Premier IT Outsourcing and Support Services within the UK

User Tools

Site Tools


Table of Contents

From: (Clarence Din) Subject: Synthesizer List 6/4/92 Date: 4 Jun 92 16:55:56 GMT

                            SYNTHS 1992
   Compilation of new and used synths, samplers, and rack modules
               Compiled and Edited by Clarence K. Din

If you would like to include a review of your synth or rack module, please send your info to I will try to update this list as frequently as possible, such as once a month or once every two months, but I make no guarantees!

YOUR SUBMISSIONS should be in the following format:

# <name of gear>

<the review>

<your name and e-mail address>

Please DO NOT include entire MIDI specs charts in your reviews unless they are absolutely necessary in proving a point in which Synth A is better than Synth B. Your reviews should be clear and concise. Humor, of course, is allowed, but keep in mind consumer and your fellow synth musician's interests!


# Alesis S4 QuadraSynth Sound Module

The Alesis S4 QuadraSynth is a 64 voice, rackmount MIDI sound module featuring built-in parallel-matrix digital signal processing. The 64 (sic, probably S4) has a large, easy-to-read, back-lit LCD display for "Composite Synthesis," which is an innovative combination of subtractive and additive synthesis. Each program can be made up of as many as four individual "sounds" (dividing polyphony by four in the process???? Eirikur) each with their own set of 3 envelope generators, 3 LPO's (sic), tracking generator, dynamic low pass filter, and very comprehensive modulation matrix. In the "Program Mode," teh S4 has 128 present and 128 user programs in which the user can program up to 7 simultaneous effects which can be independently assigned to any of four effect busses. This allows different programs or composite sounds to be routed to different effects with dynamically controllable amounts. In the "Mix Mode," the S4 has an additional 128 preset and 128 user programmable "Mixes" which can be used for multi-timbral sequencing applications. In this mode, differnt programs may be assigned to each of the 16 MIDI channels simultaneously, and eve (sic) allows you to have separate discrete effects on different sounds with complete mix control. That's why we call it "Mix Mode" rather than a simple multipatch. Programs may be routed to any of the four audio output jacks on the back and the S4 supports optional Sound Rom (sic) cards. Has OPTICAL OUTPUT, (mentioned in the S5 paragraph, summarized below).

Projected Ship Date: Third Quarter 1992 Projected Suggested US Retail Price: $995

# Alesis S5 QuadraSynth Master Keyboard

Same synth engine with a larger display in a 76 key master keyboard with zones, velocity, aftertouch (doesn't say if it's channel or key) and release velocity. Two wheels, pitch and mod, but can be "completely programmable using the modulation matrix." Both S4 and S5 have digital optical outputs and can go directly to ADAT without leaving the digital domain.

Projected Ship Date: Third Quarter 1992 Projected Suggested US Retail Price: $1495

Eirikur Hallgrimsson | This space blankly left intentional. |

# Alesis D4

My local Alesis dealer (Music Depot in Grass Valley, CA.) called me at work today to tell me that he had received his first D4, and ask if I'd like to come by and see/hear it. I wound up buying it! I paid $350. He claimed the list is $399, but I seem to recall seeing $359 somewhere. Oh, well.

My initial impressions are very good, and I'll post a full review in the near future. In the meantime, if you have specific questions, send me mail and I'll see if I can answer them for you.

I was unable to fully incorporate it into my rig tonight, as I am playing at an open mic tommorow and there's no way I can have it in the rack, setup, and adjust my sequences in time, so I'll be using the HR-16 one more night. As I couldn't play with it much, I took the time to enter its MIDI implemen- tation chart for you. Note that pith bend and controller 7 are supported!

From: (Chris Wilson)

Date: 16 Apr 92 01:33:39 GMT

Okay, I've been thinking about getting a D4, so I have tons of literature from Alesis lying around, and have played with one. Too many people have been giving incorrect or partly correct info, so here's the structure of the D4 playback system: (Note: this is done with a flyer from Alesis sitting in front of me.)

There are 500 drum sounds.  You can store up to 21 kits on the D4,

and use program change messages (or the footswitch or front panel controls) to change between them. Each kit consists of up to 61 sounds selected from the 500 available. You can set the panning, volume, and output pair (main or aux) for each sound individually. This 61-note "window" (as Alesis calls it) can be moved across the entire range of MIDI- i.e., you can use MIDI note #s 0-60, 36-96, 67-127, or whatever.

-Chris Wilson

# Casio VZ-1

I bought my VZ-1 last month for 429 US dollars, and I'm smiling ever since. I know that Casio has this bad reputation of making only toy keyboards, but this is really a pretty good intro level synth, definitely worth the money.

It is a 16-voice polyphonic synth with a 5 octave, velocity and aftertouch sensitive keyboard. It is also 8-way multitimbral with static assignment of polyphony on each channel. You can split/layer 4 voices. There are 64 ROM preset patches, and 64 ROM combi areas where performance parameters such as keyboard layers/splits, portamento values etc are stored. Also, you can store 64 patches and 64 combis in RAM. In addition, it comes with a 128 voice/128 combi ROM card. In my opinion, some of the presets are really good, and some just plain useless, with perhaps a half/half split between the two (your mileage may vary :-). Another good thing is that in addition to the pitch bend wheel, there are two user definable modulation wheels. Also, there are inputs for a foot modulator, foot volume pedal and sustain pedal.

MIDI implementation is fairly good, although there are some things missing, like there is no local on/off and no program change mapping. This makes it a little inconvenient to use it as your master keyboard.

As far as its sound architecture goes, there are 8 "modules" per voice. Each module consists of an oscillator with 8 possible waveforms, an 8 stage envelope generator for the amplitude, and an LFO that can modulate the EG. Also there is a pitch EG with its corresponding LFO, that can affect all of the 8 modules. The modules are organized in pairs, where the oscillators in each pair can phase modulate or ring modulate each other, or are simply added. Also, the output of a pair can phase modulate the next pair. So it can be in a sense like an 8-operator FM synth with some limitations (like no feedback), or an additive synth with 8 partials, among other things. I think it is a pretty flexible architecture. Also editing from the front panel is quite easy and intuitive.

The largest complaint I have about it is the lack of public domain support [EdNote: Look in the latest issue of Keyboard. Sound Source Unlimited has a MASTERAM Analog Collection for the VZ series]. There are no PD patches or editors/librarians as far as I know (I was particularly looking for IBM stuff, so I don't know about the ST). Right now I'm working on writing a Glib based ed/lib for it, but given my PhD workload, it will take a few months for me to finish it.

–Ismail Dalgic

# Casio VZ-10M

i also joined the pack that picked up a vz10m from sam ash lately. bought 2 actually, one for a friend who could use the voices. i do like it, so in spite of the comments below, i'd recommend it at the current price.

there are a few things however that i'd like to have known ahead of time. so for the rest of you, here are my beefs (so far: i've only had it a few days now. i'm a terribly critical person when it comes to music gear i'm afraid.)

1. although some of the postings make it seem that the vz10 could be an enhances cz class machine, it is actually a completely different beast. some of the internal mechanisms for generating the sounds may be similar, but any hope of tapping the vast library of pd cz-? patches is in vain. i was most interested in the vz because i've always liked the guitar and flute sounds from the cz series, but no direct port of the particular sounds i like is possible.

2. the vz10m is not nearly as quiet as the cz machines. one of the cz's best features is that it is DEAD quiet. no such luck on the vz. patches that don't have a loud sustaining portion reveal a distinct fuzz or hiss as they decay, until some sort of noise gate or internal oscillator on/off mechanism kicks in. this is not too terribly bad, and in most music this noise would be masked by other instruments, but for sparse arrangements featuring just a vz voice it is noticable.

3. the manual is (IMHO) awful! a perfect example of describing in painful detail how to program a synth, with very little info on whats going on and why you might want to twiddle this or that parameter. especially considering that knowledge of the cz series doesn't help hardly any, this hurts. i have ordered an applications guide to the vz series that i hope will do better, but as it stands i feel completely on my own in figuring out how to achieve any particular sound i might be shooting for. (not that this is necessarally bad mind you. i might come up with a unique technique or two by not having been told how to do it 'right'.) (the application guide is by Steve DeFuria, keyboard mag columnist, writing with a (japanese?) fellow whose name i can't remember. don't have the title at hand either, but it's something like "power play vz1/vz10m"?)

dan mcmullen, everex systems, inc., sebastopol, ca. (707) 823-0733 …!{well!fico2,pacbell!mslbrb}!everexn!dm

… The VZ-10M has EIGHT stages which can be numerically and/or graphically edited. The sustain point can be positioned anywhere in the envelope as well as the end (i.e. one doesn`t have to use all eight parameters and waste time setting undesired points to zero).

Are they "wonderful". Sort of. Actually programming them is a bit difficult – the interface works on levels and rates (which is absolutely absurd IMHO). I would have preffered levels and times. Programming these is like defining a time function by manipulating its derivitive – too indirect for me.

Peaves… the envelopes cannot be looped – they always finish!!!! What WERE the engineers thinking? Idiots… :-)

Paul D. DeFrain Purdue University

# Chroma

Until the Xpander came out, the Chroma had the most powerful voice

architecture of any programmable synth. It was originally designed by ARP, a company who had 40% of the synth market before they unfortunately went out of business. The Chroma had 16 VCO's (usable either as 16 individual, or 8 pairs), 2 filters (each capable of highpass/lowpass) that could be put in series or parallel, and could do FM and (I think) ring modulation. Back in those days, if you had a Chroma and a DX-7, you had one hell of a keyboard rig!

The user interface is unusual and a little crude.  It has 50 membrane

pushbuttons and a data slider (similar to DX-7 interface), but the only readout is an LED with maybe 16 characters. When you pressed a membrane switch, the instrument made a satisfying "THUNK!" so you knew you had pushed something. (Tactile feedback, anyone?) For modulation gear, it has 2 skinny levers that you push forward and backwards.

The Chroma has a bunch of LFO's and envelopes, lots of goodies to

modulate, and some sophisticated pedal algorithms. For example, you could load a chord into the Chroma and then play it later by pressing a pedal. It has an arpeggiator, and a minimal monophonic sequencer which is primarily an arpeggiator that plays back the notes in the order that you played them.

The Chroma has a velocity-sensitive 73-key weighted keyboard.  The

company (Fender) claimed that pressure-sensitivity would be added, but (I think) this never happened, or the company decided against it because it adversely affected the feel of the keyboard.

Chroma was pre-MIDI, but it could be interfaced with an Apple II to

provide 8-voice multitimbral operation and (I think) sequencing.

