A fully-functioning QuNexus prototype. It may look like the pads on the QuNeo, but Keith McMillen tells us new sensor tech should be more friendly to keyboard technique. And the fact that this is real hardware is important – Kickstarter has recently revised its rules. A look at the new hardware – and actually delivering on Kickstarter – as CDM talks to Keith McMillen.

Can a compact controller not only shrink the conventional music keyboard, but transform it, too? The layout on the just-announced QuNexus is something familiar to keyboard players. But the QuNexus assumes some new ways of playing, with keys that sense pressure and an instrument that you can tilt. Following in the footsteps (fingertaps?) of Keith McMillen’s QuNeo, the QuNexus is built around a custom-engineered set of pressure-detecting, touch-sensitive pads. But whereas previous hardware used USB for MIDI and high-resolution OSC (OpenSoundControl), the QuNexus adds Control Voltage for modular and vintage lovers, too.

The QuNexus returns to Kickstarter for crowd-funding production, a technique that raised quite a lot of money last time – and a desire to ship more quickly has even sent KMI staff to China to check suppliers first-hand – but more on that in a moment. First, a look at the hardware:

25 LED-lit “Smart Sensor” Keys with Pressure, Tilt, and Velocity
Polyphonic Aftertouch
7 Octave range
Pitch Bend Pad
2 CV/Gate Inputs
3 CV/Gate Outputs (16-bit)
Blue and White LED Illumination
14 oz, 3.5″ x 10″ x .5 ”
USB-powered, no drivers needed (Mac, PC, Android and iOS)

At first glance, this all looks just like another QuNeo with a different layout. But the QuNexus, says its makers, features new sensor technology specific to this method of playing, and the addition of Control Voltage could reshape the target audience. I also know that everyone who plays with their QuNeo and SoftStep products is impressed by just how thin, light, and compact this gear is, which could set it apart from some alternatives. The challenge for QuNexus could be competition both from KMI’s own QuNeo (and its more-capacious control layout), and ever-affordable conventional MIDI keyboards. While old-fashioned keyboards are bigger and lack the cooler features here, keyboardists are of course quite comfortable playing them. But there’s some potential here for people wanting something ultra-compact and with some different sensors; Keith McMillen himself talks about what that might be, and you can get a basic feel in the video.

The test of that market is now, as KMI again uses crowd funding to launch the project.

Kickstarter, Take Two

Behind the scenes: KMI shares some imagery of the creation process for their new hardware. Seen here, from top: a 3D-printed enclosure, the circuit board. Photos courtesy KMI.

Keith McMillen today is both announcing the QuNexus and an accompanying Kickstarter campaign. That will likely rekindle a discussion of crowd funding as a model, for KMI and beyond. For all the exploding popularity of Kickstarter, the fundamental model – backers providing money before a product ships – has caused some friction. Specifically, KMI faced backer backlash when production of the QuNeo was heavily delayed by a supplier problem.

DJ Tech Tools published a must-read article on the history of QuNeo’s bumpy crowd-funded ride, in an in-depth story by veteran music tech writer Kylee Swenson:
Creating DJ Hardware With Kickstarter: The Right Move For KMI’s QuNeo?

The tale of that one supplier breakdown is one that hits close to home for me, as CDM collaborates with engineer James Grahame on the MeeBlip. [Full disclosure: CDM is now readying a Kickstarter campaign, unrelated to hardware, and … yes, we’re also a manufacturer.] I liked this quote by Keith in Kylee’s article:

It brings into question the ethics of manufacturing in China. “It’s a land of great expansion,” McMillen says:

“It’s like the Wild West, and you’re gonna run into problems. Some of the stuff that goes into the QuNeo can only be built there because they have the most sophisticated machinery. If I tried to build this stuff in North America, it’s not even possible. So as a nation, we got ourselves into this, and there are pluses and minuses.”

MeeBlip is produced by a Canadian business and a German business, but this relationship to China still holds. And even simple parts like knob caps are often available in flexible quantities, greater varieties, and significantly lower costs in China. Dealing with China means dealing with suppliers over great distances and across language barriers, and quality and reliability can be wildly variable.

