Gee, it sure would be nice if I had a workstation set up for playing microtonal music! I can play the music, sort of, with my PC and a conventional MIDI keyboard, but the latter is way less than ideal. When you play 31-note-per-octave music on a conventional 12-notes-per-octave keyboard, a single octave covers two octaves and a fifth. Try playing fast scales with that layout!
So tonight I went scouting around for alternate MIDI controllers.
It’s a pretty discouraging field. Button grids like the monome and Ohm64 have become fairly popular as controllers for Ableton Live. (I recently reviewed the Ohm for Keyboard.) But the buttons tend not to be velocity-sensitive. You can trigger stuff, but you can’t play expressively. Playing expressively, in the world of dance music where Live finds its most natural home, seems to be obsolete, if not actively looked down on. But I digress.
A couple of years ago I reviewed the C-Thru Axis, also for Keyboard. Its buttons sense velocity, but they have a very shallow key dip, which means that it’s hard to control the velocity data output using your fingers. And the layout of MIDI notes is fixed. What’s worse, the fixed assignments are redundant (several keys per MIDI note) in a 12-note-per-octave pattern, which is truly a disaster if you want to play microtonal scales.
C-Thru has recently introduced a stripped-down version, the Axis-49, in which you can give each key a separate MIDI note number. This is an improvement, because it lets you process the MIDI notes through an application like Max to create whatever scale layout you need. But there’s a downside: No mod wheel or assignable knobs, just a bare button grid.
Plus, I’m finding that Pd, a very nice freeware music programming system that I’ve written about in the past, has a fairly grotesque latency, on the order of 100ms, between MIDI in and MIDI out when used in a fast Windows 7 computer. Maybe it’s not quite as nice a program as I thought. (It’s hard to imagine Jim Aikin being too optimistic, but you never know….) I’d rather use Pd to process the output of something like the Axis-49, because it’s embarrassing to try to coax Cycling ’74 into giving me a free license for Max. They’d probably be willing to do it, but I’d feel sleazy asking.
I’m looking at the Starr Labs Z-Board. It might be just what the doctor ordered. Or not. I’m a little nervous about spending $3,000 on a piece of hardware that I’ve never seen, when no one that I know has ever used it.
A keyboard with built-in software smarts, along the lines of the Buchla Thunder, would be super. I’ve always regretted that I didn’t buy a Thunder. These days, it’s far better to do the processing within the computer than to rely on the hardware instrument … but not if the processing introduces 100ms of latency.
Roger Linn has been working on something called the LinnStrument, but it may never go into production, because Amazon bought the company that makes the hardware finger sensor and then pulled the product off the market. Anyway, the LinnStrument interfaces directly with Max; it doesn’t output MIDI at all. And it senses pressure, not velocity, because it has no moving parts.
What would be ideal, I think, would be a tiered set of levers — four or five tiers. They could be somewhat shorter than the levers on a piano keyboard, but they should have the same key dip. I’m not sure what color scheme would be best, but it’s clear that the grouping of black keys in two’s and three’s (common in most alternate controllers) would have to go. Coloring is essential for visual grouping, but the scheme should avoid enforcing any particular bias in favor of one particular scale.
What I’m describing is, of course, the Janko keyboard. It was developed by Paul von Janko in the late 19th century. Many pianos have been built with Janko keyboards; you can see them in a few museums. Bluthner will even build a new one for you, reportedly, for a 7,000-euro surcharge. But I haven’t been able to find a currently manufactured MIDI Janko keyboard. A company in Japan was building one for a while, and may still be, but their website is only in Japanese, and the instrument is apparently some kind of home keyboard, not a MIDI controller.
Jim Aikin's Oblong Blob 
How about the Eigenharp Alpha or Tau? Those keys/buttons look like they have a lot of expressive potential. I’m not too familiar with where they are with software that allows for complete customization of the system, but it seems to have some potential.
I was talking to Roger Linn about this a couple of days ago, and he mentioned that the Eigenharp, unbelievably, can’t play chromatic scales. It’s diatonic. (Unfortunately, my Thunderbird email database has gotten corrupted, so I’ve lost the message. Ooh, what a mess!) I have two other issues with that design. First, I’m not a wind player, and have no desire to have a mouthpiece jabbing my face. Second, the buttons are not arrayed in a tiered fashion, as on a Janko keyboard. They’re on a flat surface. This is very bad if you’re a keyboard player and want to play with any facility.
As I understand it, the limitation to diatonic scales is a current limitation of the default software, but there will be upcoming software of some sort that will give you complete control over all of the key and control info. And use of the breath controller is completely optional. It’s just one of various means of control (others being ribbon controllers and, of course, the velocity, pressure, and two axis (three if you count the initial depression) key motion).
I guess I’ll have to betray my attitude problem here. (I’m having a bad morning because Thunderbird’s database got corrupted; I’m not usually this frank.)
Anyone who would design and build a musical instrument without considering that musicians might want to play chromatic scales is a moron. Or … well, I guess the harmonica only plays diatonic scales, as does the harp. But the harmonica is a wheezy old folk instrument, and the harp has a very slick system for retuning on the fly. (Even so, retuning the harp can create some awkward moments.) Possibly, the Eigenharp was designed and built by a harmonica player. But given the existence of this glaring problem, it’s more than likely that even after it’s fixed, other equally bizarre problems will remain.
No, thanks.
Hey Jim, I wasn’t trying to sell it to you. I’ve never even seen one in person. Only pointing out that there was another production alternate controller with substantial expressive potential. Maybe it totally sucks and always will. Or maybe there’ll be software that will allow it to be turned into something brilliant. We’ll see.