Octopus transmute!

I can’t in good conscience fail to mention the JUNO-60 video uploaded to the Roland How Do You Juno contest. The work of UTM, you have love that (a) it’s a video of the legendary JUNO-60, the original, analog JUNO, and (b) all those gorgeous flying imaginary graphics. Clarification: I should say that the JUNO has an all-analog signal chain. That is, the oscillators are digitally-clocked DCOs and get digital patch storage, but everything else is analog. So it’s more analog than the JUNOs sold by Roland now. And by “original,” yes, the 60 was an update of the JUNO-6.

Yeah, that’s what we’d label the parameters, too, given complete freedom.

From YouTube:

This is my entry in the How Do You JUNO? YouTube™ Video Contest. All audio was created and performed on my quarter-century-old, pre-MIDI, analog Juno-60 synth. Computer Museum Photo: Scott Beale/Laughing Squid

UTM says he’s a CDM reader, as well, so additional bonus points for that.

Deep thought: who wants to build a CV to OSC converter, and we can really pretend like MIDI never happened? (Apologies, Dave Smith.)

See also Robbie Ryan’s JUNO song. Like, with lyrics.

  • Amanda


  • gwenhwyfaer

    Would that be the "legendary, all-analogue" Juno 60 that Roland were so keen to advertise as a "digital synth" because it was one of the first to use DCOs…? *ahem*

  • Ha, yep, the same. How times change. 🙂

    I sure as heck am not going to argue against patch storage (cough, Moog Old School).

  • Pingback: Das Kraftfuttermischwerk » This is how I JUNO()

  • Andrew Turley

    // I don't have an Arduino to test this on.
    // But it should be close to working.
    // Writes OSC messages out the serial port.
    // Hook your CV to analog pin 0.
    // This is probably a little noisy.
    // You might want to drop some incoming bits.
    // Also, packet size is not included.
    // So it isn't really OSC compliant.
    // Tom Duff would think I'm dumb. Sorry.
    int cvPin = 0;
    int cvValue = 0;
    int oldCvValue = 0;

    void setup() {

    void loop() {
    // read the CV value
    cvValue = analogRead(cvPin);

    // only write the value if it has changed
    if (cvValue != oldCvValue) {
    Serial.write(0xFF & (cvValue >> 24), BYTE);
    Serial.write(0xFF & (cvValue >> 16), BYTE);
    Serial.write(0xFF & (cvValue >> 8), BYTE);
    Serial.write(0xFF & cvValue, BYTE);
    oldCvValue = cvValue;

  • Jaime Munarriz

    I just got an arduino, i'll try it on my MS20.
    But I remember korg CV use a different standard, V/octave, or something… conventional converters dont't work on korgs… but sure some noises will pop!

  • Too Brilliant! I MUST get a synth with a "Baron" switch, that goes from "Louis" to "Bebe"!

  • not a shill

    Yes, the Juno series are all based on DCO's. Also they have patch mem that is stored digitally, and do not forgot the DCB interface. Now the Roland SH-7 is an actual analog made from discrete circuitry, CV based, and all analog. The fact there is a "glide" missing from the Juno 60 shows that the time Roland did not how to achieve that effect using DCO's. I also dont get the statement about OSC to CV.. the only thing CV on a Juno 60 is the CV input for the VCF (which is really cool btw). Get of the OSC tit.. when most of the people on CDM are just using a MONOME as a overpriced bank of switches… how many actual products are people using that send out sweepable OSC values? All those lemur owners? This type of ignorance is amazing in 2009. So many users are using abelton and simply converting their OSC values into 0-127 still.. btw midi can produce more than 128 values yet so many people dont realize that because their own ignorance of MIDI. It is amusing how CDM users/editors use words and terms they dont know anything about.

  • @not a shill:
    Okay, one thing at a time.

    The CV-OSC question was a separate thought. CV *is* cool on the JUNO-60, but let's just imagine we flash back in time to the era of the JUNO-60 and MIDI hasn't been invented yet. Imagine if we could make a smooth transition from control voltage on the analog side to OSC on a networked digital side (essentially, what we have now.)

    Now, as for your apparent criticism of OSC and — what, my ignorance? Readers' ignorance? Not sure to whom you're addressing this, given that MIDI remains lingua franca and if people are ignorant of anything, it's OSC.

    The monome is indeed a bank of switches. (Overpriced — that's another question. Have you priced out low volume production with arrays of buttons and lights plus local hand assembly lately? It adds up pretty quickly.)

    The fact that the monome is only a bank of switches illustrates just how inadequate MIDI is. On the 64, it's just a convenience to have OSC – to be able to, say, refer to a button by row and column instead of a linear value (0-63 or something). No big deal. But as you approach the 256, you're already looking at some awkward assignments. Do you use different channels? Different Control Change messages? The MIDI spec doesn't even really cover what a monome is.