In China or locally, suppliers can break down, ship parts with flaws, and ship late: this is simply part of the reality of manufacturing. (I’ve gotten a crash course in that phenomenon with my first two years of involvement in custom manufacturing on MeeBlip. Ask me over a drink some time, or stick around – at some point, I hope we’ll get to talk more about our experience.)

The challenge with Kickstarter is that your relationship to the project isn’t as a buyer, but an investor. When you buy a product, any supply chain wrinkles have been absorbed in advance. Even most preorders will not take your money until they ship. When you’re a Kickstarter backer, you’re investing in a product, which means sharing some of the risk. QuNeo isn’t alone by any means. Kickstarter recently updated its terms of service, including changes like requiring imagery of actual prototypes and not mock-ups, and requiring projects to detail challenges they might face. From their blog:

Kickstarter Is Not a Store

The payoff is, by getting backers from the market for a product, it’s possible to take risks on products that otherwise might be impossible. QuNeo was and is something different, with original engineering from a small maker. You need some sort of capital to do that, regardless; in this case, the investor is you. And investment involves risk.

CDM talked to Keith McMillen himself about the project, to understand his design goals and what he’s learned as he takes Kickstarter for a second go-around.


CDM: Is there a pricing model in place for this?

We are thinking $200 list and ~ $150 street. This is similar model as we did for QuNeo and everyone was happy with it.

The QuNeo Kickstarter campaign was a success from a funding standpoint. But it also came under some criticism for shipping delays – which, in turn, have been an issue with Kickstarter campaigns more generally. How are you responding to new Kickstarter guidelines in terms of deliverables?

Here are the answers we gave to Kickstarter on this exact topic – many address your questions:

This is our second Kickstarter project and here are some of the things we have learned from our past experience, and feel that we have improved for this campaign.
1. We have identified a strong set of vendors for our components. We’ve sent KMI personnel to our Chinese factories in order to supervise and oversee production. These relationships are functioning smoothly and provide us with high quality products on a regular basis.
2. We have structured our reward levels to better track foreign and domestic backers.
3. We’ve established a system for improved communication with backers. This includes more regular updates and better monitoring of messages and comments.
4. This project is far enough along in design and realization where we feel confident we can deliver in a timely manner, even with potential bumps in the road.

Are you able to assure people that you’ll be able to ship product more quickly in this case than with QuNeo?

Yes. The delay for QuNeo came from one component supplier who would not admit they had made faulty product. Since we were talking 350,000 parts they were very slow to respond, and it took us a while to find a new supplier that could source in quantity. We now buy all of our components from approved, known-good vendors, and all of the components in the QuNexus have good second sources. I am confident we will hit our schedule.

You spoke about some of the challenges of hardware vendors in China. But given that unpredictability, does that simply weaken the incentive for doing preorders? That is, if you’re buying an already-manufactured product, then quality and availability concerns remain in the hands of the manufacturer, not the buyer, right? Or, is there an incentive here to fund a product that wouldn’t exist otherwise?

I think there are several incentives to back new projects. We made a lot of decisions on QuNeo’s behavior based on user input and suggestions. Kickstarter supporters are a pretty savvy group, and caused us to include functionality and compatibility we probably would not have been able to consider without their help. Backers get to have input at a really ideal time. The basics are in place so it is not a pie in the sky discussion, but the firmware and support software is not set. People can imagine what they want to do with the new instrument and can communicate that to a receptive design group.

And we are a small company doing original and, I feel, important work in the nascent field of musician – computer interface. Ideally we would be supported by a benevolent monarchy or an eccentric millionaire, but we are doing this out of our personal need for better and more responsive instruments. Kickstarter is a fantastic way for people to express their support and enable new musical modes.

Apart from CV input and output, what’s the advantage of this versus the QuNeo – that is, might some people simply map pitch to the QuNeo and be happier with that, given more controls?

First of all, the sensor pads are different from QuNeo. The QuNexus keys have grip while the QuNeo keys slide. This grip transfers the player’s gesture to the sensors in a more piano like manner. Even though the keys are not conventional, all of the players have said QuNexus plays faster, and with more certainty, than any of their other controllers. It is a different instrument for different purposes.

