1. Didja accidentally blow through the whole, "We're using our real names" thing on registration? No problem, just send me (Mike) a Conversation message and I'll get you sorted, by which I mean hammered-into-obedient-line because I'm SO about having a lot of individuality-destroying, oppressive shit all over my forum.
    Dismiss Notice
  2. Happy 2018, Everybody! Thanks for making Redbanned such a cool, helpful, and inviting place. Here's to another great year!
    Dismiss Notice

Custom 88 Semi-Weighted Controller. Interested?

Discussion in 'Custom Projects' started by Mike Verta, Jul 2, 2017.

  1. So...

    Since it doesn't exist, I need it, and I happen to know guys who can do it*, I'm designing a custom 88-key semi-weighted controller. There is absolutely no fantastic-feeling, solid, synth-key-profiled, aftertouch keybed being put out there - not by Fatar, or Roland, or Yamaha, or anyone. That's my primary goal. But of course, it needs a good array of wheels and sliders and buttons and pads and displays, as well. In truth, I have a lot of ideas (would love to hear yours, as well), and it's going to be fun.

    But I also know I'm not the only one who desperately misses a "performer's controller" like this. So I thought I would mention it, take an interest poll. Perhaps if there's enough interest we can talk about doing a limited production run, or a crowdfunding thing; who knows. As much as I dig the idea of having the only one out there, I dig more the idea of helping musicians be musicians.


    _Mike

    *I just came off a year of working on the new Disneyworld Avatar/Pandora theme park, with a group of guys I've known for years. They've built tons of extremely cool, robust, and mechanically/technologically complex stuff not only for Disney but for corporations world-wide for decades. I've never seen them fail a challenge, and considering what we had to pull off for James Cameron this year, I'd say a keyboard controller is about a weekend's work. With state-of-the-art prototyping and manufacturing facilities, and a profoundly intuitive sense for managing costs while maintaining absolute quality, I can't imagine having a better resource to draw upon. I should have done this sooner. I was just sure somebody was already working on it. They aren't. And if you want something done right...
     
    Jason Watts and Gary Williams like this.
  2. "Semi-weighted" can mean a lot of things, but any keybed that's not piano-weighted/hammer action-like counts. Quality is all over the map, none of them great. Hence...

    Speaking of which...

    With 88-keys, working traditional modwheels at the upper end of the range bends the left wrist in a unusual and hard-to-anchor position. I did some tests mocking up a shallow, ergonomically-comfortable surface which can be rotated at an angle to relieve tension and place controls in a natural place along natural motion vectors. It allows for eyes-free control while providing support for delicate movements and maximum comfort. This is the general idea:

    [​IMG]

    With these sorts of things, proportion and angles are everything, so after working it up in CAD, I'll have to 3D print a bunch of prototypes to find the ideal actual positioning, but this already capitalizes on something our input mice have done for decades. In fact, I thought about that index-finger slider being a wheel as well, but I think fader is better, and 60mm seems to be just about the ideal throw for that space without pulling the hand of the anchor point.

    _Mike
     
  3. Just sketching some things. This is not the form factor or aesthetic in any way. I really think the controller-as-desktop-surface is the way I'm going. The faders mock-up here would be a modular unit, and the center keyboard/mouse area can either host another control bank, or serve as the keyboard place, where its height relative to the surface would be adjustable for flush-mount, or elevated ergonomics. One thing I'm doing is putting buttons just above the keys, two per octave. I will use these about a billion times a day.

    [​IMG]
     
  4. Mike, who is doing the circuitboard design and software?

    We do development design of embedded processor boards (both hardware and software) and have built boards for several large companies and the hardware design process is fairly costly and not exactly trivial. We use Altium to design multi-layer embedded processor boards and it typically takes a few engineers several months to refine a complex project. All kinks have to be completely ironed out or botched project builds can be very costly. We also have a paste machine, a large 4-head Essemtec Paraquda pick n place and a reflow oven where we do development and up to medium sized board builds (several hundred). For larger productions we send projects offshore.