Here are a few of the interesting things a Chroma could do:

o It had pulse width modulation, but it could be applied also to the

sawtooth wave of the oscillator, not just the square wave.

o Its internal voice architecture (VCO/VCF/VCA) would be re-ordered

by pressing buttons on the front panel.  Effectively, this was like
having 10 different kinds of synth!

o The Chroma Polaris (a later instrument) pioneered an interesting

form of pitch bending.  Hold down some keys, press a pedal, let go
of some of the keys, and push the pitch wheel.  Only the keys
that are still depressed will have their pitch bent.  Great for
steel-guitar licks.

o The "chord loading" feature I described above.

Then, the Xpander came out.  It had more (and more powerful)

modulatiors, strong MIDI implementation, multi-mode filters, lag processor, tracking generators, and a host of other things. I was all set to buy a Chroma when the Xpander suddenly came out and dashed my plans. :-)

I've been able to duplicate 99% of the Chroma's functionality on my

Matrix-12, including the weird features like chord loading and steel-guitar pitch bending. The only exceptions are the re-routable voice architecture and the weighted keyboard, which are obviously hardware differences that cannot be accomplished with an M12! (Though the rerouting can be mimicked quite well with the M12's multi-mode filter, FM, and oscillator sync.)

As for sound... the 2 instruments are both analog, but the sound is

different. I feel that the Chroma's sound was less "fat" and more "buzzy" than the M12's, but these are not negative qualities – just differences.

Dan Barrett – Grad student, Department of Computer & Information Science
University of Massachusetts, Amherst, MA 01003 –

An outside software firm also created a program that would provide MIDI interface thru the Apple ][+ using the Chroma interface and a standard MIDI interface. You could run the MIDI converter or the sequencer, but not both at once. The interface also provided patch library functions.

Here are a few of the interesting things a Chroma could do:

One of the keyboard algorithms was "pitch ordered" assignment. The highest note got voice 1, next highest got voice 2, etc. This allowed the use of polyphonic portamento! For example: Play a G chord in the octave below Middle C. Now play an A minor the octave above Middle C. The notes will each slide to the appropriate relative note in the next chord. Cool and a half! Can the M-12 do this?

I feel that the Chroma's sound was less "fat" and more "buzzy"
than the M12's, but these are not negative qualities – just differences.

Given that "fat" and "buzzy" are real relative here, we won't be able to really talk about differences in sound, but I now have an EX8000 and once had a DSS-1 (the Korg sampler/synth). Each comes close to the Chroma in terms of "fat" but (hangs head in shame) I've never had the chance to play with a Matrix 12 (although I saw a guy in a band playing an Xpander once).

The real drawback to the Chroma, and to a lesser extent of the M-12, is the weight! If you want to play one live, you have to carry it! If I remember correctly, the weight of the Chroma is/was 55 lb. With an Anvil case it weighs over 70 pounds!!!

Dennis Pelton!ncsc8!dgp

One of the keyboard algorithms was "pitch ordered" assignment. The
highest note got voice 1, next highest got voice 2, etc. […]
Cool and a half! Can the M-12 do this?

Not directly, but you can fake it in multi mode, I'll bet. The M-12 and Xpander have six different modes for assignment. – | Reap all the Wages of Sin! (d.dax)

# Emu-Procussion

The Emu Procussion is a single-space rackmount device along the same lines as the Proteus, but with percussion samples with an intent to be controlled by a MIDI drum kit or keyboard. I've had one on loan for a few days….

Are all Emu boxes built like this, with a plastic enclosure that's one rack space plus epsilon wide? Phooey! Front panel is minimal, with buttons for MASTER, EDIT, ENTER, and CURSOR, a DATA wheel, and a volume knob. It has six outs, one stereo pair and two Sub pairs that can double as effects send/return pairs. The menuing system is wretched, but usable– a computer editor would have been nice.

OK, so how's it work? The hierarchy is as follows. At the bottom you have the Instrument. An Instrument has a particular Instrument Number to select a sample from the Pro's memory (there are 220 of them, various drums and ethnic instruments as well as synth sounds and sine harmonics of various sorts), tunable up to plus or minus an octave, panned, with a settable Delay amount and Forward or Reverse playback. It has a volume and accent (extra volume) setting, for two levels of dynamics, and a 3-stage ASR envelope that can work in Gate or Trigger mode. Various parts of this Instrument can be modulated by velocity, key number, trigger rate (!), mono pressure, pitch wheel, or a random-value generator, as well as up to four assignable MIDI controllers. OK so far?

Now we have a sample whose playback can be delayed, enveloped, and panned. Four of those together form the Layers of a Stack, the basic building block of the Procussion at the user's level. The Stack plays back the four Layers at once or in order, depending on how it's modulated: Layers can be switched by velicities, controllers, etc. There are velocity response curves, pan controls, etc., available, but these are often overridden by global values (more on those in a moment). There is also something called "Procussion Synthesis" which is nothing more than envelope-driven crossfading of samples, and a pseudo-reverb called Spatial Convolution (!) which actually is stored as a sample and used as a Layer of a Stack, but is actually a sort of transform multiplication taken from the Emax II. It's weird, and it sounds like gobbledygook, but it works! Basically, you layer a snare with a "snare space" and the layered sample sounds like the snare has reverb on it!

There are over 550 permanent Stacks in memory, ranging from useful to weird, and you can create 512 more of your own. But here's the major glitch in the implementation of the Procussion: you can't choose to use ONLY your own Stacks, without crippling the machine! Here's the next level up: the Zone and the Kit. Each Stack is assigned to a Zone, which has a MIDI note range, switchable for following pitch or not if there's more than one note in the range, Coarse and fine tuning, volume, pan and output assignments, a choice of whether polyphony is unlimited, limited, or has special features like chokes for cymbals or hi-hat closure, and a couple of global modulation sources. You put together up to 24 Zones to make a Kit, which covers the keyboard range. But the problem is, only 8 of the 24 can have User Stacks in them, the rest must be Factory Stacks! I consider this a bit annoying, and it will probably get owrse as one gets more and more used to tweaking Stacks to one's needs.

There are 128 Kits, 64 preset and 64 rewritable, and they have global MIDI interpretation commands, including footswitch and footpedal control options for kick drums, hihats, etc. The DrumKat and Pocket Pedal are suggested as suitable controllers, as are the Octapads, Impulse, and so on.

The unit also has a gee-whiz Demo Sequence built in. Interesting, once.

The manual is clear, well written and illustrated profusely with a decent index. Extra points on this one.

And the sound? Well, I'm not the best judge, being a guy who's had nothing in his studio but a TR-707 for drum noises for a LONG time, but I would easily recommend this unit to anyone who wanted a drop-in sound device with a fair amount of flexibility, for whom the Alesis D4 wasn't enough and a sampler was too much. The Stacks are interesting, dynamically expressive, and a lot of fun, with SFX noises and a number of good tuned sounds like marimbas and basses. But the unit's a pain to program, and getting weird results takes some doing– as Nick would say, there's not a lot of distance between the samples and your sounds. As a drum synthesizer, it falls a bit short of, say, the XD-5. But it sounds a LOT better in terms of sample quality, in my opinion.

I don't know if I'd spend the 700-plus bucks on one; I'm not that enamored of 1990s drum noises YET. But I would easily recommend it to anyone who wanted a compromise between presets and total controllability like a sampler's at a reasonable price. Let your ears be the judge, and work with it IN THE CONTEXT OF YOUR RIG for a while before deciding.

– | but a dragon can only be painted….

# Ensoniq ESQ-1, ESQ-M, SQ-80

These synthesizers form a "family". The ESQ-1 was the original keyboard version, the ESQ-M is the rack version and the SQ-80 is essentially an enhanced ESQ-1.

Sound generation:

The ESQ-1 family is a hybrid of digital and analog. The "oscillators" are digital, as is much of the rest of the synth. The filters and final DCA stage are analog, courtesy of the Curtis chips common in many synths of this vintage (1986).

The unit has eight "voices". Each voice consists of three oscillators (each with its own amplifier), a single filter, a common amplifier, and a modulator pool of three LFOs and four envelope generators. With the exception of the envelope generator hardwired to the final amplifier stage, the modulator pool can be routed to any destination, the source of most of the unit's versatility.

The oscillators use single cycle (with some exceptions in the SQ-80, stay tuned) digitized waveforms. Those that are obtained from samples are multisampled across the keyboard. The ESQ-1 and ESQ-M have a pallette of 32 waveforms. The SQ-80 has these same 32, plus an additional 43. A few of these 43 are multi-cycle samples. Waveforms include the bread and butter squares, saws, pulse, etc. from analog subtractive synthesis, plus sampled sounds including pianos, basses, woodwinds, as well as waveforms derived from complex synthesis techniques like the Karplus-Strong plucked string algorithm, etc.

Patches may be used polyphonically or monophonically; portamento is supported (called "glide"). Any patch may be split and/or layered with anmy other patch. The ESQ-1 architecture does not distinguish between a "normal" mode and a "split" mode like many synths of its vintage. It merely maintains pointers to other patches in memory. 40 patches are stored in local memory, a cartridge port allows two banks of 40 each to be accessed. Some third party carts support multiple banks, the max I have seen is 320 in one cartridge, but only 80 at a time can be used. Cartridges use EEPROM chips not RAM, so no battery is needed to hold the data.

Each patch is actually mono, but can be panned between the two outputs. Panning is at the *patch* level, so that patches can be made to move across the stereo soundstage under control of an LFO, enevelope, velocity, mod wheel, what have you. The unit is multitimbral, using dynamic voice allocation to assign the eight voices as needed. The sequencer track pages are used to control the multitimbral part assignments.


The ESQ-1 and SQ-80 both have a 61 key keyboard. Both are velocity sensitive. The SQ-80 keyboard also has aftertouch and can generate either polyphonic/key pressure or mono/channel pressure. All three units respond to both types of aftertouch via MIDI. Bend and mod wheel plus control voltage (expression) and sustain pedals are supported. One external MIDI controller can be used as a modulator as well.


The ESQ-1 and SQ-80 have an onboard eight track sequencer that is very easy to use and was at its time state of the art for an onboard unit. The original ESQ-1 stored 2400 events, later expansions were offered to allow 10,000 or 20,000 event storage. The SQ-80 stores 20,000 events and no expansion is possible. Sequences are stored in two levels, "sequences" which chain into "songs" (similar to how most drum machines chain patterns into songs). The ESQ-1 has 10 songs/30 sequences maximum, the SQ-80 allows twice as many (20/60). Sequence data is interchangeable at the individual sequence level and bulk dumps from ESQ-1 to SQ-80. Dumps from the SQ-80 to the ESQ-1 cannot be done. All units have a cassette tape interface for data storage, which doubles as an FSK synch connection for the sequencer. The SQ-80 also has a 3.5" DSDD floppy drive for faster data access. The drive can be used to store system exclusive data up to 64K bytes in length. In addition, there are hardware and software hooks to be able to save data off to the disk drive of an Ensoniq Mirage.