Secondly, I have made a lot of violins, but never felt that one of them could substitute for a guitar. If you have keyboard skills already or want to learn to play, the QuNexus is satisfying and inviting.

Speaking of QuNeo – what’s the state of this project? At this point, there’s no backlog of hardware? Is demand still high, and do you anticipate it will remain so with this new launch?

Demand is strong and growing. We have shipped several thousand QuNeos after sending out units to our Kickstarter backers. Reviews are coming in and have been positive to rave level in response. From one of the articles: “There’s a new level of control available through this device which we feel is the start of a new era in digital music control.”

Most people have never seen or played a QuNeo. When they do, the take-up is very fast, and people get their lights flipped on when they can use a single gesture and get a complex response. After a few minutes they wouldn’t want it any other way—it feels sensible.

I expect the new Kickstarter campaign to appeal to a wider range of players and shine a light on the possibilities of all of our new generation of musical instruments.

Thanks, Keith. We’ll be watching.

QuNexus Smart Sensor Keyboard Controller


  • Hogo

    Was he joking when he said the quneo was now wireless?

    • johnmoon

      exactly my thoughts.

  • slabman
  • deadwaiter

    Awesome controller – Plus the fact that Matt Hettich is in the video… CalArts grads everywhere…

    • E

      I’d need 2 hands to count the Mills grads in that video…

    • demcanulty

      And hailing all the way from sweet Massachusetts – three Berklee grads, and three MIT grads :)

  • djruv

    It would be better if they focussed on finishing up on QuNeo’s functionality. There are still some very important promised features which still aren’t working.

    • demcanulty

      I don’t want to overstep my bounds as an engineer and not a p.r. person, but I just wanted to let you know that we are absolutely cranking on QuNeo code for an update based partly on our own goals as well as on a comprehensive review of all the requests we’ve received so far. A couple of updates are in the pipeline or they might get balled into one big one soon, plus we are continuing to work on a special release for the high-def roll-your-own sensor-cooking crowd. Definitely rest assured, we are coding and testing as fast as our little fingers can tip-tap. I don’t think one should look at these new items as distractions, but rather as the indication of a healthy heart beat in a small tech company whose aim is to constantly try to sound out and push against the margins of today’s controller gear.

    • http://pkirn.com/ Peter Kirn

      Well, I also have to step in and say this… speaking as someone in the hardware business, I’m surprised people see a project that’s getting funding as a distraction. Generally, funding coming in is good for development. Now, of course, as customers they can certainly be hard on you for things they want. (And will do so anyway.) As a journalist, well – we’ve got a reviewer here now who is himself a developer who I think will have some interesting feedback and questions for you.

      Anyway, engineers welcome here, you know. 😉

    • djruv

      Well, the video shows a prototype QuNexus which is probably built with the hands and time of KMI’s programmers. This while they are still selling a product which doesn’t include a couple of it’s most important marketed features. I’m sorry but I don’t see the logic.

    • tls

      I appreciate that sentiment, and as someone who deals with devices & software, I know how funding is important regardless. I think the issues you see people bringing up as users is that the company doesn’t finish or maintain previous products very well before raising money for the next thing. Being a small hand-to-mouth company as they appear to be, this means the previously ambitious products, like the SoftStep in my example, have pretty serious issues in the real world that are not being addressed, even frankly given lip-service.

      I can see how the Company is operating here, new idea, raise money. Unfortunately they are leaving a swath of really ticked off users who bought the last product and see it being abandoned because of the new idea.

      That’s fine, products are abandoned all the time. Caveat emptor then. But it leave me personally to question why I should support them, when to date, nothing seems out of the Beta phase, then quickly left behind.

      Honestly, I am being harsh with them because of the potential of what they are thinking about. My kingdom for a foot controller like the SoftStep with good software, usable ergonomics and doesn’t require me to shut of one of it’s key features (the backlight) to be usable live.

      People can run their business as they see fit, we can choose to criticize and more importantly, not buy.