    There's a difference between something that's necessity and something that's a comparatively more elegant, more efficient, or more complete solution to a problem. So, no, you often don't need values more than 0-127. And yes, MIDI is capable of using LSB/MSB messages to send greater resolution when needed. But no, MIDI isn't capable of the resolution of OSC — period. It's also inconvenient to use LSB/MSB arrangements and as you yourself point out, software like Ableton Live also often isn't intelligent enough to know what to do with it. And no, it doesn't have time stamps. Yes, MIDI is theoretically transport independent. No, it's not as efficient as the combination of OSC with zeroconf for auto-discovery — and the OSC spec (unlike MIDI, I might add) continues to evolve and improve.

    Let's take the example of Live. I'm actually not sure what its internal resolution is – I need to look into, as well, what's provided by the API. I don't think it's just 0 to 127, but I could be wrong. Anyway, I'm the last person to say that OSC is useful when you convert back to MIDI. But is that OSC's fault, or the fault of dumbed-down MIDI implementations?

    How many actual products are people using that send and receive OSC values, including sweepable:
    * iPhone
    * iPod touch
    * Google Android
    (if you're still counting, we're now at tens of millions of devices)
    * Computers — all of them
    * Processing
    * Max/MSP/Jitter
    * SuperCollider
    * Pd
    * Various microcontroller platforms for DIY hardware (which should soon lead to available products, not just monome)
    * Most major visual packages: VDMX, Resolume Avenue, ArKaos GrandVJ, Livid, vvvv, Jitter, etc.

    In fact, I'd happily put the installed base of devices that can support UDP and TCP/IP up against the number of hardware devices with MIDI DIN ports.

    Just sayin'. I've heard this argument before — it goes something like "you must be ignorant of MIDI to claim that OSC is superior to MIDI in any way." Believe me, I know PLENTY about MIDI. Have you really looked closely at OSC, or — let's leave OSC out of it — how easy it is to transmit complex data in readable forms over TCP/IP and UDP? No one in their right mind who had ever come from that world would ever conceive MIDI as the best way to transmit and receive data. There just ain't no way.

  • B.C. Thunderthud

    Cool video, funny, good music, but the Propellerheads-logo-generator in the middle of the screen was giving me some brand dissonance.

  • not a shill

    Peter thank you for your response. My main problem with the article is that I understand CDM is being sponsored by Roland, yet just contains information that is both not accurate or relevant in regard to the Juno-60. I know CDM must have editors, it is a blog about DIGITAL MUSIC. Any fact checking would revel that the Juno 6 is the first Juno. It would also revel the Juno 60 is a hybrid synth not a pure analog. I have owned both the 60, and the 106 in addition to many analog roland synths, I am a real fan. I would just hope that mis information is not spread for the sake of shilling or blog spam. I hope the focus of CDM is on music makers and not trendy kids making sequencers from walnuts.. when a RM1X can do a much better job. As far as anything being made that can actually SEND a sweepable OSC value there really is ONLY the lemur, and some cell phone apps. There is many projects such as the MIDI BOX, that can actually make music when the COMPUTER is turned off. Abelton can handle 14 bit midi, you can check in the settings rather than the API, but people using a monome with abelton are really just using to trigger clips, or engage BUTTON presses in MAX or PD. I did not argue that OSC is not better than MIDI in some regards. It should be, its quite a few DECADES newer. However when talking about Juno-60's I dont see how OSC has anything to do with it.

  • <blockquote cite="not a shill">Yes, the Juno series are all based on DCO’s. Also they have patch mem that is stored digitally
    …except the Juno 6, of course.

  • <blockquote cite="not a clue">I hope the focus of CDM is on music makers and not trendy kids making sequencers from walnuts.. when a RM1X can do a much better job.
    Gosh, aren't we sniffy? Well, presumably any manufacturer smaller than Korg should just shut up shop and go home, because we have the big boys to give us all the music gear we could dream of…

    CDM's focus is, as far as I can see, on music makers. It's just that that category is a little wider than you're comfortable with. Deal.

  • @not a shill — sorry, this just needed clarification.

    Even with patch memory and DCOs, I think the JUNO-60 by modern standards counts as an analog synth. Our perspective has, of course, shifted. It shouldn't have used the word "original," probably. (Well, it was an original JUNO-60. Okay, lame excuse.)

    I'm not sure how it's possible for me to "shill" for a keyboard that hasn't been manufactured since the Reagan administration. (I'm also not getting a subsidy from the walnut growers of America, in case you're wondering.) I disclosed the Roland sponsorship. I don't think I crossed an editorial line, but it seems what you're really claiming is that I said something that was inaccurate, and upon re-reading that, I can see that perspective.

