 Likes
12

Can’t be deep enough as long as the interface is awesome.

Examples of awesome: Digitone II, Hydrasynth.

12 Likes
13

Some synths are just badly designed and they might seem/feel too deep.
They’re not too deep, they’re just awful.

10 Likes
14

The trend toward increasing complexity has been evident for quite some time. The amount of data, parameters, samples, LFOs, oscillators, you name it… I have to remind myself every now and then that although I have the option for complexity, I don’t necessarily have to use 8 oscillators and 18 LFOs…

But then the question is: Do I want to buy a synth like that?

3 Likes
15

I find it wild that as others have worked to simplify FM, Leviasynth decided to make it more complex…

My understanding from a couple of the videos is that morphing between algorithms and operator panning are, 1) the big new things on the table and 2) the reason the manual has a 15-page appendix of different algorithms.

How deep is too deep? I think of Brian Eno’s comments on the H3000—if the interface is good, it invites play and exploration, so that no single step into an additional dimension feels daunting…

4 Likes
16

I’ve thought about this a lot and could post my longer thoughts later, but to start with the short version is that you can increase a sound’s complexity either vertically or horizontally.

Vertical complexity basically involves layering and we see that in a lot of synths. More and more layers, more voices, more oscillators, etc. But at a certain point, I think it’s more interesting to layer different musical parts in your arrangement rather than have a massive layered sound beneath a single keypress. And digital recording with unlimited track counts means we don’t really need that kind of massive layered complexity within an individual instrument anymore.

Horizontal complexity is changes over time which means modulation, more lfos, more envelopes, etc. But when you think about acoustic instruments, those changes over time happen through performance and expression. Maybe it’s more interesting to get additional movement and detail through things like aftertouch, poly at, and other expressive performance controls rather than programming a lot of lfos and envelopes. And there’s so much you can do with post processing to continue to add complexity over time even after the part has been recorded. So again, maybe having that complexity available at the patch level isn’t entirely necessary.

12 Likes
17

I find a lot of stuff too deep, and had similar thoughts when I saw the Leviasynth. I have some soft synths that have weird and wonderful oscillators and incredibly deep modulation configuration, like chainable 8 stage envelopes etc.

I never use them because the sounds get too divergent from the sounds i want and the sounds that fit the type of music I am making.

Like if you’re making house and want a piano or organ stab, something with 3 space nebula oscillators per voice into 2 parallel filters & 3 fx buses all modulated by 7 different lfos with custom waveforms it just wont fit. More doesnt equal better.

I like the ideas but i think they are better for ppl who are doing sound design for like games or interactive installations and things like that.

4 Likes
18

Anything more than a single oscillator running through a bag of Doritos and I’m cooked.

9 Likes
19

very well said…

1 Like
20

This is a really important aspect of these super deep synths that should be better recognized by reviewers in my opinion. Most of the synth tube folks are happy to tell you about all the cool stuff a piece of gear can do and whether or not it makes it convenient to do so but most of them don’t really spend any effort judging if that would be useful for musicians or is a feature that only working sound designers will find any utility for. Like, I never use more than 3 envelopes to make a synth sound for a piece of music and usually 2 is plenty. Evolving patches have to be designed especially for their purpose or they’re guaranteed to be out of place, so there’s no point in keeping them on a preset bank for songwriting unless you build a song around the specific one.

The closest they ever get is pointing out how “cinematic” something can sound or whatever. When I first got my multi/poly one of my first day reactions was thinking how overkill a lot of its capabilities were for use in most sorts of music and that it was obviously designed to satisfy people doing really elaborate sound design work for multimedia applications and that’s been one of its go-to uses for me; replicating things like soundtrack/film score sounds to understand how they were made and generally serving as a test bench for experiments in synthesis concepts. It rarely ends up on a song though.

5 Likes
21

I think in the current day and age there is little justification for a hardware synth or sound module to be really complicated. Unless it is something endlessly usable in every context and replaces many synths (like, maybe, Iridium). Life is to short for endless scrolling of parameters!
It was really refreshing to see Wavestate, ModWave and MultiPoly live, I already had OpSix, but as much as I like their sound and capabilities, my GAS for this trio expired immediately. MultiPoly is too complex even as a plugin, it’s closer to a groovebox with all these lanes and layers.

While I frequently wish for Elektron engines to be a little bit deeper or work differently, they definitely win here with “1 to 1” mapping of parameters to knobs.

Among my first very first hardware synths were Roland Gaia and Blofeld. For me the sweet-spot is definitely somewhere in between. Gaia was three identical layers of classic subtractive engines with knob per function, but with only 1 LFO. Blofeld is actually quite intuitive (and much much more powerful), but requires significant amount diving and scrolling. More knobby Blofeld or Roland with proper mod-matrix, somewhere here.

1 Like