Roger Linn LinnStrument + Dave Smith Prophet 12

What happens when two electronic instrument design masters, Dave Smith & Roger Linn, get together?


Roger dropped by the office to give us a sneak peek of his great LinnStrument prototype. The LinnStrument is triggering our Prophet 12 Module prototype as the sound source.

This is not the final design or form factor of the LinnStrument, but it was fun to see it in action and play around with it. It is extremely intuitive and expressive.

See Linn’s site for more info on the LinnStrument.

19 thoughts on “Roger Linn LinnStrument + Dave Smith Prophet 12

  1. Now I’m starting to like it.

    Make a sling over version so I can play it like a chapman stick and I’ll have a conversation with my bank account!

    (Or I’ll install it myself)

  2. These guys are the masters – the yardstick by which other designers are measured.

    The LinnStrument looks like it’s going really cool!

  3. I’m impressed about how the masters of the past (creators of no less than the MIDI standard, the Prophet-5, the Prophet VS, the Linn LM1 and the LinnDrum) are still at the bleeding edge of awesomeness more than three decades later.

    Now have Tom Oberheim kick in and we’ve got the living synth pantheon!

    1. Because it’s rectangular and has square buttons in a grid … hmmm? Have you done any research about the sensor technology and about what the user interaction model is like? They’re both extremely different and will play very differently also.

    2. The soundplane is indeed a cool controller. There is a design doc that describes how Soundplane works physically. Think: little moving squares that change resistance in the corners that they are mounted on, and get signal processed to find the current surface shape. Linnstrument is based on resistive touchscreens (ie: versus capacitance touchscreens like iPad).

      There are a bunch of requirements to be balanced:

      The x,y,z,w positions all need to meet basic constraints:
      1) accuracy
      2) precision
      3) latency
      4) throughput
      5) accurate touch separation (ie: blobbing of fingers together when too close)

      x,y are the vertical and horizontal
      z,w are actual finger pressure, and a fingertip area sensing
      high throughput may be required to track all 10 fingers at the highest sampling rates for these parameters. on an Apple macbook trackpad, it is even possible to track the orientation of the finger area oval (I have seen the code for it), such that you can detect another axis (u axis?).

      Such a controller (an ideal controller IMHO) would probably smoke a violin in terms of accurate expressiveness with at least 4 axes of expression per fingertip.

      An iPad (ie: capacitance screen) has really awesome precision/accuracy in the x/y plane, and no z at all. A sound plane has problems with touch separation (as does a Continuum), but it definitely has awesome latency/throughput characteristics.

      I would imagine that Linnstrument, being based on resistive screens, comes out well in the various tradeoffs. Roger had seemed to imply in earlier videos that since he can get pressure samples at about 1hz rate on one of his prototypes that you don’t even need to do anything about an artificial ADSR envelope…there is a pretty high-def envelope coming directly from your fingers.

      And this is just thinking of it as nothing more than a controller that emits OSC or MIDI signals. The MIDI signalling has to be ridiculously complex to accurately represent a 3D continuous controller.

      You then have to deal with the brain processing the messages (ideally a custom OSC protocol) in real-time and deal with sample-accurate timestamping of the gestures to that there is zero jitter for minimum latency.

      1. …and the other constraint…that precision and accuracy of the x axis (which will generally be pitch, but y might be part of pitch if you emulate guitar string bends)…

        if the instrument is to be an octave wide (or a fifth, or whatever)… to play accurate fretlessly, the precision of x,y needs to let you pick the exact pitch within +/-5 cents; ideally +/-1 cent. (A cent is 1/100 of a semitone). It is easy to hear the difference between a Pythagorean third interval versus a Just one, which are about a fifth of a semitone apart. When playing the same note in different rows, you can easily hear being off by 1 or 2 cents; where you get a chorusing sound that is unpleasant if you can’t precisely control it.

      2. oops.. i meant 1khz of touch updates. ๐Ÿ˜‰ 1000 pressure updates per second is the thing that majorly differentiates it from ipads, which are good in (almost) every other way. on that note, ipad latency for touches probably cant match what Linnstrument can do.

    3. They are very similar(although Linn has changeable note layout overlays), but I’m not sure who got the idea first.

      As a customer, though, I want to see the price first, and THEN decide which one is a rip off.

  4. Yeah, totally…because Roger Linn and Dave Smith just sit around dreaming of exotic instruments and playing with monolithic prototypes for fun. That totally happens…oh wait…

Leave a Reply