Keith McMillen has been creating new types of instruments for decades.
In 1979, he founded Zeta Music, and created new guitar effects, the first programmable MIDI audio mixer, and modern violins for the likes of Jean-Luc Ponty and Laurie Anderson. He’s also founded BEAM, a non-profit dedicated to promoting music that explores new technologies.
But McMillen is best known as the head of Keith McMillen Instruments, a company that’s created a wide variety of expressive MIDI controllers, like the QuNeo pad controller and the QuNexus keyboard controller; and the unique K-Mix audio interface.
Their latest MIDI controller, the K-Board Pro 4, combines the expressive continuous control of KMI’s earlier designs with a traditional keyboard layout.
This interview is one in a series, produced in collaboration with Darwin Grosse of the Art + Music + Technology podcast, talking to people in the world of expressive new musical instruments. Previous interviews in this series have featured Roger Linn, Geert Bevin, Lippold Haken and Jordan Rudess.
In this interview, Keith McMillen shares his views on why he’s not interested in creating ‘easy-to-use’ instruments – but, instead, more interested in making instruments that bring new power to musicians. You can listen to the audio version of the interview below or on the A+M+T site.
Darwin Grosse: Keith, thank you so much for taking time out of your schedule to talk with us. This series of interviews on expressive controllers has been fascinating. We’re getting some really interesting voices, and I’m really glad to be able to add yours to the list.
Why don’t we start by talking about some of the advances in your product line and some of the directions that you’re seeing your team go?
Keith McMillen: Well, we’re determined to make every traditional music playing mode capable of communicating in subtle and wonderful ways with computers.
I started this 30 plus years ago with the violin, even before MIDI. I built violins that would talk to multi-voiced analog synths that I’d built. It just always felt unfair that the people who hid things, keyboardists, percussionists, were able to participate in the synthesizer revolution and all of the other great tools that we now have on our laptops and iPads.
We’ve been focused on getting all these instruments capable of communicating. Since we last talked, we introduced the BopPad, which is a percussion pad. It looks like a practice pad, but has our smart fabric censors underneath.
It allows you to play pretty much any style. You can play with mallets, fingers, sticks, and of course, depending upon the sounds you’ve selected, you can be a tabla player or play a timpani. Because of the knowledge it has – in terms of where you’re striking, how close to the center, how far away, continuous pressure – you can get all kinds of great subtlety out of synthesis.
Something I realized not too long ago is, as we’re making this transition, trying to get all of the basic archetypal musical forms of play, that some people lost things once they plugged into a computer.
A drummer is a really good example, I think. Drummers are used to getting a lot of subtlety out of their acoustic drums and have a very intimate relationship with it. They can use their thumbs to stretch the skin before they hit it. They had all of this subtlety. Then, once they plugged in with a drum controller, they lost it, most of it. It was reduced to the on and off basics that is the foundation and entry point for a MIDI instrument.
It’s really wonderful to watch these people light up when they’re able, now, to play with all the subtlety and nuance on an electronic instrument that they have spent their 10,000 hours learning on acoustic instruments. Yeah, it’s interesting that some people, like guitarists, have certainly lost out.
Keyboard players, on the other hand, never really had an intimate relationship with the string or the pipe. For them, the move to synthesis control didn’t surrender much. Initially, it surrendered polyphony and we soon got that fixed.
That’s what has caused us in the area of the keyboard to have to go beyond what was originally capable with the acoustic instrument because they really didn’t have any subtle control of pitch bend, timbre, and things that are pretty basic to having gestures turn into music.
With the K-Board Pro 4, which is a four-octave keyboard with traditional size keys, you can just walk up to it and start playing it. And a lot of people, within a minute or two, get it. They start accenting notes in a phrase. And using the ability to bend a note, with the chord held with their left hand not bending.
Basically, it’s like they tell you when you learn how to drive, keep both hands on the wheel. You can keep both hands on the keys now, because it’s configured as a traditional sized keyboard. There’s very little transition time. Once people learn that the front of the key is the horn and the back of the key is a string, left, right is pitch, up, down is brightness, they’re off and running. It’s really been great to see musicians light up.
I know others have been promoting keyboards of a higher dimensional nature and I think it’s really great that the community is aware of this and that musicians now have the ability to access all this subtlety.