Sequences may be edited via overdub or by step editing. Controllers may be stripped but there is no single event editing of anything but note data. Program changes, tempo changes, time signature changes and volume changes may be programmed into the sequence but can only occur on sequence boundaries (essentially on any bar line). A very useful feature is that all sequence changes may be auditioned before any existing data actually is modified.


The sound of the ESQs is halfway between an analog synth and early digital machines (like the DX7). Although some sampled waveforms are included, the sounds created with them are not overly realistic compared to more modern sample-based machines. Many people describe its sound as "grainy", although it seems to be free of the high frequency aliasing artifcats of the early FM and L/A machines. The modulation pool and resonant filters make this unit well adapted to classic synthesizer textures. The SQ-80's additional waveforms include attack samples and acyclic loops which allow it to produce some sounds similar to the Roland D-50. It also includes some drum samples, but the "kit" is limited to kick, snare, closed hi-hat and toms. These samples are, however, quite useable and can be processed through the amplifers and filter to create numerous "synth drum" timbres. The SQ-80 can reproduce any ESQ-1 or ESQ-M patch. SQ-80 patches using the additional waveforms not found on the ESQ-1 will not sound correct on the ESQ-1 or ESQ-M.


The MIDI implementation is quite good. All front panel functions can be controlled by system exclusive. The units send and recieve on any MIDI channel, in fact these channels need not be the same. A special mode for use with guitar controllers is included. The ESQ-1 has MIDI IN and OUT only, the ESQ-M and SQ-80 also have THRU.


Three years before Korg proclaimed the M1 a "workstation", the ESQ-1 was one. It is still a good choice for someone looking for a self-contained synth/sequencer package. The SQ-80 is worth the extra $$ for the disk drive, aftertouch keyboard and extra waveforms if these are of interest. The ESQ-M module is not as common (it sold poorly when in production) since many users were not as enamored with the ESQ-1 sounds as with the built-in sequencer. The sequencer itself still holds up well against some other on-board units in competing machines (I'm thinking specifically of the Korg M1 and Roland D20 here). The sound is somewhat generic, in that it doesn't have an obvious "signature" the way the DX7 and D50 did. However, it has an architecture well suited to being able to create a wider range of sounds than other synths of its vintage. In that sense, it's of more interest to folks willing to learn to program it as opposed to just calling up some hot presets. There is still a reasonable amount of third party support for the ESQ-1, you can still buy patch sets, get memory cartridges, etc. although that support is bound to dwindle in the coming years. There is a good amount of public domain support for the machine, including plenty of patches and patch librarian software for a number of computers.

Brian Rost

# Korg 01/W

Re: Is the 01/W a sample-playback or a synthesiser?

I finally got my hands on one, so therefore can argue more intelligently than before. Unfortunately, I stuffed up my own experiment to see how much "distancing" this machine is capable of.

(1) Going through the combinations, I was very impressed by the sound of this synth, probably the best I've heard yet. Korg gets an A+ for market strategy (I must look into buying shares). This is especially when you consider that this is a tweaked M1 with probably little development cost (unless they busted a gut getting the extra polyphony).

(2) After familiarising myself with the factory programmes, I took the piano 8' patch, put the filter wide open, removed as many effects/envelopes/ etc as I could find, and went through the P.C.M. (Note: This is where I stuffed the experiment). In the first 255 PCM waves I went through, I found very little that could explain the excellent sounds. There were a few (e.g. "Thing" (several) and "Crickets" (several) where some of the good bits of the sounds I had been hearing were basically just playing back PCM.) Unfortunately, it wasn't till I had left the store that I remembered that there are a lot more waves on the machine (another two banks?), and the key parts of other patches may have been in there. (I will have to check this out). The PCM was useful-sounding, but hardly imaginative material (Disclaimer: I am not claiming that the "sounds", i.e. patches/combinations are unimaginative). All the usual PCM was in there, and of course I didn't look at all the PCM, and it sounded (this entire sub-review is highly IMHO) like a well-thought-out basis for synthesis, rather than a collection of gosh-wow sounds to be played straight back in the store to impress people.

(3) Using mainly the "Sine" and "Saw" waves I went through the different waveshaping tables. I wasn't able to actually figure out what this is meant to do, as far as I could see it added some grunge or distortion to the sound, or, let it through basically unchanged. My IMHO conclusion from this is that it doesn't add much to the synthesis process, certainly less than putting real resonance on the filters would.

(4) The filters did sound nice (IMHO), but people may have noticed my bias for a certain sound-shaping tool that's missing here.

Anybody reading this far? At least you've got to give me one point for actually going and listening to the machine.


# Korg 01/W

The waveshaping can add a little more realism to some of the acoustical sounds. It O.K., its just nothing to write home about. It might stand out in a solo with a quite accompaniment. You can also exaggerate the waveshaping effect to create some weird sounds from some not so weird samples.

True, you don't have the wave molding capability of the Wavestation, but with a little bit of creative effort you can get away from some of the boring playback sounds. Also, if you have a computer you can have it dynamically change some of the 01/W patch parameters as you play (simulating the dynamic sound color of the Wavestation). But, no - its not a Wavestation - and was never intended to be one. One the other hand, if you are in to composing, scoring for acoustical instruments, want a super grand piano sound, etc., the 01/Wfd is great. It has a 50,000 note sequencer, a disk drive to save sequences, very clean acoustical sounds, excellent drum sounds, and a great effects section (effects very similar to the Wavestation). If you don't have a computer/song-sequencer then the 01/W 16 track sequencer may be an important feature for you. If you really want to compose at the wave (sound) level, and are not so much into song writing (or already have a good sequencer), then you'll be a lot happier with a Wavestation.

One other point. The 01/W, unlike the T series EX models doesn't have sample RAM - you either run with the samples installed at the factory or buy additional samples on PCM cards. I've heard, however, that Frontal Lobe makes a device that pretends to be a PCM card but has its own sample loading hardware. I know the Frontal Lobe works with the M and T series, and would guess that it is compatible with the 01, as well. In fact, I've even wondered if the sample RAM wasn't left off of the 01/W, because this 3rd party solution is now available?? That helps Korg to hold down the price of the 01/W, since many folks don't care about creating/adding their own samples.

# Kurzweil 1000PX

The 1000PX is the rack mount version of the K1000. The 1000Px was replaced by the 1000PX Plus about 1.5 years ago. Basically the 1000PX Plus is identical to the 1000PX except that it has the PXA soundblock installed and some better sounding programs. The two machines are essentially identical if the PXA block is added to the 1000PX. I believe that the PX+ is the currently available rack mount. The K1000 is not available anymore, I don't think. It was replaced some time ago with the K1200, which is esentially a K1000 with an 88 key, somewhat more pianolike keyboard and, I think, the PX+ voice architecture. At sometime there was a device called the K1000 SE, I'm not sure how that differed from a K1000. Anyway, each of the K/PX family can be upgraded to essentially identical devices (using PXA and PXB soundblocks [called KXA and KXB for the K series]). I can highly recommend Kurzweil modules to anybody who is looking for some sampled sounds, in a very elegant package to boot. Also, contrary to what some people believe, the Kurzweil 1000/1200 series devices are capable of some very serious synthesis. Now, there are no resonant filters for sweeping, however there are some number of hundred parameters of each voice which are controlable or mappable or both. There is a very flexible control routing system and some sophisticated envlos and amplifr tricks. Actually, you can probably do anything with a Kurz you can do with an M1, or maybe even a D50 (since the almost nonexistant D50 filter can be easily simulated with the Kurz). So, not only do you get some incredible samples (the piano is still to be beat, in my opinion), you actually get a very flexible synthesizer (sans filters, but many people who read this stuff obviously don't care and wouldn't know what to do with one, anyway) as well. Nick, nonwithstanding, the Kurz 1000 series are synths.


# Kurzweil K2000

[EdNote: The following is a reprinted advertisement from Kurzweil]

The K2000 is a high quality portable synthesizer designed for the professional musician in performance or in the studio. The first incarnation of an all new technology that Kurzweil engineers have been developing for the last several years, it features the highest quality ROM samples ever created by Kurzweil as well as per voice Digital Signal Processing far more powerful than that of any other digital synthesizer on the market today. In addition, more samples may be loaded into RAM, and an optional board will allow you to sample your own sounds. The K2000 combines powerful MIDI controller features, stereo effects processing, multiple polyphonic outputs, expandable hardware design, and a friendly user interface to make it the most powerful, versatile, and expandable instrument ever created by Kurzweil.

The Sonic Possibilities are VAST ————————— ==== "Variable Architecture Synthesis Technology" allows you to take any multisample, waveform, or noise and process that sound using a variety of known synthesis techniques, as well as many new ones yet to be explored. Using any of these DSP functions is like using a unique synthesizer. In the K2000, each note can be processed using its own individual set of these synthesizer functions. It's like having any and all known synthesizers under one control panel. What's more, these different synthesis techniques can be used *simultaneously*. With the K2000, you can start with any sound you want and transform it into something completely new. The possibilities are *truly* VAST.

As Your Needs Grow, So Can the K2000

The K2000 was designed to be extremely flexible and open ended. The ROM is stuffed with 8 megabytes of top quality multi samples, but you may add RAM for loading more samples, using standard SIMMs, up to a total of 64 megabytes of RAM! The on board high density disk drive and SCSI connector make it easy to load and save sounds, programs, setups, and sequences. And Kurzweil and several other companies will be offering extensive libraries of samples. In addition, an optional circuit board allows you to do your own sampling and will accept both analog and digital information.

The Power is Beneath your Fingertips

At Kurzweil, we have always believed that the true power of a keyboard lies in its ability to modify its sound according to the performer's whim. With extensive realtime control of various parameters, from the front panel or through MIDI, both dramatic and subtle performance nuances are possible. Performance Setups allow the keyboard to be used as a powerful controller, transmitting on up to three MIDI channels. And as with all of our instruments, the K2000 is multitimbral on all 16 MIDI channels simultaneously.

Many extra features

How about portamento on any sample or waveform? Up to six octaves of pitchbend range? Sample and Hold? Or the ability to load and playback a MIDI type 0 standard sequence file? These are just some of the many features that you may take advantage of.

Break the sound barrier with the new K2000.

[*page two*]


* 31 VAST algorithms, each with up to 4 configurable digital signal processors per voice. DSP functions include: Filtering (lowpass, highpass, band pass, notch, allpass, parametric EQ, shelving EQ, 2 pole, 4 pole), Resonance, Continuous Panning, Amplitude Modulation, Crossfade, Distortion, Digital Wrap, Waveshaping, Pulse Width Modulation, High Frequency Enhancement, Low Frequency Oscillation, Hard Sync Oscillation, and Mixing Oscillation.

* 8 Megabytes of ROM samples and waveforms, organized into 168 keymaps. Includes: Grand Piano, Electric Piano, Ensemble and Solo Strings, Voices, Acoustic and Electric Guitar, Electric Bass, Flute, Tenor Sax, Trumpet, Trombone, Mallets, Extensive Drums and Percussion, Exotic Percussion Loops, Attack Transients, Percussive Effects, and Waveforms.