    Its hard to envision how to accomplish cost-effective small runs and maintain a high level of quality with a complex project such as this. In addition, there is a software component that's probably somewhat involved.
    Obviously there are lots of things I can't say publicly but at some point you might find it useful to talk with my son who is the lead engineer/president to bounce ideas. Our website is very out-dated as its not really a point of sales for us. https://revotics.com/

    Me? I'm just the musician Dad of a very ambitious son who started building his business above our garage, then took over the garage as well. He now has his own house (two houses away) with seven garages/office/shops. Fortunately, I still have my studio though.

    I'm not proposing anything at this point but if you have questions on the hardware design process, feel free to inquire.
     
  5. Hey Greg - thanks, we should talk because I always absorb whatever I can from whomever I can. In the short run, the electronics design is being done by a guy who, among many other things, has designed and built some of the most insanely complex and robust multi-tiered custom systems, rides, attractions, etc. that run in all the Disneyworld/land parks worldwide. But yes let's talk!
     
  6. Hi Mike, my son Aaron is the one you really want to talk to. He's on vacation right now and should be back tomorrow so maybe sometime later this week might be good.

    My gut feeling is if you have a design guy already that's great. What we might have that could be of most interest, is the ability to build the boards on small runs (tens to hundreds). We've had trouble in the past with outside designs though if they don't use Altium Designer and adhere to proper design procedures such as proper pad sizes and such. I run the pick n place and the layouts have to just perfect or the components can tombstone (tip on end) or end up with poor connections. Everything has to be just right. For this reason, we do mostly in-house designs only these days, but if you're designer discussed requirements and build issues with Aaron, it might be something that could make sense for both of us. Let me talk with Aaron first and then I'll e-mail you his ph number so maybe we can set up a Skype meeting or conference call.
     
  7. I'm very interested in this project. I've been looking for a decent weighted controller for years. Fatar was OK when they made their keybeds in Italy (SL-1100 and the Doepfer everyone at RC uses.) But now they are terrible.

    The best keybeds I've found are from Kawai, used in their digital pianos. They came out with a controller keyboard, but didn't add a mod wheel so it was worthless. But they have the legit piano feel, that satisfying "knock" that pushes through to the key.

    I like the traditional pitch and mod wheels. Everyone messes with this design but the big "original" from the Prophet 5, DX-7, Wavestation, etc. are my favorites. But it's your party so build them how you like. One suggestion would be to make them a bit bigger. I think you get better control with a larger throw, like a big versus small fader.

    One suggestion to your design would be to move those controller faders over to the left. I tend to play with my right hand and ride CC1 and 11 with my left. Having them in the center of the keyboard is not as nice as putting them far left.

    The Disney engineers sound great but it's a whole 'nother thing to build with manufacturing in mind. The Imagineers are probably more accustomed to 1-off prototypes, versus building something that can be mass produced. You ight have them do the first pass, then take it to some factory engineers who can make small changes to save cost. Contact G&J Manufacturing in Altadena, they could probably build this. I can get you their contact info if you need it.

    Last suggestion, sell some replacement leather pads so we can swap them out when they wear out. Looks great, I can't wait to see more.
     
  8. Actually, our engineers are merely Disney contractors; they build mass-produced things all the time. But that said, this won't be "mass produced." That's not the goal; the goal is for it to be the best; to be worth the money. That alone doesn't guarantee people will pay for it, but I'm not interested in being in the business of producing controllers, merely for offering the best one I possibly could. I asked about price to see if people were generally thinking what I thought in terms of value, and they were. I'm not going to make it expensive for its own sake, but any choice which comes down to measurable, obvious quality versus cost will come down to quality. The action itself is where the energy is going right now, and that's the key.
     
