Latest version (1.8)

Firmware update procedure



  • Rewrote from scratch the band-limited waveform synthesis code used in the analog waveform models (CS-80 saw, waveform morphing, square-saw morphing, sync square, triple saw, triple square). As a result, the sound quality of these models has been improved. The generated waveforms are crisper, with a more consistent amplitude, and less aliasing.
  • Increased the range of the TIMBRE and COLOR settings on the CSAW model. It is still possible to synthesize the CS-80 "sawtooth with a notch" waveform of the previous revisions, but this model now covers a larger range of saw/square hybrid tones.
  • Added a new synthesis model, HARM, which generates a mixture of sine harmonics. COLOR modifies the distribution of the amplitudes of each harmonics, around a central frequency set by TIMBRE.
  • Improved the quality of the Z?PF models when they are played with fast envelopes, by adding linear interpolation to the TIMBRE parameter.
  • Improved the quality of the VFOF model.
  • Removed some of the "clickiness" of the built-in VCA.
  • Improved the DRFT setting, which generates analog-style frequency drifting and detuning. The random process generating the noise has a more realistic behavior; and the amount of frequency drifting can be adjusted (instead of a plain on/off switch).
  • Reworked the SIGN setting, which was designed to give each module a unique tone coloration, and glitches, generated from its serial number. The generated curves are now more "fuzzy" and "organic"; and the amount of coloration can be adjusted (instead of a plain on/off switch).
  • Exposed many of the parameters of the internal AD envelope. The TDST setting has been replaced by 4 new individual settings to control the amount of FM, TIMBRE, COLOR and VCA modulation from the internal AD envelope – making it dead easy to create drum sounds with Braids. The TENV setting (the preset envelope shapes with silly names) has disappeared and you can now control both the attack and decay of the internal AD.
  • Added many scales (about 50!) to the built-in quantizer, whose code has been rewritten from scratch to allow all kind of stuff not based on the 12TET system. A new setting, ROOT, selects the root note upon which the scale is built.
  • Reversed the polarity of the output signal, so that the sawtooth wave ramps up.


  • Added 808 kick and snare models.
  • Added cymbal noise model.
  • Added triple sine and triple triangle models.


  • Fixed an issue causing rare glitches with the sawtooth and square modes.


  • Added a new TDST option to assign a behavior to the trigger input: phase reset (sync), AD timbre modulation, AD amplitude envelope, and AD timbre/amplitude modulation.
  • Visual feedback added to the META mode.
  • The CV-controlled model change in META mode is now relative to the current model, as selected by the encoder.


  • Fixed a bug that prevented the system settings to be written in memory in some situations.
  • Added a new setting to the RANG menu that locks the output frequency to 440 Hz (useful as a reference for tuning other modules).
  • Altered the phase and amplitude of some waveforms in the WTBL and WMAP modes to make the interpolation/scanning smoother.
  • Added a new wavetable mode, WLIN, which allows the wavetable data to be scanned linearly by the TIMBRE control; while COLOR morphs between various interpolation methods and resolutions (between waveforms or within waveform sample data).
  • Added a new wavetable mode, WTx4, generating four voices of wavetable synthesis. TIMBRE morphs through a representative selection of the waveforms; while COLOR selects various intervals between the four voices (unison with detuning, octaves, chords…)
  • Added two new synthesis modes, with 3 voices of anti-aliased sawtooth or square waves. TIMBRE and COLOR control the interval/detuning of the second and third voices.
  • Added a new synthesis model, DRUM, which is a variant of BELL generating an additive timpani-like sound. TIMBRE controls its decay, while COLOR the tone brightness.
  • Reworked the FOLD mode to provide more destructive wavefolding on its triangle/sine internal inputs. The amount of wavefolding decreases when note frequency is increased to prevent aliasing.
  • Reworked the FBFM mode. The range of the FM index control (TIMBRE) is dynamically adjusted as a function of the COLOR and main frequency settings. This allows a greater amount of destructive FM in the low and mid ranges, while keeping aliasing under control when the frequency reaches the higher octaves.


Initial release.

Miscellaneous builds

v1.9 (unreleased)

After the release of version 1.8, new sub-oscillator models were added, but development halted here without any further changes that would have.

Download (1.9)

Odd FM input range fix

On some modules, the FM input does not reach the full range, causing some models to be unreachable in META mode.

Download (fm_range_fix)