* 32 bit internal precision into 18 bit DACs, 20 khz audio bandwidth. 16 bit linear sample format. 1 cent tuning resolution.

* Stereo multi effects processor with up to 4 simultaneous effects, including reverb, delay, chorus, flange, multitap, graphic EQ, Rotary, and more.

* Ergonomic, friendly user interface includes: 240x64 pixel backlit graphic display, mode LEDs, soft buttons, alpha wheel, alphanumeric keypad, and programmable Jump/Mark editing shortcuts.

* 24 voice polyphony, with up to 4 oscillators per voice, for incredibly fat timbres without layering.

* 16 channel multitimbral.

* 6 polyphonic outputs configured as two stereo pairs and stereo mix outs. Stereo insert plugs may be used to route dry signals to and from external effects devices and back through the mix outs.

* 4 slots for expandable sample RAM up to 64 Mbytes, using standard SIMMs.

* 3.5" floppy disk drive, 1.4 Mbyte, MS/DOS compatible.

* SCSI port.

* Optional digital audio circuit board allows for stereo or mono sampling directly from DAT, CD, or analog audio input. Includes Optical, AES/EBU, and stereo 1/4" connectors. Sampling rates are 32k and 48k for analog input, and 32k, 44.1k, and 48k for digital input.

* 200 preset programs in ROM.

* 128 kbyte battery backed RAM for user programs, setups, maps, songs, and other data - enough room for hundreds of user programs. MIDI Type 0 standard sequence files may be loaded into memory and played back.

* Portamento and Mono Mode on any sample or waveform.

* Up to Six Octaves of Pitch Bend range.

* 61 key synth action with velocity and aftertouch.

* Programs may contain up to 3 layers/splits. Drum programs may contain up to 32 layers/splits and each drum may have its own custom DSP treatment.

* Internal modulation sources per layer include three 8 segment envelopes with realtime rate control and looping, two attack/sustain/release generators, two LFOs, two velocity triggers, and more.

* Functions for processing control source inputs include mixers, negators, invertors, sample and hold, quantize, lag, ramp, shape, and more.

* Performance Setups allow the keyboard to be split into three different zones and transmit on three separate MIDI channels. Performance controls include two switch pedal inputs, one continuous pedal input, two wheels, and one slider - all programmable.

	   [--------end of quoted section--------]

Well, that's it. One question I have is this: if there's 8Mb of ROM, and only 128k of RAM in standard config, and if Kurz and other companies are going to release new voices, then where do they go? They didn't mention card slots. I can only assume that you HAVE to get more RAM. Not that that's all that expensive. Should be ~$144 for 4Mb. I hope it'll take 2Mb SIMMs.

Also, considering all their hoopla about how they want the performer to be able to control the nuance during performance, doesn't one slider seem a bit skimpy? I suppose you can always get one of those MIDI slider boxes…

The Mono Mode sounds promising for wind controller applications.

Well, I hope this is interesting/helpful to some of you. I'll be on the phone with Kurzweil Monday to see if I can find anybody who will give me a few more specific answers.

Seeya, –Mike

I want to clear up a few questions about the K2000 from a previous posting.

(1) Can you do LA synthesis on it, e.g. layer a large number of sounds to
create a patch (e.g. four or more). The specs I saw in Keyboard magazine
said up to four oscillators per voice, but I assume only one filter.

Well, only Roland synthesizers can do LA synthesis [tm]. As far as four oscillators per voice, marketing hype aside, it is possible. As on the 1000 series, for each program you have several layers. (3 on the k2000, 4 on the 1000 series). One voice is taken up to play a layer. Briefly, The layer starts out with a sample [acoustic, waveshape, partial, attack, etc.] that is processed through an algorithm. There are about 30 or so algorithms, and each one defines a "flow" of signal processing. There are five components in the chain. Pitch and Amplitude control are fixed. In between are three selectable DSP segments that can be various filters, wave shapers, oscillators (sine, saw,and square), among others. Some of these DSP functions use up more than one segment and do more complex processing. So, if you devote these segments to oscillators then you can effectively have up to a four- oscillator voice.

(2) Can it resample a wave passed through the effects?

No. However, a staggering variety of effects can be created with the DSP processing.

(3) Can you have effects and filters at once, or would you have to use an
outboard effect to do this?

In addition to the DSP effects, there are built-in multi-effects which can be used simultaneously with the programs being played.

(4) Does it have extensive sample editing functions and/or automatic looping
of samples etc.
That software will be part of the sampling option. I doubt that you'll be
able to edit ROM based samples although who knows.

Correct. Comprehensive sample editing features will be included with the forthcoming sampling option. You will probably be able to edit ROM samples. The basic machine will accept samples through MIDI sample dump.

(6) Does the keyboard have aftertouch?
Mono pressure if my memory serves me correctly. It does support poly pressure
via MIDI.

This is correct.

(7) Did you get any impression of programming capabilities, I think you said
that you could get any sound you wanted but what about the programming effort/
results ratio?
Although its prommanging capabilities were limitless (their motto is that
it takes 300 years to explore all of it options), I think with reasonable
practice (considerably less compared to FM) one could acheive some pretty
impressive sounds. On the other hand, I must admit I was pretty overwhelmed
at first. This is a problem inherent in any new device, IMHO.

The programming effort/results ratio has been calculated to be 1.34997. The 300 years was calculated by hiring temorary personnel to sit at the beta-test units and scroll through every parameter to play every sound. This was done while a stopwatch was running, and the "lap time" was multiplied by the number of people, to get 300 years as a final value. In fact, this is the major reason it has been a long wait for the K2000 to be shipping.

(tongue out of cheek now)

Seriously, the user-interface is very intuitive and very easy to navigate. It is very satisfying to program your own sounds on the K2000.

If you are familiar with the 1000 series programming interface, rest assured that the K2000 is heaven in comparison. Remember though, I have never been in heaven, athough playing a Matrix-12 comes close. :-)

It is hard for me not to be biased, of course, and even harder for me to convince you net people that I am not biased, so you all will just have to see for yourself since this is an extremely subjective area. I have a k1000,

# Kurzweil K2000

I, too, was somewhat disappointed after listening to the ROM sounds of the K2000–largely because some of the killer samples that were in the K1200 were not present (vibes and acoustic bass for example). There is a "vibes" patch in the K2000 but it is really a synthesized waveform, not a sample.

In talking with a technical engr. at Young-Chang, he explained that these samples would be included in the next ROM set this summer. Too bad, it's over $700 to add another ROM set (expansion board, $260; 8MB ROM, $500).

HOWEVER, (and this is why I would buy the 2000) supposedly, you will be able to purchase subsets of ROM samples on diskette for $20/10diskettes. If you can truly mix and match any diskettes to get the samples you want, that would be ideal. Or, hopefully, they will offer the entire new ROM set in a diskette package. To add 8MB of RAM to the K2000 should only cost about $300. at today's SIMM prices.

		eugene beer

# Mellotron

The Mellotron was perhaps the first polyphonic "synthesizer" to be made available to the masses of musical performers who did not intend to haul around a real piano or orchestra. The unit is rather large – smaller than a string section though.

It works like this. For every key on the console there is, essentially, a tape player. When a key is pressed a mechanism is engaged which pulls a pre-recorded tape across the head. When the key is released a spring "instantly" yanks the tape back – hopefully before the key is pressed again. Walla – a sample player. But that`s not all…

These tape loops came in racks that could be interchanged so one minute your mellotron would be a string ensamble and the next – an organ or piano. I don`t know how log it actually took to swap a rack but I know allignment was sometimes a problem. I doubt anyone tried this live.

The mellotron, as mechanical and absurd as it was, was the only choice many musicians had if they wanted more than two finger chords on a synthesizer – until, of course, the ARP String Ensamble hit the market – so long mellotron – should have invested some time in research and development.

I write in past tense because the mellotron is no longer a serious instrument for professional use is and more a piece of synth history. I am sure there are plenty of people who have mellotrons in excellent working condition – hopefully someone who does is with us and will post something.

the mellotron, a noisy, but strangely curious, machine…


Re: The Mellotron is a piece of history.

I think Crowded House's producer (Mitchell Froom) uses one. Neil Finn claimed that few or none of the samples on Woodface were from digital samplers. He also claimed that MF used a machine that used spinning analogue optical disks (spinning disks with the waveform etched in somehow) to imitate musical instruments. (Somebody please write in with the proper name for this instrument!).

People who are interested in this sort of thing should read the series "IT came from the music industry" that ran in _keyboard_ some time ago (last year I think).



  To be more 'intuitive', the Mellotron was a ANALOG SAMPLE PLAYER.

ANALOG: Instead ROM and RAM to store the sound, Mellotron used tapes. SAMPLE: Any sound could be recorded in the tapes, but in 'real time' not in

      little divisions called samples. I think this isn't the correct word.

PLAYER: Each key in the keyboard had a tape associated. When you press the key,

      the tape runs, like a 'PLAY' button. A mechanism make a 'FAST-REW',
      when you loose the keys. Cause this, it not recommended for fast solos.
  The Mellotron was sucessful because it was polyphonic, in a age of monopho

nic instruments. But, you was limited to one sound, or tape set, in a time in a live show. No 'Now-I-swap-to-sound-number-x' stuff. Swap sounds = swap tapes, that implies in mechanical troubles, precision, alignments, etc, things that you don't make live.

  There is a man that created a museum-like to Mellotron tapes/parts.

I don't remember the name now. Too many in the net can talk about better than I.


By the way, as has been pointed out in previous Mellotron threads, it would be difficult to do a completely realistic digital sample-playback version of the machine. Each key was a separate, 8-second sample. There were something like 37 keys, so, at, say, a 32kHz sampling rate, that would require something approaching 10 megabytes of sample memory per sound. That's also assuming an 8-bit sample width. And there were three sounds per tape rack, so you'd need 30 megs of sample memory. Hey, this would be a wonderful application for data compression, eh? Low-quality tape machine imitation… And then there's the issue of touch, where, with a Mellotron, if you press the keys lightly, you get a slight reduction in treble and volume. And, the tape rewind time. And, I suppose Ensoniq should manufacture it, in order to reproduce one of the Mellotron's most famous features, namely unreliability…

  1. Jim Smith

Yupatupata da yupadupa chickida, Icktang icktang,

          Ickitack tangdow, Rickitickatar ticka chingtar da.

# Minimoog

The Minimoog was one of the first commercially available synthesizers.

The Mini has a three and a half octave keyboard. It plays one note at a time. If more than one note is played, the lowest note sounds ("Lownote priority"). To the left of the keyboard is the "left hand controller" section. It consist of a pitch bending wheel, a modulation wheel, and two switches, labeled "decay" and "glide". The PB wheel has a center detent, and normally remains centered. The Mod wheel normally remains at 0. Neither wheel is spring loaded.

