Yamaha FS1R

Yamaha's fine tradition of free product‑support software continues with a Mac editor for their unique FS1R synth.

Looking back at the December '98 SOS review of Yamaha's excellent new FM/formant‑shaping FS1R synth, two things become clear: first, it's not the easiest synth to program, and second, some desirable features are not accessible from the front panel. The solution to both problems is computer‑based editing software.

As with most popular synths, profiles for the FS1R should become available in generic editing packages, but now there's also a dedicated option for FS1R owners which offers distinct advantages: it's been developed by Yamaha themselves, for the Macintosh (System 7.5, 4Mb RAM and 4Mb hard disk space minimum), and it's free!

Dubbed simply FS1R Editor, and currently at version 1.01b (beta), the software provides on‑screen control of all FS1R parameters and a convenient way of organising banks of Performances and their constituent Voices. The software's presentation is functional rather than sophisticated, but its layout is logical, especially given the complexity of the synth. With the FS1R plumbed into my MIDI interface (under OMS; the software is also compatible with Apple's MIDI Manager and PatchBay), I was able to edit sounds on screen and audition them from my master keyboard. In almost all cases, the edits were immediately reflected in the FS1R's edit buffer.

The main Voice edit window: double‑clicking on any of the icons takes you to its edit page.

A Great Performance

Under normal circumstances, when using a synth editor for the first time you'd have to transfer the synth's sounds into your computer. To save users the trouble of doing this (which, by the way, takes a long time), Yamaha have included all the FS1R's factory Performances with the software.

FS1R Editor is a librarian as well as an editor, and makes it very easy to manage Performance and Voice banks. Several Performance or Voice library windows can be open at once, and it's very easy to cut and paste Performances and Voices to compile custom banks. Each Performance bank has an allied Voice bank composed of the Voices that make up those Performances, and when you begin to move Performances around it's nice to see that the relevant Voices are moved with them, automatically.

Editing is similarly straightforward, though as there is a huge number of parameters available in the FS1R, only a few will ever be visible at any one time. Double‑clicking on a Performance opens the Performance Common window, and from here it's easy to point and double‑click through the four Parts (each comprises a single Voice), effects and Formant Sequence (FSeq) assignments for that Performance.

The FS1R has six memories for user FSeqs, but there's no way to actually edit them, or create original ones, using the synth by itself. Enter FS1R Editor, which can do the job, but with limitations.

The same method is used for editing Voices, though the main Voice edit window is arranged rather like a web page set up to work with frames: double‑clicking on text or graphic icons causes theselected parameters to appear in the 'frame'. Icons for the Operator parameters of the current Voice's algorithm are arranged in columns, and can be freely dragged and dropped between Operators — if one Operator's frequency EG, for example, is suitable for any of the other Operators in the Voice being edited, simply drag it to the other Operator's frequency EG icon to instantly copy all the data.

Formant Ferment

The FSeq editor is the only part of the program to use colour, and it really does come in handy to give you an overview of how the various operators are interacting.

One of the most appealing aspects of the FS1R is its Formant Sequences, which make the synth speak or sing. The FS1R has six memories for user FSeqs, but there's no way to actually edit them, or create original ones, using the synth by itself. Enter FS1R Editor, which can do the job, but with limitations.

Just to recap, the FS1R's Formant Synthesis can use eight voiced/unvoiced pairs of Operators to approximate vowels and consonants. An FSeq then sequences these elements to form words or phrases by manipulating the frequency, fundamental pitch and level for each voiced/unvoiced pair over time. FSeq editing has its own window, where curves for pitch, Operator frequency and level are drawn with the mouse. These are colour coded, making it easy to keep track of them during editing. However, even with this assistance, customising FSeqs is a tricky business. There's little help from the insubstantial documentation, and really the prospective FSeq creator is left to figure it out by experimentation. If you know nothing about the spectral patterns of the human voice at the moment, you certainly will if and when you manage to coax the FS1R's user FSeqs to generate more than silence or broad‑band noise! Another inconvenience is that FSeqs are the one instance where edits made in the software aren't immediately reflected in the synth: an FSeq must be transmitted to the FS1R after every edit, if you want to hear the result.

On top of all this, the FSeq‑creation facility that I was really hoping to see isn't available with this software. As mentioned in the FS1R review, Yamaha's in‑house development tools allow WAV or AIFF audio files to be transmuted into FSeqs. The processing involved must be pretty sophisticated, and perhaps it's too much to hope for in a piece of free software.

Gift Horse

Though I found the FSeq editor a bit of a let‑down, intrepid sound designers with time on their hands will no doubt make the most of it. All other aspects of FS1R Editor are admirable and make it a must‑have 'purchase' for Mac‑using FS1R owners: it allowed me to access everything much more easily than from the synth's display, made rearranging Performance banks simplicity itself, and never crashed. FS1R Editor isn't perfect, but it's hard to complain when the price is this attractive!