  9. As far as price, in my experience, under $2000 is where you want to be for success. Over $2000 slows things down considerably, but could work for the composers you're aiming for. Over $3000 will be tough to sell any. When I was at Alesis, we tried very hard to keep the Andromeda under $2000 but I think it ended up at $2499. It still did better than expected.

    Even at $1999, most people would tell you that there's no market. It's hard to find a controller selling for more than $499, the Native Instruments one and the Kawai are the only ones I can think of. But I think there's a market there at the right price. There are many MANY guitars that sell for north of $2000, why not a controller keyboard?

    I know this is a pet project and you're not too worried about sales, but you don't want to be stuck with a garage full of parts you can't use either. I think Kickstarter is great because people have to commit before you order anything. You will see hundreds of people telling you this is great but never buy one.
     
  10. #11 Sam Miller, Sep 21, 2017
    Last edited: Sep 24, 2017
    This is a really intriguing idea. I love the idea of doubling the rear portion of the keyboard as a platform for peripherals like the mouse and keyboard.

    Regarding features, one thing that's missing which I love on my Impulse is the mute/solo buttons at the base of the faders and another button to their side which allows me to switch between the faders affecting volume and affecting MIDI.

    Regarding keybeds, I made a lateral move against that problem and bought a digital piano with MIDI ports - Kawai ES100. I love the key action, but it needs some wheels and faders to be complete. Have you considered having the wheels as a standalone peripheral? It would address the ergonomics issue without closing you off to those who have grown accustomed to their keyboard's action.
     
  11. One problem with many keyboards I've used in the studio is that the mechanical key action is too loud. I often monitor quietly, and run into this problem: the sound of the keyboard interferes with my ability to clearly hear the attack of the sample I'm playing. I'd imagine there would need to be some kind of engineering compromise between great action and silence, but it would be wonderful to have a keyboard that has both to some reasonable degree.
     
    Jason Watts likes this.
  12. Love the concept! My band just bought some Roli Seaboard Blocks which are crazy (but not fitting with your design plan). If I hear you right, key-feel and ergonomics are a big driver for your design. A couple of things I think would help the (virtual) composer crowd:
    1. Mappable & storable Velocity/CC curve presets
    2. X/Y (Z? pressure) pad.
    3. Custom key/range split
    4. Simple DAW controls
    5. Assignable aftertouch/vibrato key wiggle
    Before I jump into a quick rundown of these feature requests, let me say that I know a lot of this can be done in modern DAWS on the back end, but having it saved or accessible on the instrument could speed up the workflow and improve sample playability. I also know there are a ton of standalone iOS/Android or touch screen interfaces that offer some of these features, so it may be redundant or cost prohibitive.
    1. To help match inter-library expressiveness. Playing a library with odd velocity programing? Load your preferred velocity curve & play. CC1 jumps from mf to f at the slightest shift of the Modwheel? Load your preferred CC1 curve.
    2. Not a deal breaker, but a lot of folks combine CC1-CC11, or Volume/Vibrato, and having a multi-CC controller can be useful. If the faders are set to the left like @Jeff Laity mentioned, then somewhere close to where the faders are currently located could work.
    3. Useful for keeping the keyswitches and the playable range close together. Or output to more than one channel.
    4. One cool thing about the current design is keyboard/mouse space, but almost every major DAW accepts MIDI control to initiate key commands (play/punch in/go to marker/next track/etc.)
    5. Could totally throw a monkey wrench in the keybed design, but check out polyphonic aftertouch. It is a trip playing sustains with vibrato at the key level.
    Could scrap all of that and reserve some space for a touchscreen (*not included) with an adjustable mounting surface, then pay a programmer to develop an app that interfaces with the keyboard/DAW. Analog "I can touch it!!" faders/knobs are my preference, but hard to beat the flexibility of a digital solution.

    @Mike Verta - Thanks for dreaming big! Any new updates on the development?
     

Share This Page