Darwin Grosse: Sure. It must be interesting for you when you are test driving. Especially, I think of with the BopPad, it must be really pretty fascinating when you’re first putting this together, and you first bring in people to test it out, and you see people be able to revisit percussion techniques that they haven’t been able to use for a decade because they started focusing on pads and had to relinquish things like surface tension or drag functions that you could actually now reintroduce into the electronic sphere again.
Keith McMillen: It’s really great. We missed a lot of smiles as people first started playing. We did catch a lot of wonderful moments in some of the videos on the site. Just being able to do a controlled press roll and as you move the sticks towards the center and further away to have the timbre change, to not have it mess up the rhythmic quality of what is a very hard to do drum technique, to make it sound beautiful and under control.
To see those people ‘come back’ is fantastic.
Darwin Grosse: In the course of talking to people about these expressive controllers, one of the things people often decry is that MIDI keyboards have devolved into being no more than on/off switches. Now, with these new tools, we’re reimplementing movement that requires some virtuosity in order to make it not sound like a train wreck, right?
Do you find people are initially uncomfortable with it, or is it just such a fantastic new set of options that they dive right in?
Keith McMillen: That’s a good question. It really is a broad range of reaction from the range of people that have been able to have interactions with on the K-Board.
For some people, it’s like it’s always physically been there. They start playing and then they start bending notes, because they’ve heard this. They’ve heard guitar players bend notes, and they’ve heard timbres change over time.
It’s not like a foreign language – they just never have been able to produce those things.
A lot of people, within a few minutes, are off and running. Usually, they have to stop, laugh, clap their hands, and wonder if it’s really happening. On the Kickstarter site for this project, we actually captured four peoples very first playing experience and it’s fun to watch.
K-Board Pro 4 First Impressions:
Keith McMillen: These are people, obviously, though, who are more open.
They’re not raised only on Bach fugues, they have some awareness of what synthesizers can do and keyboardists, really, we’re the first ones in, since the early Moog gear had a keyboard attached to it. They had a bend wheel and they had pedals.
A lot of people have very short learning curves, in terms of adding this new functionality to their voice. Then you get people who have played certain things, certain ways, all their life and they’ll dip their toe into bending two notes opposite directions for the first time. Then, the training takes over and then they’re just playing.
That’s also good. If someone who’s a very accomplished keyboardist can walk up to the K-Board – which is not normal – and just start playing their repertoire, without any transition, I’m pleased with that.
Many of the attempts to capture 3-Dimensional gestures from a keyboard have been out there. You read where it takes three months to play in tune. I think that’s not good. You get both types. Seeing the person light up, kind of like running into an old friend, and really playing phrases with articulation and a new spirit that they haven’t had before is a real treat.
Darwin Grosse: It’s interesting that you use the phrase ‘coming back to an old friend’, because one of the things I’ve noticed is that, since the keyboard is the main entry port into doing MIDI data entry, people from all kinds of backgrounds end up banging away on keyboards, just because that’s the easiest way to do it.
One of the things I notice is when you see a guitar player using a MIDI keyboard enter stuff, they still sort of naturally apply finger vibrato while they’re playing, even though they know full well that it’s not doing anything.
There’s something about that that gives them the innate sense of playing the note, rather than pressing a button.
Keith McMillen: It’s like ‘phantom limb syndrome’. It used to be there!
Darwin Grosse: It used to be there – I can still feel it!
Now you talk about making a broad range of instruments available for interaction with the computer. What are the things that you’re doing for other types of musicians? You mentioned guitar players already as having some missing pieces to their puzzle, then you have string and wind players How do you imagine bringing them all into the game? Is this part of your grand master plan for the future?
Keith McMillen: Yeah. It’s like everyone gets to get on the bus.
There’s really no reason that, just because you play a bowed string instrument. that you should not have access to this incredibly fantastic technology – what you can get today for 30 bucks on an iPad, people would have jump out a window for 20 years ago.
Everyone should be able to access these tools.
Some of them are harder than others. Bowed strings and pluck strings are, in my opinion, the hardest and they’re the ones that I’ve been working on the longest. The Zeta Violin is still being used, still being produced, and used for basic MIDI entry.
