A series of three polyphonic analog synthesizer models produced by Roland in the 1980s. All three have the same basic voice architecture, with one DCO having two waveforms and a suboscillator, an OTA lowpass VCF, a high pass filter (not voltage controlled), a VCA, and an LFO. They also had, as a global effect, a built-in version of Roland's excellent analog chorus circuit. All had six voices, as indicated by the model numbers.
The first of these, the Juno-6, was introduced in 1982. Aimed at a price point far below the high-end Jupiter-8, it was only a moderate success because it lacked patch memory. Additionally, it did not have MIDI, being that the initial specification was still being developed at the time, and anyway, the Juno-6 did not have microprocessor control over the patch parameters. The Juno-6 was Roland's first synth to incorporate DCOs in place of analog VCOs.
The user interface design followed the trend that had been set by Roland's Jupiter-6 and -8 high-end synths. Nearly all patch controls were in a single row across the panel, with all continuous parameters being controlled by sliders. Unlike the later models, the high pass filter was continuously variable (on the Juno-60 and 106, it has four discrete settings). Two other features not found on the later models were the ability to route the envelope generator to pulse width modulation, and an external control voltage input jack for the VCF cutoff frequency. The chorus had two rate settings plus "off", but users soon discovered that a third setting was obtainable by pressing the '1' and '2' buttons simultaneously.
The Juno-6 had a built-in arpeggiator, with up, down, and alternating up/down patterns. It could be clocked via an internal oscillator or via an external clock input jack on the rear panel. A five-octave, non-velocity or aftertouch sensitive, C-to-C keyboard was incorporated. Performance controls to the left of the keyboard consisted of a Roland pitch stick, which was routeable to both DCO and VCF frequencies; a separate button for modulation (on/off only), and an octave shift switch (in addition to the keyboard transpose controls, which allowed 1-12 half steps in either direction).
The Juno-6 did not compete in the market well with the Korg Polysix, which had patch memory. It was almost immediately superseded by the Juno-60, which corrected this fault; it had 56 patch memory locations. It also came with the Roland DCB control bus, a precursor to MIDI. (Later, a DCB-to-MIDI adapter extended MIDI to this synth.) The -60 retained the -6's arpeggiator, keyboard, and performance controls, and the basic voice circuitry was the same as the -6.
The main addition was the patch memory, which memorized all of the voice parameters (but not the arpeggiator parameters). A red border on the panel indicated which controls were remembered. The basic panel layout of the Juno-6 was retained, with the patch memory controls added to the right. The memory also included a cassette interface. 56 available patch memory locations were indicated on the panel, but other "hidden" ones were accessible by pressing (undocumented) combinations of patch selection buttons. The Juno-60 was one of the first synths to be equipped with a test mode, which the technician could use to calibrate the circuitry and diagnose various circuit faults.
In 1984, the -60 was superseded by the Juno-106, which increased patch memory to 128 slots. The other major addition was MIDI; although this was not the first Roland synth to come with MIDI from the factory, its MIDI implementation was far more extensive than almost any other synth on the market at that time. It included a facility for remote editing of patch parameters via sysex and transmission of movements of the programming controls via same. This made the 106 one of the first remotely programmable synths on the market. Note that there are a few aspects in which the MIDI interface does not conform to current standards; this is due to the incomplete nature of the standard at the time the -106 was designed. The user should also note that the -106 boots up in omni mode; selecting a MIDI channel number takes it out of omni mode.
To get the price and number of components down, Roland took the voice architecture of the 60 and miniaturized it for the 106. They bought unpackaged ICs (the bare chips) from the chip vendors and encapsulated them into two custom hybrid IC designs, one containing most of the DCO circuit, the other containing the VCF and VCA. Unfortunately, the latter (the infamous 80017A) had a design fault that caused the ICs to gradually fail over time, a situation that 106 owners today are still having to deal with. Additionally, the LFO and envelope generators, which has been in hardware on the previous Junos, became microprocessor-generated software functions. These measures did get the list price of the 106 down to less than half of the all-conquering Yamaha DX-7, which made the 106 one of the few analog synths to go head-to-head in the market against the DX-7 and become a hit; Roland sold over 40,000 units.
The -106 lost the arpeggiator of the previous models, but it gained another interesting feature: polyphonic portamento. To gain some control over this, two different voice allocation algorithms were provided, either of which could be selected via buttons on the panel. The "poly mode 2" setting allowed the performer to control voice assignment by carefully arpgeggiating chords, making the voice assigned to each note, and hence the behavior of the portamento, predictable. A third mode, unison, was of limited value due the phase-synchronized characteristic of the DCOs (they are all timed by the same master oscillator), which produced an unpleasantly buzzy and non-moving sound.
The same keyboard from the earlier Junos was retained, but the performance controls were rearranged. The separate modulation trigger button was replaced by adding a second axis to the pitch stick; pushing the stick away from the performer activated the modulation. (It was still only an on/off function.) The octave select switch was replaced by an on-off switch and rate knob for the portamento. On the panel, the patch selection, cassette interface and MIDI interface controls were moved to a row underneath the parameter sliders. Patch memory was divided into two "bank groups" each of which could be loaded or saved to cassette separately. Each bank group contained 8 banks of 8 patches each. Plugging a foot switch into a rear panel jack allowed the performer to rotate through the 8 patches of the currently selected bank by pressing the switch once for each patch change.
Although the two synths share a voice architecture, cassette patch data dumps are not compatible between the -60 and the -106.
This was the end of the Juno series proper, although the MKS-7 rackmount synth shared the Juno voice architecture. In 1985, Roland superseded the Juno-106 with the Alpha Juno series, which used a completely different voice architecture. The Juno-60 and 106 remain valued by performers and collectors (the 60 slightly more so by some), with the Juno-6 somewhat less highly regarded. Both the -60 and -106 have the useful feature of being able to changes patches instantaneously, a feature seldom found on later synths.