It is significant to note that the PB wheel directly controls all of the oscillators. There is no center "dead band" as there is on most modern synths. (The "dead band" would be an area of travel at the center where the wheel would have no effect.) This makes it possible to do accurate pitch bending that sounds good. It is also possible to use the PB wheel to introduce a slight phasing or a vibrato, which is not possible on wheels with a "dead band".

The electronics consists of three voltage controlled oscillators (VCO) and a white/pink/red noise source, which feed into four inputs of a five channel mixer. The fifth mixer input comes from an external input on the top of the case. The output of the mixer feeds into a 24db/octave voltage controlled variable resonance lowpass filter (VCF). The output of the VCF feeds into a voltage controlled amplifier (VCA).

There are two identical envelope generators in the Mini. One controls the filter cutoff frequency, and the other controls the loudness of the VCA. Each envelope is an ADSR with three knobs labeled Attack Time, Decay Time, and Sustain Level. The release time is always equal to the decay time. Both release times can also be set to zero by turning off the "decay" switch on the left-hand controller section of the keyboard.

The ADSRs are only triggered when a note is played on the keyboard while no other notes are down. If a note is down, hitting another note may change the pitch (only if the second note is lower than the first), but will not retrigger the ADSRs.

The ADSRs are of the non-return-to-zero variety. That is, each attack starts from the current voltage of the EG, not necessarily from 0. Combined with the non-retriggering nature of the keyboard, this makes phrasing possible. To do phrasing, the player must be careful to lift the last key of a phrase fully before playing the next note, and to play the notes within the phrase legato. For instance, setting a slow attack and decay time on the VCF EG will make the filter tend to open up more and more as successive notes of a phrase are played. This a pleasing effect which "rounds out" the phrase.

The mod wheel controls the amount of modulation which is sent to both the VCOs and the VCAs simultaneously. A knob labeled "modulation mix" controls whether the modulation source is to be the output of VCO 3, the noise generator, or some mixture of the two. Two switches, one at the left of the oscillator section, and one at the left of the filter section, can entirely disconnect the output of the mod wheel from that section.

Each of the three VCOs is a one volt per octave analog VCO. Each VCO has an octave switch labeled "low 32 16 8 4 2" which can be used to change the octave of an oscillator in performance without retuning. Oscillators 2 and 3 also have a tuning knob, which can be used to tune (or detune) the oscillator relative to VCO1 over the range of about +/- a 6th.

Each VCO has 6 waveforms: triangle, trianle/sawtooth mix, sawtooth, 1:2 square, 1:3 retangular, and 1:4 rectangular. VCO 3 has a reverse sawtooth instead of the tri/saw mix. Only one waveform of each VCO can be used at a time. VCO3 has a switch that can disconnect it from keyboard control, which is useful when it is being used as a modulation source instead of an audio source. This switch also has the effect of increasing the range of the detuning knog to about 6 octaves.

The noise source is represented by a switch in the mixer section labeled "white/pink". With the switch in the "white" position, white noise is sent to the mixer and pink noise is sent to the modulation mix knob. With the switch in the "pink" position, pink noise is sent to the mixer and red noise is sent to the mod mix knob.

All routing in the Mini is accomplished using switches and knobs. No patchcords are used. The Mini has no memory, so changes to parameters must be made in real time.

One characteristic of all analog synths which is also present in the Mini is that of tuning instability. The Mini will be flat if it is cold. The unit comes up to within a quarter tone of its final pitch after being on for about 10 minutes, after which it will slowly drift up the other quarter tone over the next 20 minutes or so. It is recommended that when preparing to use the Mini in the studio or at a gig, the Mini be powered on before the rest of the equipment is set up.

If the temperature changes during performance because of cold weather or stage lighting, the pitch will be affected. The instrument is easily tuned by a master tuning knob on the front panel.

The Mini has four qualities which make it a viable contender for solo line and studio work, even today: Fat sound, the pitch wheel, the ASDR phrasing, and the ability to change all aspects of the sound easily during live performance.

The "fat sound" results from the type of filter used, and from the tuning instabilities. The pith wheel feels good because the deten is weak and because there is no "dead band". The ADSR phrasing was discussed above. Finally, each adjustable parameter is (necessarily) controlled by a dedicated knob or switch on the front panel.

Earlier Minis use discrete components (transistors) for the VCOs. Later Minis use ICs. Although the ICs improve tuning stability somewhat, they don't sound quite as "fat".

It is interesting to note that the Prophet 5's front panel and architecture was modelled after the Mini. However, the Prophet doesn't have that "fat" sound, probably mostly due to differences in the filter design (the Moog filter was patented.) Even now, after the filter patent has expired, there are no synths that (to my knowledge) sound as "fat" as the Mini. This is probably due to the use of "better" VCOs, which track each other closely and do not drift as much as the old transistor ones did.

Every couple of years, the oscillators need to be tuned. Inside the case there are 7 trimmer pots used for tuning. Each VCO has two, labeled "range" and "scale". The seventh pot is used to calibrate the octave switches.

The VCOs are tuned by first tuning the "range" pot so that an extremely high note is in tume with an external pitch reference. Then the "scale" pot is tuned to adjust the frequency of a low note. Since adjusting the scale affects the range, tuning is somewhat of an iterative process. Each VCO is tuned independently.

I tune VCO1 exactly, using a strobe tuner. I then tune the other two VCOs to it by ear. I stretch tune VCO2 a little by tuning the high end to VCO1 exactly and then tuning the low end so that it is a little flat, beating against VCO1 to a pleasurable degree. I tune VCO3 exactly to VCO1.

Steve Runyon (

# Moog Liberation

About a week ago, I was wandering through Johnny B. Goode's, my local used-gear store. John looked up at me as I came in, and smiled That Secret Smile. I KNEW that grin; it meant "Sorry, Metlay, I have something so WEIRD you CAN'T pass it up!" This was not good, especially considering the money I was about to spend on CDs, as John 3 has gleefully noted….

I wandered through the store, seeing what lay in wait. A Casio VZ-1 that he'd managed to sell for $500; a Yamaha DX-11 for about the same, a Roland S-10, a Yamaha DX-1 for $1000 (see how the mighty have fallen!)….

…and there it was. Nestled in a road case lined with blue fur, complete with multipin cable, power supply and manual– a Moog Liberation. I ran a hand over it disbelievingly, then hauled it out and hooked it up to a guitar amp and played with it for a while. Bliss! Quickly I checked it over. "Hey," I complained to John, "the pitch ribbon's broken."

"Yeah. So?" The Grin was still there– he knew I'd buy it anyway.

And I did. I'm crazy but I'm NOT stupid.

So now it sits on a temporary stand in my studio and gets a lot of love and attention. I don't know if I'll keep it; I cut my teeth on EMS Synthis and ARPs and Oberheims, and Moogs don't have a lot of mystique for me. But it IS kind of fun to have around.

You youngsters out there in digitoid MIDIland probably don't remember the Liberation. It was a comparatively late design in Moog's history, and quite an innovation at the time– while a number of strap-on keyboards existed already (the Syntar, the Clavitar/Clavitron, and the Probe come to mind immediately, as well as the gizmo Rick Wakeman played, anyone recall the name of it?), the Liberation was the first foray into the field by a major maker for a major market. It was available in 1979 and sold until about 1981, when Moog died. No strap-on keyboards were marketed after that until the MIDI era. (PArenthetical note: I wonder how hard it would be to get one of every strap-on ever, not including custom designs like the Probe? I'll append a list to the end of this review.)

The Liberation was designed to give the keyboardist a reasonable palette of sounds in a strap-on form. It was designed to be light, expressive, sonically powerful and flexible. Let's see what that meant in 1979, Okay?

The Liberation has a main body containing the keyboard, controls and synth guts (well, most of them), a long multipin cable, and a rack box with audio output and a monophonic CV/S-Trigger output set for running other synths from the Lib. The Lib has straplocks so you don't drop it, a guitar-like "neck" with performance controls, and a full front panel. It has the following features:

THE KEYBOARD is a 44-note F to C unweighted synth keyboard with full-sized keys. It feels kind of weird, because it's pressure sensitive! Yes, kids, one of the first monophonic pressure sensors ever is on the board– it's called a "force" sensor, terms like "mono pressure" and "Aftertouch" still being science fiction.

THE SYNTH is a standard Moog two-oscillator beastie, functionally identical to the Rogue, Prodigy, Taurus II and Taurus III– and also similar to the Realistic synth Moog built for Radio Shack. It has two VCOs, each with triangle, sawtooth or pulse wave selectable, each with a three-octave range and up to a fifth of detuning sharp or flat. There is oscillator sync, but no pulse width modulation: one oscillator has a square wave, the other a 10% duty cycle pulse. The sound sources include the two oscillators, a pink (!) noise source, a ring modulator that outputs sum and difference frequencies of the two oscillators, and a "poly" section. The latter is a fully polyphonic divide-down square wave setup with limited filter mod and no envelope shaping; it's good for thickening leads and having some chordal backing for the synth voice (which is high-note priority). The filter is a standard Moog post-Mini 4-pole lowpass jobber with cutoff, "emphasis" (i.e. resonance) and envelope amount controls.

MODULATION CONTROLS include one LFO and two envelopes, one for the filter and one for the VCA. The LFO has a range of 0.3 Hz to 30 Hz, and can output either a triangle, square, or random S&H voltages. It can be routed to either the oscillator pitches (both at once) or filter cutoff, and controlled by either the force bar or the mod wheel. It also has a retrigger mode switch, which triggers the envelope rhythmically in time with the LFO without any need to touch the keyboard. The envelopes are the typical braindead Moog "Contour Generators" as they're called, with individual sliders for attack time, sustain level, and a COMMON time shared by decay and release. The only way to defeat this is to flip a switch that sets both envelopes to cut off immediately at key-off. Oh, and there are LEDs that light up when the LFO is running and the envelopes are triggered. Way cool. The unit also has a Glide control (portamento time) and separate tuning controls for the synth and the poly section.

LEFT-HAND CONTROLS are where this machine really shines; it puts the rest of the controllers out there to shame, even nowadays. Only the Roland Axis comes close to jamming as many controls onto the left hand as the Liberation managed. From the tip of the "neck," there is a rocker switch to determine whether the force bar would apply a direct control voltage or serve as a sidechain control for the LFO (where the modulation goes is controlled on the front panel), a wheel for setting the amount of force-bar voltage, a second rocker switch to enable or disable glide (rate is set on the front panel), Moog pitch ribbon (mine needs to be replaced, alas, and like a fool I deleted that address for Moog parts that's been posted here a good 23 skidillion times), and THREE MORE WHEELS– a modulation amount wheel, a filter swell wheel, and a volume wheel. The filter wheel has a short throw and is spring loaded to return to zero, and the other two are free turning with a lot more travel than the wheels on any MIDI controller. They were obviously designed to be set and left as is for a while, so their resolution is very good. Nobody seems to thumb mod wheels and leave them thumbed much these days– on Roland and Oberheim boards, it can't even be done, as the mod devices are springloaded. Sigh.

