What it is and what it was built for
A 1997 digital synthesizer built to prove physical modelling could be polyphonic, multitimbral, and performable. It extends Korg’s MOSS system — introduced in the 1995 Prophecy — into a full keyboard instrument with performance controls and expanded modelling engines derived from OASYS.
How it works, in terms a player would use
It uses physical modelling synthesis, not samples or analog circuits. Each voice runs two physically modelled oscillators, a sub-oscillator, and a noise generator. You shape sound with parameters tied to physical behaviour — string tension, tube length, reed hardness — not abstract filter cutoffs. The XY touchpad moves two real-time parameters at once. The 61-note keyboard gives velocity sensitivity and aftertouch. The polyphonic arpeggiator triggers across all 12 voices simultaneously.
What it does better than anything else
It was the world’s first multitimbral physical modelling synthesizer. Nothing else in 1997 let you run six different physical models — a plucked string, a blown flute, a struck metal bar, a bowed violin, a resonant vocal tract, a rubbery synth bass — all at once, each responding to aftertouch and XY motion.
What it does badly
It cannot layer more than 12 voices. It has no built-in effects beyond what MOSS engines include. It offers no sampling, no wavetable scanning, no granular control. Its interface relies on menu diving and soft buttons, not dedicated knobs for every parameter. There is no sequencer or audio recording.
What people made with it
The sources do not state what people made with it.
Who should use it, and who should not
A player who values tactile physical modelling over convenience or modern workflow. A collector or studio builder who wants the first instrument to deliver resonant, dynamic, non-repetitive tones from algorithmic synthesis — not playback. Not for producers who rely on recallable plugin chains or instant timbral variation.
Is it worth your time
Yes — if you want hands-on physical modelling in a pre-VST era instrument that still behaves like hardware. No — if you need deep sample manipulation, seamless DAW integration, or more than 12 voices. It does one thing first, and does it without compromise: multitimbral physical modelling in 1997.