    • demcanulty

      Thanks Peter for the welcoming words, and thanks to all others for the responses and the thoughts. In response to tls, I’d love to be able to personally put my name on the line for every single thing I possibly could – past, present, and future, but I can’t go so far because I’m not currently, nor have I worked on these other projects. But I can say that the Qu* projects are of a family and the team that is working on those projects is committed to them all together, and really we couldn’t even be reading and posting here if we weren’t personally vested in trying to make them everything they can be. As Peter implies, all money coming into a small company like ours supports the development of all our products. It’s not possible, as far as I can tell from my limited perspective, to just fall back and stop developing new things as a small company. But I know that the better and the harder I work, the more I support people working on things even unrelated to what I am working on. All of that is what keeps the company running and keeps the development happening. I can’t say I’m not sensitive to complaints about projects that I’m not working on, but I do know that the better I do and the better my work does the more money is available to make all of the other projects better.

    • djruv

      Nice words. But by selling a product without it’s marketed features makes it a little hard to believe the next product full of promising features will actually contain those features. It rather gives the impression the company is just using very ‘smart’ marketing strategies to raise money.
      I think your way of doing business is vey rude (if not illegal) and asking the community to fund this is extremely arrogant.

      From KMI’s website:

      “Class Compliant and Open Source Development Kit

      QuNeo works with USB, MIDI or OSC and will communicate with your favorite music software environments right out of the box. More advanced users and programmers can use the development kit and API to create their own code to respond to QuNeo’s sensor data. Hack away to control your world in ways never before possible!”

      I’ve bought what I thought to be one of the first OSC controllers ever. Turned out KMI is selling an (I must be honest here, very interesting but) ordinary MIDI controller with a Max patch that is able to convert MIDI messages into OSC. I’m sorry for being so harsh but that’s just ridiculous! First of all there is no f*cking use for doing so and second, I can do that with every single MIDI controller I have.
      And what about the “development kit and API”? Does that stand for “With a screwdriver the device is very easily opened.” or something?

    • Dan Overholt

      “I can do that with every single MIDI controller I have.” —- no, actually, you cannot.

      You fail to consider the fact that MIDI-over-USB completely obfuscates the original 31.25kBaud speed-limit of original 5-pin MIDI devices. I’m not sure how fast the firmware of the QuNeo sends MIDI-over-USB data, but it’s a fair guess to say that it’s easily sending it much faster than would have been possible in years past. Then, really, what is the difference between MIDI and OSC? Just a data-formatting thing actually! (as long as the data rate is no longer limited, and the bit-depth of the ADC is not truncated to MIDI’s 7-bit standard – but there are ways around this, eg. sending 14-bit pitch-bend messages or custom MIDI messages).

      This is not to say that there is no point in doing the translation over to OSC or even sending the OSC data as Ethernet Control Model (ECM) which is basically Ethernet-over-USB. This could be pretty slick, actually – as it would bypass the need for any translation programming running in the background on the host machine, by exposing the USB device as a network adapter to the host OS:


    • djruv

      Haha I’m sorry, I should have mentioned I only have MIDI controllers which send MIDI over USB (I assumed that’s kind of obvious in 2012, how could I have been so stupid). I know at least 6 languages which allow me to translate that data into OSC.

      MIDI is a prehistoric protocol, hacking around it to get 14 bit values doesn’t come anywhere near the features of OSC:
      – 32-bit two’s complement signed integers
      – 32-bit IEEE floating point numbers
      – Null-terminated arrays of 8 bit encoded data (C-style strings)
      – Arbitrary sized blob (e.g. audio data, or a video frame)

      And I wasn’t asking for anybody’s opinion on wether I need OSC or not. I wanted to point out the way KMI is doing business is pretty wrong and an article on their new supposedly ground breaking product for which they need funding is a little weird in this context.

      Apparently laws in the state were KMI is based are very flexible but where I come from it’s pretty damn illegal to sell a car which is supposed to be able to fly but in practice lacks that one little detail.

    • http://pkirn.com/ Peter Kirn

      That’s an extreme interpretation. On the QuNeo description, for instance, they say only this: “QuNeo works with USB, MIDI or OSC.”

      “Works with” is accurate. There is absolutely no clear specification for what OSC hardware should provide. Claiming that they’ve not only said something inaccurate, but in a way that’s illegal – claiming this site is then supporting an illegal claim – is a leap I just can’t make.