I have some designs that will get much more timbre and subtly out of it. That’s been one of the problems with string instruments, bowed or plucked, guitar, violin, is that you have to abandon that nuance.
Darwin Grosse: Right.
Keith McMillen: The first violinist with Hong Kong Philharmonic was over and he was playing a Zeta Violin into a MIDI synthesizer.
The guy is a phenomenal player and I really had to say, “Okay. Think like this. When you play that note, someone on the other side of the room has seen you play that note and pushes a doorbell button.”
That’s the level of intimacy string players have over MIDI.
The new StringPort system, which has fret scanning, which is always been, in my book, a needed component. I had the Mirror Six synths out 20 plus years ago and this latency thing is not acceptable.
For a lot of play, you put your left hand down before you pluck the notes, so that information gives me everything I need even before the string vibrates. Then, taking the timbre of the string, the brightness, and just the volume, and mapping that to the synthesizer, which is what’s happening with MPE-defined mappings.
All of a sudden, people start playing like a guitar as opposed to a device that enters notes into a computer.
Darwin Grosse: Yeah, my first and foremost instrument always was guitar and I was a early user of the old Roland-GR systems. This idea of having a guy across the room pushing a button, that resonates so clearly with the feeling of disconnect.
The question I have is, if you’re doing fret sensing, you’re doing a bunch of other stuff, is there going to be a way to do that’s practical for an instrument that doesn’t require the transatlantic cable coming out of the backend in order to get it into the computer.
Keith McMillen: I certainly hope so.
There are issues, because it’s a lot of data and you have to send all the audio and Bluetooth, at it’s best, has about 10 to 15 milliseconds of delay.
And WiFi is highly variable, so you get jitter. There was ultra-wideband that was proposed four or five years ago, which was, essentially, wireless USB. That held a lot of promise, but they could never make it work.
I keep my eye on everything and it’s still not cost effective and/or convenient with the wireless systems that our computers have in them at this minute and with the transceivers that are readily available.
With WiFi and Bluetooth, having 15 or 20 milliseconds of delay in a conversation is not horrible. But, if you’re a musician, that kind of delay becomes a problem.
I would love very much….to free ourselves (from cables). StringPort right now is a USB two wire. Like I said, I really have my fingers crossed for wireless USB, because I think they will do it right.
Darwin Grosse: Now, in addition to your work making controllers with different kinds of expressiveness, one of the other things that you’ve long been involved in, is real-time group collaboration and interaction.
I’ve seen your work with your modern performance group TrioMetrik. It’s clear that you have a vision for this – not only coming from the individual instrumentalist and their virtuosity with their instrument, but this vision for a group of musicians in a collaborative mode, sharing information with and react to information from other people. With networking, these devices could open up new doors for collaboration or new doors for interaction.
Can you talk a little bit about how you see that working and what your experiences with TrioMetrik were that give you hope for that?
Keith McMillen: If you look back the history of western music, there have been several major improvements, changes, that are usually the result of a technology getting refined – by being able to have a piano, the strings don’t break, building five flutes that all play in tune together – the things that allow symphonies to work.
I feel that every generation – generation could be anywhere from 20 to 100 years – produces a new technology that becomes the basis for a new music. We haven’t quite done that yet. You go listen to some EDM music, which is really fun to listen to and dance to, but when you leave, it’s kind of impossible to even whistle a melody that you heard.
There’s no synthetic soloing, yet, that would be on par with a violin solo or a guitar solo. Electronics has made us more insular in that, if you want to perform a piece, you need to be composer or programmer and the performer. Where in the past, those functions can be done by three different people.
I think if we take our instruments and hook them to the most important technology of this century, which is networking – that is the new ocean. That’s where fun will be.
I’ve been keeping this as a theme, a guiding light, in my work and I was able to get a version of it up and going about 10 years ago, after about 20 years of programming and modifying and improving a bunch of instruments. Basically, if I can solve the problem, I’ll do it, and if I can’t, I’ll start a company.
It was fantastic! We had people writing pieces for us – the ability of one instrument to change the timbre of another, for the score to sit there and hover until the bass player plays a specific line – it becomes almost like a video game, in that you’re interacting with the computer and you’re interacting with other players, but it’s music.