    We don't have editors. I'm the editor. And, in a sense, you're the editor – but that's why these conversations are fruitful. It's very possible that in print we could have nitpicked over my copy, but very often you don't get into in-depth conversations like the one we're having. So I do appreciate the corrections; I hope I don't appear that I'm taking it personally, because I do this primarily in order to talk to people.

    As for what you're saying about OSC, it's simply not accurate, unless you mean "currently commercially available." There's an OSC implementation for various microcontrollers (including but not limited to the Arduino), so you can absolutely make hardware that sends sweepable OSC values. The code above shows you how you could do it with control voltage.

    Also, it's just plain wrong to claim that MIDI allows you to work with the computer turned off and OSC doesn't. That's not true. You can use OSC between two pieces of hardware. I think that this is a wide misconception, though.

    You're right, though – this has nothing to do with the JUNO, necessarily, although as I said, if you look back to the time the JUNO-60 was made, there were transformations in play that would not have been intuitive at the time. So the engineers of the JUNO-60 might not have imagined that they'd go to an all-digital architecture, let alone that at a NAMM in the 21st Century a new Moog would use the absence of patch memory or MIDI as a selling point. I don't think people had a clear sense that MIDI was coming. And while MIDI worked shortly thereafter, I don't know that it was the only — or even the best — solution, or that it was seen that way at the time.

    It was off topic here, but … yeah, that "no editors" thing.

    By the way, OSC may be newer than MIDI, but many of the underlying technologies are not. TCP/IP evolved around 1973-4. Serial technology is, in turn, much, much older, but MIDI doesn't conform to what were standard serial implementations even at its inception.

  • UTM

    Hi folks. I'm the one who made the video. I just wanted to point out that though I regularly use my Juno as well as a Moog Prodigy and an Etherwave Theremin, I'm no analog purist (I digitally recorded, processed and mixed all the sounds you hear in that song). I have often wondered why there is no darn Glide on that Juno! I'm really happy someone liked the "Baron" switch. Might as well point out one more thing regarding the analog vs. digital argument: There's a little "solo" of sorts starting around 1:25 that's actually the tape output of the memory banks. Sounds like a modem. That's about when the entirely coincidental Propellerheads logo looking shape appears. That's actually my hands playing the keyboard through a kaleidoscope filter. I thought it looked kinda like a transmuted octopus beak.

  • @B.C.: I think he’s just using a nuclear power mod.

    Also, thanks for getting us back on topic. It’s clear to me people don’t understand OSC’s place in larger history, where it’s going… but yeah, that’ll be a separate post.

  • Sean Drinkwater

    How are the Junos more analog than the Alpha Junos?

  • gwenhwyfaer

    Knobs, darling. 😉

    Seriously, I believe the Juno 6/60 used hardware LFO and EGs, whereas the Alpha Junos generated them in software.

  • @gwenhwyfaer: Yeah, and without splitting hairs, I would say that makes the Alphas a hybrid analog/digital and the JUNO-60… how shall we say… "analog enough"? Given the things supposedly "analog" synths get away with, I think the perspective has changed – and maybe rightfully so.

  • gwenhwyfaer

    Fair enough. I'd actually still class software mods as "analogue enough", myself, in every way that mattered; as far as I'm concerned, it doesn't become a hybrid until there are both digital and analogue elements in the audio signal path (PPG Wave, Ensoniq SQ80, Prophet VS). And even then… what right does, say, a Novation K-Station or an Access Virus have to call itself a "virtual analogue", given that what they actually model is an architecture as hybrid as the Evolver's?

    I think the divisions are meaningless in themselves. Only the sound matters, and a good sound is a good sound whatever makes it; other than to understand why something sounds good, when something almost identical sounds quite undistinguished, any claim of superiority verges into the clouded lands of audiophilic thinking.

  • Pingback: mustakl » Let’s all dump MIDI()

  • everything goes up to eleven. i approve.

  • cubestar

    LOL @ Midi nerd anger!
    "The MIDI spec doesn’t even really cover what a monome is."
    A monome 256 could be notes on 2 midi channels. You could even make velocity/release velocity/pressure assignments.
    My main gripe about the Monome is that it's not available to the general public, except in insanely overpriced auctions. It seems like there would be more "cells" of craftspeople to keep up with demand. I'll be interested to see if the Ohm 64 helps with that.

    I wonder if you could really get Korg/Roland/Akai/M-Audio interested in adding OSC?

    Seems like if the big HW makers switch over, things will flow from there.

  • UTM FTW!!!!!!!