So how does it sound? Great! I wish I'd had one in 1982. The synth voice is classic Moog– rich and gritty and smooth at the same time. The poly and ring mod add a lot to the sound as well, and it overdrives nicely. I plan to try it through a flanger and fuzzbox at some point.

How does it FEEL? Incredible. The controls are well-thought-out, many use color-coded sliders that run both vertically and horizontally, so it's easy to find where you are by feel, and the neck works like a dream. The only board that comes close ismy Yamaha KX-5, which has a smaller neck and fewer easy-to-reach controls.

Do I have a gripe about it? Yes, the same one that forced me to get rid of my Prophet T8– it's HEAVY! They claim that the Lib only (!) weighs about 14 pounds– compare that to a Rickenbacker 4001 bass at 12 pounds, a Les Paul at under 8 pounds, or my Yamaha SH-101 which barely weighs THREE pounds. Ouch! I can't wear it for more than about 20 minutes before my back starts hurting. This may, above all, force me to either let it go or perform radical surgery on it to lighten it a bit. The force bar is very loose and springy with a LONG travel, and when it's depressed it changes the throw of the keys a bit. And its sound, while great, isn't quite up to par with my Xpander. | but a dragon can only be painted….

From the University of Chicago "Chronicle":

Eaton, Moog to unveil 'revolutionary' keyboard

John Eaton, Professor in Music, and Robert Moog, designer of the

Moog Synthesizer, will introduce a revolutionary microchip-enhanced keyboard during a Music Department colloquium at 3 p.m., Friday, Jan 31, in Goodspeed Recital Hall. Moog will present a lecture on the instrument, "The Evolution of the Multiple-Touch-Sensitive Keyboard." The lecture is open to the public.

The result of a 20-year collaboration, the new Multiple-Touch-Sensitive

Keyboard was developed by Moog, based in part on ideas suggested by Eaton… The [device] is "the world's most sensitive musical instrument next to the human voice," according to Eaton. "Playing it is a kind of combination of playing a a very sensitive stringed instrument and playing a keyboard instrument."

Each key on the 49-note keyboard has a microchip in it that responds to

five specific motions: the precise distance it is depressed; the finger's front and back position and motion on each key; the finger's side-to-side position and motion on each key; the total area a flattened finger covers on each key; and pressure on a key after it is depressed fully.

These five, fully independent controls send signals in digital streams

of numbers to the computer, which routes the signals to affect any possible aspect of musical continuity desired–loudness; vibrato; tremolo; reverb; tone color or instrumental change; the speed, pitch, and any other application that can be dealt with by a modern sound synthesizer or sound-generating computer program. The keyboard is one of three that Moog will build for the University's Computer Music Studio.

"The keyboard can be connected to any sound-generating apparatus,"

said Eaton. "The possibilities are endless."

To demonstrate the instrument's possibilities, Eaton plans to use three

synthesizers that generate sound according to different principles. Eaton will present the first concert performance of the instrument on Friday, May 29…

The article continues, giving a brief bio of Eaton and his awards, etc. I am afraid that I couldn't attend the the demo, so this is pretty much all I know. Maybe if someone at UC knows more, they could tell us?

– Sean McCreary University of Colorado at Boulder

The sound of a Source is quite close to that of a Mini-Moog. It can store 16 Min-Moogs in fact. hehe… The synth has tape capabilities. It can also be midied for about $225 from Encore Electronics. Sysex data is available. I'm thinking of doing that quite soon, before they go out of business like so many other companies! hehe… Sometimes they can be finicky though with their Incremental Controller, but they are pretty easy to use. Comes with 2 sequences avaialable, arpeggiator. The sequences can pull the sounds from ANY OF THE 16 on board. Sort of multi-timbral (?) in a linear fashion. I haven't seen any other monophonic synths like this do that trick yet. Well, what's the price they're quoting?

tlb – Uucp: …{gatech,ames,rutgers}!ncar!asuvax!stjhmc!267!14!Tracy.Barber Internet:

# Roland Alpha Juno-2

DESCRIPTION: Digital-analog hybrid synthesizer (DCO/VCF/VCA). SYNTHESIS: Subtractive. KEYBOARD: 61-note, C to C. Velocity- and pressure-sensitive. SOUNDS: 128 onboard sounds (64 preset, 64 user-programmable). VOICES: 6-voice polyphonic. EXPANDIBILITY: 1 RAM cartridge slot (64 additional sounds). FEATURES: Octave transpose, chord memory, polyphonic portamento, "biometric" parameter adjustment. PRICE: $350-$700.


The Roland Alpha Juno-2 is a five-octave, six-voice polyphonic digital-analog hybrid synthesizer (DCO/VCF/VCA). It has a warm, analog sound, but its oscillators do not suffer from the tuning problems that often accompany voltage-controlled oscillators. The keyboard is velocity- and pressure-sensitive (channel aftertouch, if I am not mistaken), although you almost need a Mack truck to get anything at all out of the aftertouch sensitivity. The velocity sensitivity is ok, though; the Alpha Juno-2 synth was the first (and only) touch-sensitive Juno Roland made. Its little brother, the Alpha Juno-1, receives velocity and channel pressure information over MIDI, but its four-octave keyboard is not velocity- or pressure-sensitive. The rack-mount version, I believe, is the MKS-50.

It has only one oscillator per voice, but its sounds are plenty thick, thanks to its built-in chorus, sub-oscillator (a pulse wave 1 or 2 octaves below the main oscillator), and pulse-width modulation. It is a subtractive synth which sports pulse waves, sawtooth waves, and white noise. The four-stage envelope generator has adjustable levels and times (an improvement over the standard ADSR), but there is only one envelope and LFO per voice. Aside from the chorus, it has no built-in effects.

It is a performance-oriented synth; left-hand controls include the standard Roland pitchbend/modulation paddle (although the mod wheel is not throw-sensitive), two octave transpose buttons (normal and down an octave), polyphonic portamento, and chord memory. It also features real-time control over four groups of related parameters (a feature which Roland calls "biometrics" for all you sci-fi fans): Modulation Rate, Modulation Depth, Brilliance, and Envelope Times. "Biometrics" groups related parameters and makes them simultaneously adjustable via the Alpha-dial (were you wondering where that Roland term originated?…). For instance, adjusting the mod rate alters the LFO rate and the chorus rate; adjusting the envelope times lengthens or shortens all four of the time stages of the envelopes; adjusting the brilliance alters the waveform and/or the cutoff frequency of the lowpass filter; and so on.


Although it sounds like another Juno, it looks like a DX-7. Its predecessor, the Juno-106, had a ton of front panel sliders and knobs for real-time adjustment of parameters, but the Alpha Juno is much more streamlined: one dial (the Alpha-dial!) for modifying parameters. While the Juno-106 had only a two-digit patch display, the Alpha Juno could display patch names and parameters on its LCD. It was clearly designed to look like a DX-7, and while its touch- sensitivity and larger display are appreciated, it is unfortunate that Roland got rid of all of the front panel sliders.


It has 128 sounds onboard (64 preset, 64 user-programmable) with the option of accessing 64 more sounds via a much-too-expensive ($80) RAM cartridge. But the onboard sounds are great (some of them, anyway). Two of the pipe organs are very good (especially Pipe Organ 1); the acoustic bass and the synth basses are quite full-sounding; the Piano 2, although it sounds nothing like a piano, is a very usable PWMmed sound throughout the entire range of the keyboard (as is the Electric Piano 2); the steel drums are very good; and the saxophone sound is a surprisingly good mellow sax (for a non-PCM-sample-playback or -digital-waveform-generating synth). Its "synthy" sounds are good, too: lush filter sweeps and punchy synth sounds (has anyone else played "Jump" on P-54, Fat Synth?) Its weak suit is DX-7 style electric pianos (or acoustic pianos, for that matter). Don't look for them here. It has a few good, usable full-keyboard sounds, which I like a lot, but it is not a DX-7.


The back panel has stereo and headphone outs; MIDI In, Out, and Thru; sustain pedal input, pedal switch input (assignable to program switching, activating portamento, or activating chord memory mode), and a RAM cartridge slot. The Alpha Juno-1 also has a volume pedal input; I forget whether the Juno-2 does or not.


I bought an Alpha Juno-2 in 1986, about a year after it came out, as my first synth. It had gotten decent press, but was mostly upstaged by Roland's Super JX, Ensoniq's ESQ-1, and other fancier synths of the time. I liked the sounds a lot; I had compared it to the similarly-priced Korg DW-8000. As a first-time synth buyer, I found the punchy, warm sounds of the Juno to be preferable to the thinner sound of the Korg (although the Korg has probably retained its price and popularity more than the Juno has, and certainly has better electric piano sounds, as well as onboard effects). I remember comparing the pipe organ sounds and deciding that there was no contest, for the types of sounds I was looking for. I bought the Juno and have been happy with my purchase ever since. It is the only analog synth (or mostly analog synth) in my rig (I have a D-20 and a U-20), and I love the sounds I can get on it. I only wish, though, that it had onboard reverb. Oh, well. I guess you can't have everything. It was a good first synth, though (especially given the lousy Roland manuals – believe the rumors!), and I learned a lot about programming synthesizers from working with this synth.

Now, if I could just pick up a PG-100 programmer, so that I don't have to scroll through all of the parameters with the stupid Alpha- dial….

Nick Velharticky

# Roland JX-3P

This is a hybrid DCO/VCF/VCA synth released in about 1985. It was later rereleased as the Planet-S MIDI module. The architecture consists of two DCOs (the second being tunable), two VCFs (one static high-pass and one dynamic low-pass) and a VCA per voice. Polyphony runs to six voices. There is only one EG and LFO per voice and this can be irksome. Moreover, PWM is only possible on the second DCO which is also used as a noise generator. Hence it is impossible to get PWM and noise in the one patch.

At first listen, I fell in love with this synth. The sound was just so warm and vibrant that I couldn't help but buy it. Much of it relies upon the built- in stereo chorus effect which is almost indispensable in my own patches. The presets are competent but uninspired. There are strings, some brass, organs (including a very good B3 but no Leslie), electric pianos and the usual complement of analog synth sounds. None of these were awesome, but gave me plenty of scope to work with them. The preset "Filter Flow", a standard Moog-style filter sweep, was improved considerably after some minor tweaking. Overall, if one isn't prepared to program this synth, they might as well forget it. This, of course, means the almost compulsory purchase of the PG200 programmer. If a second-hand JX-3P doesn't have this then it might be a good bargaining chip to get the price down, as programming is extremely painful without it. I mainly use mine for meaty synth basses, filter sweeps and other analog synthetic textures. It really fills out a mix, especially when I play thick chords, and it complements most digital gear nicely.