Usually there’s a score involved, there’s still notes, but you get to do all of these other things and as our computers get more and more capable, I think they’ll become stronger players.
People such as Ray Kurzweil and others speak of the future when a thousand dollar computer will be as smart as a human. They put it off in the 2040’s, so I think that’s a very good target. In order to have this work, you need to be able to get not just the audio, but the gestures and the notes, and the volume, and the timbre, of what a person is doing on that instrument, at a high level form.
Well-modeled, so that there’s a number that relates to how brightly the violinist is playing or how syncopated the guitar player is, phrase-following, so that the computer can respond after you’ve done your part. These are all really challenging technologies and require a lot of computing power, but it’s happening.
Think about voice recognition. Talking to your stupid phone and it actually gets it right. That’s new!
Darwin Grosse: What are the things hampering forward movement on this? I sometimes wonder if one of the things that hamstring advances in music, music technologies and music instruments, is the fact that nobody can ‘own the saxophone’, so nobody wants to put that much intellectual property into the saxophone, right?
Keith McMillen: Right.
Darwin Grosse: Where someone can own Facebook, so they get to pour a lot of effort into turning writing notes to your parents, into a computer paradigm.
Do you think there are things like that, things on the business side, that actually hamper expansion on either the technological or the artistic side?
Keith McMillen: Yes, certainly.
Let’s take that same saxophone, there’s really very few practical ways for a saxophone player to control a synthesizer in a meaningful manner. That’s part of the chicken and egg problem, where if the synth doesn’t give you that intimacy of response that your actual saxophone does, why bother?
If you look at the history, we’ve had a lot of really fantastic composers who wrote pieces for synthesis and/or computers along with western musicians – John Adams, Terry Riley, they’ve all given up. This is another part of what I’m working on.
It’s because, if you write a piece that uses technology, in five years nobody will be able to play it.
That hurts. That’s one of the things that has to change. It’s also one of the main goals of my life – to create this environment that will allow a musical piece involving technology to be playable in 20 or 50 years. That way composers won’t shy away from putting nine months of work into something that’s played once and, at best, you can listen to a recording of it. Composers like their music to have a life.
Darwin Grosse: I think this gets back to where you talked about describing things in more high level terms. That allows compositions to float above a lot of different things. Not only can it float above technological changes, but it can also even float above the idea of the composition of the performing organization.
A Paul Simon song, for example, can be played on the piano, or it can be played on guitar, you can sing it, or you can just play the melody on the top note, top string of the guitar. And it’s always still that Paul Simon song, because the description of the melody, the melody is done in a high level type way.
When you’re doing things that are timbre-oriented, or you’re doing things that take advantage of the kind of glitches or whatever that come out of a computer, all of a sudden, it makes it very difficult for someone to say, “Well, yeah, we don’t have a guitar synthesizer player, but we have someone who could conceivably do this on a KMI 12-Step or whatever.”
How does that translate? It’s not always clear that that can happen.
Having this higher level description and then allowing people to be able to map that to actualities seems like a pretty good play, at least for composers.
Keith McMillen: Yeah. One way that I think about it is that, when computers first came out and you had printers, you had one font. It talked over that fat cable, using something called ASCII.
Then computer technology evolved and we got Postcript (ed., a computer language for drawing vector graphics that made possible desktop publishing). Now, not only can you print an ‘A’ on a page, you can add a drop shadow and make it bold.
I think MIDI as the ASCII of music, in that an ‘A’ is an ‘A’, if you’re lucky, but that nuance has not become part of the normal daily experience. I think being able to play pieces in the future on different instruments, that just makes it normal.
Tonight, I bet I could drive less than half an hour and hear a Beethoven string quartet or a couple of them, and maybe some reductions for piano. It’s lively. Pieces for electronic music, though, generally don’t have a performing life.
If you look at it, the causes are fairly clear. I believe they’re treatable. I can open up a 20-year old webpage, that’s pretty amazing. I think we just need to have that same ability with an electronic music score that subsumes all of the syntheses and processing and allows people to play a piece that’s not necessarily written for their instrument. That’s very good for the mind. There’s a lot that has yet to happen.
