A dream hardware controller for soft synths?

[noiseboyuk]

Yes to everything there, you described very well what's in my head!

[Teksonik]

the question always is if would it be actually preferable workflow-wise over using a mouse and/or generic manually assigned controllers.

For sound design I can't imagine anything more efficient than changing parameters with a mouse in my left hand while playing notes with my right hand.

I have an Akai MPK261 and it has 3 banks of knobs and sliders for a total of 24 each. I could set up presets to control various softsynths but I don't for sound design. Much easier and faster for me to use the mouse and on screen GUI.

For controlling parameters while playing or sequencing yes the knobs and sliders are very useful. Sweep the cutoff with a knob, control ADSR with the sliders etc but no matter how complex a controller may be you're most likely to still need to have mouse in hand to edit some softsynth parameters (how to open a drop down menu then scroll up/down then select the desired value etc)

There was this:
http://www.gearjunkies.com/2011/05/ulti ... ontroller/

[CrystalWizard]

About 24 or 32 knobs around 2 or 3 touch screens plus some wheels, joysticks, a keyboard, a ribbon, and maybe a theremin like sensor. Also softsynths with 24 macros instead of 8. Inputs for 2 pedals and 8 floor switches. Even just a keyboard with many knobs and faders with two slots for ipads would be a start.
While i'm waiting i can combine a midi floor controller, a keyboard like SL MKII, and two ipads running touch osc plus hours of programming. Yes, there's plenty of room for better controllers and i see this as more valuable than better software synths and effects. It's not a trivial task to design something customizable yet general. The software needs some sort of general interface, so in that sense the software could be improved. As posted some sort of standard would facilitate matters.

[Funkybot's Evil Twin]

As a System-8 owner, I still believe it's the best controller for softsynths available. I'm constantly looking for better hardware control of software, and I've got lots of thoughts on the subject. I'm not responding to anyone in particular, but a few points:

1. Any controller is about compromise. There's no way to build a controller that will map to everything perfectly. It's not a realistic goal.

2. Blindly auto-mapping plugin controls to hardware will never work, but a standard like NKS (now for effects too) is a fine approach. I hope NI allows other manufacturers to create NKS compatible hardware and allow it to truly become an open format. In the absence of developer support, users should be allowed to create templates via some kind of editor. MIDI Learn in the plugins is the next best thing.

3. Layout is important, but so is flexibility. Having the System-8 mapped to softsynths works fairly well. Filter cutoff, resonance, HP, ADSR's, LFO's, reaching for these has become second nature. Where I get tripped up is around Osc parameters when some synths offer on/off buttons for waveforms and other synths use knobs for that task. A good synth controller would have a generic synth layout, with a strong enough combination of parameters to be flexible (example: knobs and buttons in the osc section). A good synth controller would also have a robust effect section in its layout.

4. Different tasks benefit from different control layouts. A good generic synth controller keyboard may make your hardware synths feel more like an instrument, but it would make for a clumsy effect controller box. The task matters. A hardware synth controller, a knob box for effects, and a mixer control surface would all be complimentary. I feel like splitting out controlling into those three specific tasks (rather than trying to make something good at everything) would be a better approach.

5. I agree with everyone who thinks high resolution screens with hardware knobs and faders are the way to go. I'd love a synth with a Roland JD-XA style layout, where the surface ran something Lemur/TouchOSC where I could rename/assign parameters and have the synth GUI change to reflect the current layout. Build something like that, make it waterproof (spilled beer is a thing at gigs), add a five-octave keyboard, pitch/modwheels, and lots of knobs/buttons/faders. If a touchscreen, physical buttons wouldn't be necessary, allow users/developers to create virtual buttons, but I'd want physical knobs and faders.

6. How many physical controls for a synth controller? We have 127 MIDI CC's right? Let's start from there and work down. The System-8 has something like 70 controls that transmit CC's and I still feel limited at times. If banking from one layout to another could truly change the entire UI because the contnroller was built on a screen, then fewer controls would be necessary.

7. A good controller like that won't be cheap. There's probably a good reason nothing like this exists. I'm sure it's occurred to someone at Native Instruments, but the closest they've come up with are their Komplete Kontrol keyboards and those aren't inexpensive. I'd expect something like that to be arouond 2.5k. Then who'd still be interested?

Just some random thoughts. For now, get a synth that transmits MIDI CC's if you want a synth controller and accept the compromise. Get a knob-box if you want to control effects. Get a mixing control surface if you want to mix. We're probably still a ways away from a truly excellent solution. That said, I think Native Instruments is on the right track with NKS.

[recursive one]

2. Blindly auto-mapping plugin controls to hardware will never work, but a standard like NKS (now for effects too) is a fine approach. I hope NI allows other manufacturers to create NKS compatible hardware and allow it to truly become an open format. In the absence of developer support, users should be allowed to create templates via some kind of editor. MIDI Learn in the plugins is the next best thing.

I don't know how NKS actually works since I don't own any NI hardware but one of the ideas discussed here was how to make auto-mapping less blind.