MIDI implementation is atrocious (especially considering Yamaha's MIDI implementations of about the same period). The synth sends on channel 1 only and receives on omni. Reception can be changed to channel 1 by turning the synth off, connecting MIDI IN to MIDI OUT, turning the synth on again and playing a few notes. The JX-3P does not seem to respond to an omni off command sent over MIDI which is extremely irritating. No sysex, of course, and there is only a MIDI IN and MIDI OUT interface (no THRU). The keyboard is 61-key with no velocity or aftertouch although there was a ROM upgrade available that allowed the JX to accept MIDI velocity data and respond to it.

There are only two modulation controls on the outside of the synth - a pitch bend lever (like all roland ones) and a LFO trigger button. This is indicative of the rather limited modulation routings of the synthesis engine, as well. The one EG can be routed to pitch of either oscillator, filter cutoff (LPF only), VCA level and that's about it. Likewise the LFO can be routed to all of the above. You end up stretching your rescources a hell of a lot when you want a filter cutoff LFO acting on a voice using PWM with an EG controlled pitch variation on one oscillator while that EG also controls filter cutoff and VCA level…

Overall, I like my JX-3P. I like its warmth and character. I also like its gaudy appearance and quirky nature (the idiosyncratic way that it is so sensitive to power spikes and drops - try turning on/off an appliance in the same powerpoint as a '3P if you want to see what I mean). If I had the money (and the synth had a somewhat less sparse MIDI implemenation and I could program a decent choir sound) I'd buy ten of the things and a mixer and call it quits as concerns buying any more gear ever…Or maybe I'd just buy a Prophet T8 and be done with it. Despite its limitations, the JX-3P would have to rank in my "10 Greatest Synths of All-time" list.

Jon. | Every yo-yo wants one.

# Roland D-70 Super LA Synthesizer

The D-70 Super LA Synthesizer is a powerful 76-key, multi-timbral keyboard featuring unique sound creation capabilities, innovative performance features, and incredible new sounds. The D-70 introduces a new approach to the "Linear Arithmetic" concept, enabling you to develop complex, exceptional sounds and providing a new level of creative flexibility which makes it the perfect instrument for live performers or for studio musicians who need instant, complete control over sounds.

With its extensive multi-band filtering capabilities, the new DLM (Differential Loop Modulation) process which creates thousands of new distinctive, modulated waveforms, and a wide selection of new PCM samples, the D-70 provides you with extraordinary power. The revolutionary Tone Pallette editing system makes it easy to create new sounds instantly by allowing you to modify main synthesizer parameters such as level, attack, release, resonance, cutoff, and panning. By pressing a single button on the D-70's front panel, you can use the Tone Pallette's four sliders to control parameters for any four Tones at once. No other synthesizer offers you such power and flexibility.

The D-70 also contains a wide variety of impressive performance functions. Offering a large 40-digit x 8-line LCD which makes operation and editing easy and full MIDI control capabilities, the D-70 is the ideal master keyboard for any MIDI system. The keyboard lets you control release velocity in real time, in addition to velocity and channel aftertouch, and responds to polyphonic aftertouch through MIDI, giving you incredible expressive responsiveness. The Tone Pallette can be operated in real time as well, providing even greater expressive possibilities.

The D-70 contains PCM wave data, such as white noise, spectrums, and sawtooth waves, that can be used as sound elements. A variety of high- quality, multi-sampled PCM sounds including pianos, brass, guitars, drums and distinctive synthesizer textures are provided internally as well.

In addition, a new sound card developed exclusively for the D-70 features a large selection of PCM wave sound elements. The D-70 is also compatible with SN-U110 Series Sound Cards which provide access to a variety of exceptional multi-sampled instrument sounds that may be used as is, or edited as desired. Waveforms from the external cards as well as from internal PCM sounds can be modified extensively, giving you incredible power for creating sounds. DLM (Differential Loop Modulation), a unique, new editing feature can be used to truncate a waveform and process it with loop modulation. This allows you to create a wide assortment of totally new waveforms from the original PCM wave. These waveforms can then be filtered with a high-quality multi-band TVF filter. The D-70 enables you to control and modify sounds at every level, giving you the ability to create unique, sophisticated, and personalized sounds never available before.


Superb Selection of Built-In PCM Sounds

The D-70 includes a variety of sound elements such as white noise, sawtooth wave, and PCM short loop sounds, similar to those of the other D-Series synthesizers. In addition to these versatile sound elements, the D-70 contains a wide selection of multi-sampled PCM sounds ranging from lifelike acoustic instruments to analog and digital synth textures that offer sound quality comparable to professional samplers. The D-70 also allows you to modify these PCM sounds using DLM and TVF filtering for an even greater range of tonal possibilities.

Dedicated PCM Sound Elements Card and SN-U110 Series Sound Card Library.

In addition to the D-70's internal sound elements, a completely new Sound Card designed exclusively for the unit featuring PCM sound elements such as white noise and sawtooth waves can be inserted in the D-70's two ROM card slotsfor additional creative possibilities. The D-70 is also compatible with SN-U110 Series sound cards, a diverse collection of multi-sampled PCM sounds. PCM wave sound elements from the dedicated cards and sounds

from SN-U110 Series cards can also be TVF-filtered on the D-70 allowing you

to create an even wider variety of sounds.

DLM (Differential Loop Modulation) Enhances Sound Creation

DLM is a revolutionary feature that allows you to truncate PCM waveforms, define the start point and loop length, then process the waveforms with loop modulation, resulting in wave data with a new harmonic structure. The mode, start point, and loop length of the waveform can be set to create integral or non-integral harmonics, producing sounds that span the entire frequency range. The D-70's built-in PCM waveforms provide the raw materials suitable for effective filtering and DLM radically alters the

harmonics of these waveforms prior to the TVF filtering stage. By using a single PCM wave to create a varitey of different waveforms, DLM produces random, modulated sounds completely different from the original similar to a ring modulation effect. With these high level editing possibilities, you can build sounds from the ground up and customize PCM waveforms with unprecedented control.

From: Commander Brett Maraldo bmaraldo@WATSERV1.WATERLOO.EDU

Subject: Re: A D-70. Yeah. Or a WS. Hmm. Maybe a D-70…

I have owned a D70 for 7 months now.  Let me tell you about it.

First, the voice architecture: first you start with a 'tone'. A tone is a PCM sample that has an associated pitch ENV, Filter and env, and amplifier and env. Once you have your tones created you assemble up to 4 of them into a patch. A patch is 1 to 4 tones layered, or plit over the keyboard. Tones can be set to vel switch or mix. Once the patches are created you can specify up to 5 of these in a multi-timbral 'performance'. A performance is a set of five patches mapped to various midi channels, volumes and effects routings. the 6th voice in a performance is for the rhythm section. The rhythm section is actually quite powerful: you can specify, for each of the 76 keys, all the parameters of a tone. that is, there are 76 tones specifically mapped to the keys in order for the rhythm section so once you have your basic kits programmed you can create other unusual sounds and use those in the rhythm section (reversed sounds, elctronic zings and zaps, etc).

In my setup, I have local off set.  In the past (as of a week ago)

I would just play the keyboard and not care what channel the D70 was sending on because I would tell my sequencer to send all notes to a specific midi channel, controlling the performance patches or other midi devices. But I have been using the MIDI OUT page recently and now see how the D70 is a superior controller. With the MIDI OUT page you can, by a one button press, specify either LAYER, SPLIT, or ZONE mapping. You can also select which channel a zone will send on. And, you can turn out-sones on an off with the zone buttons:

s s s s upper 1 ch x mode

l  l  l  l	upper 2 ch y	   sw/mix/norm
i  i  i  i	lower 1 ch z	    mode
d  d  d  d	lower 2 ch $	   sw/mix/norm
e  e  e  e
      r  r  r  r	ZONE  LAYER  SPLIT 	     slider are physical sliders	
  • * * * * * * * are physical buttons

You use the alpha-dial (data dial to select the channels the various zones of the keyboard will send on for ch x,y,z and $). If you select LAYER (Simplest) then you will send on all of those midi channels. You can press the buttons under the sliders (each mapped to the uppers and lower zones), to turn on/off that midi out. Example: is x,y,z and $ are 1, 2, 3 and 4 and you are set to Layer and all the LEDs above the zone buttons are on then you will send to all of 1,2,3,4. If you press the 3rd zone button then you turn off the send to ch 3. If you select Split, then you can specify the split point of the upper1,2 and lower1,2 zones. so if all the sones are active (LEDSs on) and the split is at C4, then you will send on 1 and 2 on the lower and 3 and 4 on the upper. (Note: the midi channels can be set to control the internal performances patches or external midi devices. I use ch1-5 for the D70 patches, 6 is minimoog, 7 is prophet 5, 8 is MKS30, 9 is M1000, 10-14 is TX16W, 15 is XR 10, and 16 is D70 rhythm). Finally, if you slect Zone, you ca specify independent zones for uper 1, upper2, loer 1 and lower 2. You may have zone overlaps.

The way you specify the zones is great:  if you press the ZONE

or SPLIT button twice you get a graphic diaply shwing the keyboard and the zones as lines above the keyboard. By pressing and holding the sone button below the slider and pressing the upper and lower keyboard key you can specify the length and psotion fo the zone and immediately see a graphical rep of the zones. If a zone is deselected you will see it as a dashed line. An active zone is solid. very fast.

Finally, you can specify if the two voices in UPPER or LOWER will

be layed (norm), vel switched (SW and approriate vel threshold) or mixed (MIX and threshold). You can also use the LEVEL and TRANSPOSE buttons with the four sliders to set volumes and pitch transposes for the MIDI output even for external devices.

Sorry about the typos... I am feeling lazy.

ciao, Brett L Maraldo Plexus Productions

So, I've had this machine at home a day (the D-70, not the drummers' stool), and I'm slowly finding my way around the user interface. I thought I'd give you my first impressions.

PHYSICALLY: very nice. The 76-note keyboard is a *big* point in its favour. (I was after a partial replacement for the pf85, after all, and 5 octaves isn't really enough for a controller.) It's not very deep (front-to-back) so I'll have to get some longer Apex arms to bring it out so I can see the display when I'm using it live below the Wavestation. The D-70 has a very small cross-section, and looks a little strange alongside a Wavestation. The controllers feel much nicer than the D-50 - less spongy. The keyboard is better (but, it could not have been worse), and passable for piano solos (I couldn't possibly play a piano part on the D-50). The front panel buttons are so-so. Most are fine, but the function keys are shallow travel, creak a little, and feel a little flimsy. I have a phobia about broken microswitches after having a Mac mouse die on me and a fault develop in my replacement trackball, so I have a feeling I might get a failure here sometime.