Darwin Grosse: Now, when I look over the work that you’ve done over the years, whether it’s the things that you’ve been involved in designing, the instruments that you’ve built as part of companies that you’ve developed, or even the instruments that you choose to use yourself, you have always avoided two things.
First, you’ve avoided designing a new instrument. We talked previously about this – there are reasons that you don’t want to develop a new instrument. You believe in the pedagogy and the availability of teachers and the amount of time that people have already spent on their instruments, right?
Keith McMillen: Absolutely. I solidly stand that there will be no new instruments, just because the environment’s gone from that.
Darwin Grosse: The other thing is that you have steadfastly avoided the ‘easy-to-use’ stuff.
It seems like, right now, the music industry is a little bit awash with easy to use stuff. It’s like I can bang my head on the keyboard and a beautiful melody is going to come squirting out, or I can beat away on my pad and it’ll always be in rhythm, no matter how I hit the thing, right?
Keith McMillen: Yeah.
Darwin Grosse: Even in the things that are not exactly instrument-like, things like the 12-Step, or the SoftStep, or the K-Mix, or the QuNeo, there’s still a lot of expression, a lot of movement, a lot of capability in there. It isn’t like push a button for instant music.
I’m curious about why you’ve chosen that tactic, because, if you look at the market place, it seems like what people really think they want is the easiest to use possible thing. You’ve focused on tools for the accomplished musician. Why have you made that choice?
Keith McMillen: If you look at the instruments that are easy to play – recorder, slide whistle – they just don’t go very far.
A lot of electronic versions of ‘Tickle Me Elmo’ instruments, they’re fun and they’re great for kids, but they end up in a closet.
Musicians who are serious about their art – and most musicians are – they want something that will grow with them, or at least respect their talents. That means it’s not necessarily going to be able to be picked up and played on the first day.
I remember reading an article written a long time ago by Bob Moog, someone asking him, “All those wires, all those knobs, it’s so hard.” He said, “Well, here’s a violin. Let me hear what you can do.”
I don’t necessarily want to recreate the instruments I’ve created. I feel if I can get all my licks in and make them cover the palette of that particular instrumental form, I won’t necessarily have to go back and do it again.
I think that music – as in really good music that persists over time and space – people pick it up and play it who didn’t write it. As you mentioned, if it was written for a flute, I can play it on a guitar. It’s flexible. It’s welcoming.
But yeah, you got to pay a price. You got to learn the language and a lot of people have. That’s my target audience and what I want to offer them is more. I don’t want to take anything away.
I want to add new timbres, new possibilities to those instruments, and then for composers, I want to invite them back into the world of writing for instruments and technology.
There was more going on in the 60s and 70s than since. While it may address no need expressed or felt right now, I do think once it’s available and behaves nicely and people start writing some good pieces for it and other people, who maybe never met the composer, can download all the tools they need and play those pieces, and interpret those pieces. Well, then we’ll have another species of living music. It’s been a long time since that’s happened. I’d say 60s’ with popular music, rock music, was incredible. We really haven’t flipped over, the clock hasn’t flipped over yet to the next one.
Darwin Grosse: What advice would you give to a 10 year-old who’s got the hunger for music? What direction would you send them on?
Would you send them on the direction of here’s an instrument that has a great potential for a future, learn the violin, or the piano, or the drums. Or would you say, “Focus on composition because that’s the gathering point for musicians.” Or would you focus on production as being sort of like the locale where everyone comes together? What advice would you give to somebody if you wanted to help position them for how you see the future?
Keith McMillen: It depends on what the person is looking for.
Music has a large social element – the stories of everyone in high school getting a red guitar so they can pick up girls – and that’s valid.
But if a person wants to express something that’s important to them that music, being the only real-time art out there, and music would address that need – then I think you have to pay your dues.
I do think you can easily start with a ukulele. I think they’re great. There’s the 3-string guitar, the Loog. That’s fun.
I think anything you expose a young person to shouldn’t be so challenging that it makes them give up, but it should at least be an entryway to more serious instruments and more styles that you can choose from and learn from other people. I forget what the number is, but it’s something like 80% of all guitars end up in the closet. Well, that’s okay – 20% of the people are still playing.
