DX7 to Synergy Converter

Motivation

As one of the most popular digital synthesizers of all time, there are thousands of DX7 voices available online. The DX7 and Synergy share some common characteristics and sounds developed for the DX7 can be translated to similar sounds on the Synergy. The converter hopes to make some of those DX7 voices available on the Synergy, or more realistically, make some of those voices an inspiration and starting point for your own unique Synergy voices.

SYX to VCE and CRT

For each SYX file, the converter converts each voice in the sysex and produces a VCE file and then creates one or more CRT files containing all the converted Synergy voices. VCEs, DOCs and CRTs are created in a folder named via the SYX base filename. For example, given a SYX named foo/bar.SYX, the converter would produce:

foo/bar.SYX
foo/bar/bar.CRT
foo/bar/bar-2.CRT
foo/bar/voice1.DOC
foo/bar/voice1.VCE
foo/bar/voice2.DOC
foo/bar/voice2.VCE
...
foo/bar/voice32.DOC
foo/bar/voice32.VCE

Synergy voice names attempt to preserve the original DX name, however this sometimes involves compressing the 10-character DX name to the available 8-characters on the Synergy. The converter attempts to preserve ‘significant’ characters and prefers to eliminate spaces, punctuation and vowels. If more than one voice have the same name after such conversion, the converter appends a numeric suffix to ensure that each name is unique. For example,

DX name Synergy Name
”–woohoo–” ”–woohoo”
“FabVoice12” “FabVoc12”
“Brass␣ ␣ ␣01” “Brass␣01”

Since a DX7 sysex can contain up to 32 voices and a Synergy CRT can support only 24 (and depending on their complexity can sometimes fit less into a CRT image), the converter puts the first set of VCE’s into the bar.CRT and then creates a bar-2.CRT for any voices that could not fit in the first one.

User Interface

Click the Load/Save button and select a Convert DX7 menu item to initiate the conversion. On Mac, there’s a single option since the Mac file selector allows either folder or file selection. On WIndows and Linux there are separate menu items for selecting a folder or selecting an individual SYX file.

When selecting an individual SYX file, only the voices in that file are converted. When selecting a folder, the converter recursively walks that folder’s contents (including sub-folders) and converts every sysex it finds.

Conversion Approach

The DX7 and Synergy synthesis architectures share some characteristics, but are not identical. In some cases there is a one-to-one correspondence of a parameter in a DX7 patch to a corresponding parameter in a Synergy voice. However, there are some DX7 parameters that have no equivalance on the Synergy.

The Synergy carrier and modulating patches are, for the most part, identical to the original DX7 algorithms, subject to the following constraints:

  • Due to the order that oscillators are evaluated by the Synergy osc board, oscillator numbering is reversed in Synergy vs. DX – DX oscillator 1 converts to Synergy oscillator 6, DX 2 converts to Synergy 5, etc.

  • The Synergy has no way to feed back a signal to an oscillator modulation chain. Oscillators with feedback use the Triangle wave as an approximation of of the DX sound. On the DX, the waveform produced by full feedback level (7) produces a Sawtooth wave, so the Triangle wave is closer than the standard Sine wave.

  • DX pitch and amplitude envelopes are directly converted to Synergy envelopes. The Key Acceleration value on the DX is translated to the LOW (Yellow) Envelope, to give the sound change from soft to hard key strike. You may need to scale the Lower or Upper Envelope gain to your playing.

  • DX key level scaling are converted to Synergy filters.

  • DX algorithm 16 and 17 cannot be implemented directly on the Synergy. Instead, we use a special 5-oscillator patch which produces similar sounding voices.

  • Modulation levels for carriers have been scaled to emulate the sounds of the original DX voices. Some algorithm-specific adjustments are made automatically during the conversion. One of these is due to the fact that 3 and 4 Oscillator ‘towers’, as they are decribed in the DX7 world, are not handled the same in the Synergy. The 3rd and 4th upper Oscillators are lowered to about 25% of the original level. Since this isn’t a one size fits all fix, some voices may be helped by raising the gain on these Oscillators.

  • One of the features on the DX7 that is not on the Synergy is being able to set the Harmonic for the Oscillator at 0.5. This is emulated by setting the Synergy Oscillator at 1, and raising the other Oscillators by 1 octave, then setting the Transpose down 1 octave so it matches the DX7 voice.

Recommendations

While some DX voices convert very accurately to Synergy, do not expect a 100% accurate conversion in general. Although both are considered to be FM Synthesizers, there are multiple ‘flavors’ of FM Synthesis. The Carriers sound identical in additive mode, but when modulated, the DX7 and the Synergy do sound different. After creating the Synergy voice you may need to tweak it before it sounds right.

Things to try:

  • Adjust oscillator gain levels to change frequency modulation and oscillator balance
  • Adjust envelope rates. Add extra envelope points.
  • Tweak filters

In some cases, the initial Synergy voice may not sound much like the DX voice at all – but may inspire you to create a completely new sound for the Synergy.