For example, the "dream controller" will have two big knobs called Cutoff and Reso and two small knobs called Filter Control 1 and 2 grouped together. Then compatible plugins should have their own specific charts which could tell the hardware that this parameter is the cutoff, that parameter is the resonance, and these are, e.g., filter drive and filter FM, so they should be mapped to Filter Control 1 and 2. There should also be some software allowing users to create and distrubite their own plugin charts. Something like that.

[rod_zero]

OF all the problems discussed to me automapping is the last one, as long as you only need to map it once and set it for life I am ok, it is not that big of a hassle for me. I would also expect different users will ahve different ways to map some controls.

[Funkybot's Evil Twin]

OF all the problems discussed to me automapping is the last one, as long as you only need to map it once and set it for life I am ok, it is not that big of a hassle for me. I would also expect different users will ahve different ways to map some controls.

Yep, exactly my feelings too.

[acYm]

there was already such a controller on the market, a very mainstream product from a major brand, one that worked so well that you realized it's actually most often the VST's that are the limitation with hardware control and not the interface itself, but it was largely rejected by the community despite being highly powerful and versatile. not sure why. nobody ever took the time to understand it or develop it from the user side, except maybe 3 or 4 enthusiasts. old forums are full of wrong or silly information about what it can and can't do, that was never corrected. it's quite unbelievable, really. and now, here we are, with this thread.

[aMUSEd]

there was already such a controller on the market, a very mainstream product from a major brand, one that worked so well that you realized it's actually most often the VST's that are the limitation with hardware control and not the interface itself, but it was largely rejected by the community despite being highly powerful and versatile. not sure why. nobody ever took the time to understand it or develop it from the user side, except maybe 3 or 4 enthusiasts. old forums are full of wrong or silly information about what it can and can't do, that was never corrected. it's quite unbelievable, really. and now, here we are, with this thread.

I did

[chk071]

Which controller was that? KORE?

[aMUSEd]

2. Blindly auto-mapping plugin controls to hardware will never work, but a standard like NKS (now for effects too) is a fine approach. I hope NI allows other manufacturers to create NKS compatible hardware and allow it to truly become an open format. In the absence of developer support, users should be allowed to create templates via some kind of editor. MIDI Learn in the plugins is the next best thing.

I don't know how NKS actually works since I don't own any NI hardware but one of the ideas discussed here was how to make auto-mapping less blind.

For example, the "dream controller" will have two big knobs called Cutoff and Reso and two small knobs called Filter Control 1 and 2 grouped together. Then compatible plugins should have their own specific charts which could tell the hardware that this parameter is the cutoff, that parameter is the resonance, and these are, e.g., filter drive and filter FM, so they should be mapped to Filter Control 1 and 2. There should also be some software allowing users to create and distrubite their own plugin charts. Something like that.

You can do all that already in Komplete Kontrol (and before that Kore 2)

[acYm]

I did

for sure. in a better world, I bet you would have gotten hired as lead template designer.

Which controller was that? KORE?

indeed. my weird decision to re-install it and make it the center of my studio in 2014 is paying off big time. everything is so much more fun now and it lives in perfect symbiosis with ableton.

believe me, the real tragedy here is that my epic Kor3 concepts will stay in my head because NI are completely deaf to my arguments.

[Touch The Universe]

It's 2018. I'd like to see some more futuristic touch screen sort of devices to use as controllers, morphable layouts where the knobs can be adjusted and moved at will, layed out on a grid on top of touchscreen held in place by electro - magnetism. Theres got to be some way to place an object on top of a touch screen, have it stay in place, and send and receive messages through. If I want to load up sylenth, the parameters will load up a map of sylenths parameters on the grid and move the knob plates on there on there own. Knobs can be added or removed as needed.

For instance, there could be a knob placed directly on top of the cutoff in sylenths gui on the touch screen and it will control that parameter, stuck in place like a fridge magnet. If the knobs can be moved freely, I'd imaging only 8 or so would be needed. Need to tweak resonance, simply move the knob over the resonance on the gui and tweak. No need for a cluttered layout, it can be as simple or complex as desired. Since no one probably tweaks more than two knobs at once, this might be all that is needed. Just two interactive modular knobs that can be placed anywhere on a touch screen interactive surface with mapped synth guis.

Probably impossible and sounds expensive, but might be cool if it existed. I'm sure the op had something more realistic and practical in mind, but, hey, its fun to dream

A reaktable controller for synths.

[zenophilix]

God, yes. I'm not the only one wanting this. Although, granted, I'd just be happy with a small display under each control on my keyboard that changes depending on what MIDI/Mackie control is assigned to it at the time. I find myself not even using the channel faders anymore because it's easier to just use the mouse than remember what fader is controlling what track.

[Dunc]

There are Knobs you can stick onto your touchscreen. They've been out for a couple of years I think, but they are supposed to be a bit laggy and unreliable ( never tried them myself, so I'm not to sure about the truth of the matter ).

http://www.pepperdecks.com/products/28-tunagear


Both of the below products are not touchscreen based, but look interesting none the less:

Sensel Morph at NAMM is starting to look good for a future make your own controller ( with programming involved ) -------

Also, if they ever get the damn thing out, is -----https://www.play-joue.com/en/