Those are the people who take great enjoyment from it and find satisfaction. And because of that, they are willing to put more time into the instrument. I just don’t know that there are shortcuts.
Darwin Grosse: Keith, I want to thank you again for taking the time. This was a fabulous chat, as is the case every time you and I get together and talk.
Keith McMillen: Always a pleasure. Thanks for the time.
Additional Resources:
Keith McMillen has been featured previously in these interviews:
- Keith McMillen on the Arts + Music + Technology Podcast #54
- Keith McMillen On Kickstarting A Music Controller Revolution
- Keith McMillen On The QuNexus And Why New Music Needs New Instruments
Darwin Grosse is the host of the Art + Music + Technology podcast, a series of interviews with artists, musicians and developers working the world of electronic music.
Darwin is the Director of Education and Customer Services at Cycling ’74 and was involved in the development of Max and Max For Live. He also developed the ArdCore Arduino-based synth module as his Masters Project in 2011, helping to pioneer open source/open hardware development in modular synthesis.
Darwin also has an active music career as a performer, producer/engineer and installation artist.
Additional images: KMI, Beam
“I solidly stand that there will be no new instruments, just because the environment’s gone from that.”
Thank you for this sentence, Keith McMillen! I will post it under every article where “new” “instruments” are featured.
One of the reasons that McMillen’s prediction has held true is that most electronic musicians are, sadly, untrained musicians and technological luddites, these days.
Everybody likes to think they’re cutting edge, but what they actually want to buy is cheap knockoffs of 40-year old gear. And they want endless novelty – a new button that they can press that causes a new sound to be spat out.
The most original new instruments get ridiculed because they’re ‘rip offs’, compared to the pricing of disposable mass-produced midi controllers. Do you see trained musicians going for cheapy instruments, though?
The other reason is that musicians that actually can play want to leverage their skill. So they want something familiar – like McMillen’s new keyboard – instead of completely new instruments, like the Eigenharp.
Kudos to people like Buchla, Linn, McMillen, and Haken – all making something that’s actually new and capable. But the ones that are the most likely to succeed will probably be instruments that build on familiar forms.
I don’t know how much he understands of interesting electronic music;
the key point is not to play a lot of notes – a melody you can whistle,
but to play with the timbre.
We are 50 years way past the discussion of “is this music” 🙂
He’s is also talking about replaying pieces from notation,
Surprise me, but if I look at the MPE output that is just to hard to follow with everything that is going on.
If people can play it, why shouldn’t people be able to read it?
I think he makes a very important point here – establishing a language future generations might be able to interpret.
Notes I can read from the sheet,
MPE includes pitchbend, vibrato, velocity, sliding up and down and release velocity and aftertouch per note on top
Good luck with reading and playing all that from sheet in realtime.
It’s not that I don’t like MPE,
but let’s stay on the carpet.
Would be interesting to read an interview with a composer who uses MPE and has others play the stuff (this excludes 99 percent of YouTube demos 😉
You can write MPE music down, but it sure won’t be easy to play prima vista.
its exactly the same as playing a violin… do they put in notation for vibrato and emotion and all that? thats up to the player, not the sheet music
MPE just gives you the ability to play the synth with the same amount of expressiveness as a traditional physical instrument – thats the point of it
No it isn’t. You don’t play more than 2 strings at a time on a violin.
btw. A piano is a traditional physical instrument too. 😉
So, play a chord and wiggle just one finger to do vibrato
There is no way to notate this that I know of.
Slide up and down to mix and unmix to layers of sound …
Stuff gets complicated with 10 fingers and 5d touch.
No, it won’t get much more complicated. Sheet music is no midi, MOST of expression is left for the player to figure out and interpret. It won’t be much different with MPE – just a few more symbols to remember.
that has nothing to do with music notation
you asked how MPE would be notated… and my point is that it has nothing to do with notation
just like the individual aspects of virtuosity are not notated for any other instrument
If you can’t notate it you can’t replay it without having heard or seen it.
That’s one of the points he is going on about. (Composers like to be played, playing Beethoven)
Im sure we can agree that you can do a lot more essential meaningful stuff that goes far beyond – do as you please with mpe, because it drastically changes the music.
you still dont get it
oh well, good luck with all that
I don’t think you saw the roli mpe demos.