      If you’re unhappy with KMI as a customer and have a specific grievance, tell us how you’d like their product to work. But that’s another matter.

    • djruv

      I did not say this site is supporting an illegal claim: “I wanted to point out the way KMI is doing business is pretty wrong and an article on their new supposedly ground breaking product for which they need funding is a little weird in this context.” I’m suggesting here that it wouldn’t have been a bad idea for the article to mention somewhere KMI’s previous “ground breaking” product still lacks it’s most “ground breaking” features. I could have sated it more clearly, sorry, I wasn’t out on offending anybody.

      Besides that, claiming it “works with” OSC in the way it does now is the same thing as claiming my flying car example will fly whenever put into an airplane! As I previously mentioned, every single MIDI (over USB) controller “works with” OSC, as well as every car will fly when put into an airplane, which doesn’t make it legal for me to sell it as a flying car.

  • rgb

    To be a practical keyboard for me it needs 3 octaves. I like a lot of the features, esp poly aftertouch, this could be a fun device.

  • sdgsdgsdgsgs

    >2 octaves

    • lel

      >implying the text will ever be green
      >implying anyone here knows what the hell we’re on about

  • tpnga

    They should finish the products they have out. My softStep is nearly useless in any practical way

  • http://twitter.com/tomerbe tom erbe

    april 2013…. sooner please.

  • Jason Duerr

    As a manufacturer of electronics, i hear these horror stories from customers all the time. Making things here doesn’t cost more when you factor in risk.

  • a

    Controllers aren’t much good with out quality software.

    The QuNeo software is anemic and KMI hasn’t demonstrated they can or know how to make good software. With the QuNeo they had all this extra time due to hardware sourcing issues and didn’t put it into making software that takes musicians into consideration.

    KMI support says they won’t do our work for us. No, our work is to play the instrument. Your work is to make software that lets us translate musical thoughts into control as intuitively as possible. Kinda tells you everything you need to know about the company.


    • demcanulty

      Since I’m posting here already in another thread and this has sat over the weekend, I have to say a bunch of us have been scratching our heads over this. I’ve seen KMI support regularly erase computers and do fresh installs of all kinds of software just to figure out how to answer people’s questions about ‘how do I make your x hardware do y function with z software?’. I think they do amazing support, far beyond what I would expect any other company to do. And all the Quneo editor stuff I’ve seen loads up fast and allows you total control over all Quneo sensor outputs. But still, a comment like this makes all of us think and ask each other, maybe something is not enough and we’re failing our users in some way. So please, hit us up, get in touch with support and let us know who you are and what we’re not doing. I know that Keith reads all of the support emails and keeps close tabs on those things, support is one of the things we pride ourselves on doing well to a fault.

    • MattKMI


      My name is Matt and I am a product specialist at KMI. I’d like to know what you think is anemic about QuNeo and the software that we ship with it? Mostly, I’m interested in how you think we could make it more useful for musicians? We’re trying our best to fit both customizability and simplicity into our editors and all supplied software integration, but I’m sure we could still learn a thing or two! Though we’re a small company, we listen to our users. Our products are things that we want to perform with, and we’re interested in making them work as well as possible… Feel free to respond to this comment with your ideas! I’m sure other CDM users would appreciate the ability to join in the discussion.


      Matt @ KMI

  • http://www.facebook.com/tom.byrer Tom Byrer

    There is absolutely no incentive for me to buy early via KickStarter. Their KS price is the same as street, no solid demo of the “Pitch Bend Pad” that I can see, SoftStep has a bad reputation, & I think the QuNeo is better for triggering samples due the XY nature.

    I would have liked RGB LEDs (would be help as a teaching tool, check modes/readouts). I think either a better layout to allow (more or less) seamless octave expansion by placing 2 edge-to-edge, or a XY pad (like from QuNeo) for all-key pitch-bend/modulation is missing. IMHO it would have also made sense to add E# & B# keys for transposing tricks & make the step sequencer better.

    CV & small size in seems interesting, but not enough to put off the purchase until QuNexus 2.0 or I can find it on sale.

  • http://www.facebook.com/rudi.dutschke.967 Rudi Dutschke

    I wouldn’t pay no more than 50 Euros