SOUND: Sounds like $1,000,000. Well, sounds like $3500 which is what it lists for. Well, sounds like a D-50 but with a *lot* more punch and clarity, a smoother and more blended sound (since it's applying multimode filtering to the sample data rather than layering synth and unfiltered sample partials), a superb bottom end compared to the D-50, with the choir and bell samples transposing down well to the bottom of its 76-note board (I don't know yet whether it challenges the VFX as digital-synth-with-the-biggest-bottom until I get into programming it) - I can make sounds like on T. Dream's LEGEND soundtrack. The high-end is different to the D-50, mainly because it's using highpass resonant filtering for animation rather than lots of LFO mod. This is really where the D-50 and D-70 diverge in terms of capabilities. The '70 seems to have more warmth and character than the WS, and out-basses it as far as I can tell.

VOICE ARCHITECTURE: Quite respectable. Less LFO modulation than the D-50, simpler envelopes, no synth features like PWM. No ring-mod (Differential Phase Thingummy instead). But, of course, the thing which sold me on it when I heard it last year is the filtering. It's mistakable for analog resonant filtering unless you listen closely. How closely is a matter I have to investigate. Doing multimode resonant filtering on sample data is sweet. Very very sweet. I like it a lot. It suits the way I work, which involves putting sound elements into specific parts of the frequency spectrum. I enjoyed this on the VFX, and find it's lack to be my single biggest cause of frustration with the WS. Oh, the D-70's filter resonance can be overdriven in an interesting way, I notice.

Here's where the D-70 has its only serious weakness. It only has 30 voices. Without more LFO modulation facilities, I'll probably be wanting to do more layering of tones to add animation, and I'll be running out of voices pretty quickly, especially with multi-channel input. How much of a problem this is, remains to be seen.

PERFORMANCE HEIRARCHY: Despite KEYBOARD's slamming, I find this to be rather good. The Tone Pallette sliders make incremental changes at the patch level (I think) to already-programmed tones, so that sounds can be modified in most major respects very quickly. (The LFO's aren't here, which is a bit of a shame - they're lower down.) All four tones are available for these changes, with graphical feedback, in parallel. This is very nice. KEYBOARD slammed the machine for not having cross-referencing tools for the tones. In fact, the Wavestation has the same problem, but has a clearer user interface for seeing what's where, which is why it's not a problem so much on the WS.

The multitimbral operation looks fairly sensible, although the user interface for walking around the MIDI parameters is hellish at times. I'll investigate this further, but so far all I've done with it MIDI-wise is checked controller transmission and SysEx'ed off all the internal memory.

Hey, the word 'SysEx' verbs quite well.

USER INTERFACE: A combination of superb and frustrating. The performance-level stuff (Tone Pallette, dedicated editing buttons) is superb, making sound design and modification quick, intuitive, painless, etc. Getting around the lower levels of the interface heirarchy is like having teeth pulled. It's not possible to jump around between tones as you'd like. Some parameters appear in several pages. Some pages appear more than once on separate programmer buttons, with different page titles (Gnnnh!). So, it'll take some getting used to. Like I say, a nice performance interface with some sh*tty interfacing below ground level…

The manual is fair. Nothing more. But then, it would have to be a very good manual to explain how the user interface heirarchy works - and it doesn't explain this very well at all. It took me a few hours and several cups of tea to figure it out; the most helpful part of the manual was the SysEx memory maps in the appendix.

MIDI CONTROL: Surprisingly good, given Roland's usual reputation here. The standard controllers can all be remapped at the performance (or is it patch?) level for the incoming reception parts. In addition, a performance has a transmission MIDI pallette which maps all the outgoing controllers. Zones can be layered for MIDI (although local patches can't be layered in ways you might expect) and send volume and patch changes. Notes aren't held under performance changes, which is a shame but not surprising. Oh: solo and portamento are supported, and the D-70 has *fingered portamento without retriggering*! Hooray, the feature that until now only Yamaha keyboards have gotten right. Now I have no excuse to buy a TX81Z. (But, who needs an excuse?)

SOFTWARE: Yes. It's true. The LFO's slow down when you play chords. I have to tell you, there's nothing more discouraging than powering up a $3500 synthesiser and having it announce "Software Version 1.00" at you. But then, I've been burned by Ensoniq. So, there's the LFO performance problem and a couple of the interface shortcuts seem not to work. According to Roland UK, I can just install the 1.17 ROMs into it; there's some confusion about whether there's a faster CPU upgrade as well. Roland tell me that I can just put 1.17 into this machine without a fast CPU, but wouldn't deny the existence of a fast CPU upgrade. I'll take the machine to 1.17 and see how well it performs. If it's too sluggish, I'll call Roland again and see what a new CPU will cost me.

FACTORY PATCHES: Usual assortment of Roland Rubbish with silly names. The previous owner had made an attempt at programming with about half a dozen patches called PIANO, CHOIR, SYNTH but otherwise the factory patches were intact. The worst thing is the bldy demo sequence. I don't want bldy Eric Persing playing sappy New Age fusion music on my synthesiser. I object to the legend "PCM Play" being inscribed on the front panel and the potted biography of Mr. Persing, one of L.A.'s hottest session players with TV commercials to his name. I mean, well, *yawn*.

STUPIDITIES: The D-70 doesn't have an IEC/Euro connector on the back. The mains cable is grommetted directly into the case. STOOPID for a live keyboard. This will be coming off in exchange for an IEC as soon as I can find my wire cutters…

– Nick Rothwell | "That's the Waldorf MicroWave set for General MIDI."   |   "Yeah. You got a problem with that?"

I bet you thought the D-70 was a Super LA synth, didn't you? Replacement flagship to take over from the D-50 and all that. Wrong-o. It so happens that I had the hood up on my D-70 to upgrade the ROMs, and I noticed something interesting. On all the important boards (main digital, analog subboard, PSU) there's the usual printing of the Roland logo and so on, but the D-70 legend is stuck on. Hmm, thought I, and peeled off one of the labels. Wanna guess what it says underneath?


So, the much-vaunted mega Linear Arithmetic synth is no such thing. It's not the top of the LA range, it's the top of the PCM sample player range. Roland obviously decided at the last moment that marketing a new "synthesiser" in the D-50 line was a better idea than another sample player, so the name was changed everywhere (well, apart from the id on the main chassis, which they missed).

This is no great surprise to some of us. I always thought the D-70 was a glorified U-series machine. Brett Maraldo agrees. It has the same sort-of profile as the U-20, and even comes boxes with the same end cheeks. But I never realised that it came so close to being the Roland U-50.

Now, I hope you'll excuse me as I slope off and hide my head in shame at being, after all my efforts, the owner of Yet Another Sample Playback Board…


The sound of the D-70 was the first thing that struck me when I had a play with one. Not the warm (will I get flamed to ashes for calling a digital synth "warm"?), clear sound of the D-50 and not the crystalline, thin sound of the D-5/10/20/110 (although it's *much* nicer than the latter) but a pleasant (if slightly dull) sound that reminded me of an M1's pads with a little more grunt. The acoustic instruments were about as true as the U-20's. In fact, don't they sound rather like the…? Oh, yeah. There was that (finishes aside). The D-70 takes U series sample cards. Surely that would have caused some suspicions?

One might also have noted the lack of pulse width modulation, attack transients (as I recall, all of the D-70's "tones" are single cycle waves or complete samples) and most of the EQ was another sign that all was not as it seemed.

It must, of course, be noted that memory is a lot cheaper than it was in the heady days when the D-50 was released. The main reason for the existence of L/A synthesis (Roland's version thereof, anyway) was that the cost of memory required for the storage of complete samples was prohibitive. Without the cost factor we see complete samples being included.

What about the filters? I've heard it rumoured that they're quite good. One might be forgiven for thinking that Roland looked at the U-20, gave it some filters, did some jiggery-pokery with the LFOs (are there two on the D-70?) and released the beast. Roland might be losing its technical edge but the D-70's still (in my most humble opinion) one of the nicest sounding boxes out there. I'm not sure how useful DLM and "analog feel" are as synthesis tools but I think that the prescence of the former might just allow the D-70 to be called "synthesizer", even if the synthesis method isn't really L/A.


Jonathan Elliott | Sole and founding member, AMALGM - University of Tasmania | Analog Maniacs Against Lamentable | General MIDI - Join today!

# Roland U-20 keyboard

This unit is a PCM sample playback unit.  It has good effects

offering both chorus and reverb. It has the ability to modify the ASDR of the samples. The unit has a combination pitch and modulation "wheel". You have to push back to use it for modulation and simultaneously move it right and left. It has a volume slider and two uncommitted sliders that can be used to control about 15 parameters. The keyboard is 61 keys. The rackmount version is the U-220.

The architecture is particularly hard to understand and

gives new owners fits. The ROM PCM samples are called "tones" that are assigned to "timbers" where various modifications can be applied. These "timbers" are then assigned to a "sound patch" which has six slots for regular "timbers" and one slot for a special "rhythm" set, for a total of seven slots each of which can be a unique MIDI channel. Here additional modifications can be applied to change the final sound. "Keyboard patches" determine how the various transmitters of MIDI data perform as companions to the "Sound patches" that determine how the various MIDI recievers perform. If you are now confused, you get the idea.

There is excellent control over assigning splits and

layers using these six "timber" slots. You can also use velocity to switch or mix layers based on how fast you attack the keys. The keyboard can be mapped onto any MIDI channel and it does send aftertouch information. There are two jacks on the back for plugging in a sustain type pedal and an expression type pedal.

There is generally good controller programming; however,

the U-20 has not worked well with wind controllers due to an obscure problem with how it is set up to use controller information to control "timber level". Thus it does not follow controller #2 messages in real time. A wind controller would need to send either aftertouch or controller #7 messages instead of the usual breath controller message #2.

The PCM cards are unique and a very limited third

party offering exists. However, the 15 cards that do exist are good. There are about five Roland synths that use these same cards, so there is some motivation for Roland to issue new ones occasionally. The U-20 offers excellent sounds, especially of "real" instruments. It may be the best "real" sounding instrument in its price range.

The manual has excellent technical information, but

is of questionable use in understanding the architecture and how to program the U-20. There are two third-party manuals available that, while lacking in technical data, do offer chatty and readable descriptions of U-20 button pushing. The programming of the unit with the buttons and the tiny screen is still difficult. Only one company, SoundQuest, is known to make an editor for this machine. It is extensively buggy, but much better than nothing. The ability to see the architecture all together on a PC screen helps clear up how things fit together to make the final sound.

A RAM card slot and two PCM card slots are available.

Performance combinations are stored in "keyboard patches" all of which are initially set the same. There is one "keyboard patch" for each of the "sound patches". "Sound patches" are arranged in two 8x8 arrays, or a total of 128 internal and another 128 on the RAM card. There is room for four of the special rhythm sets. There are two definable chord sets available at a time from a total set of eight defined chord sets stored. The U-20 does not support microtuning in hardware.

PCM ROM cards are a problem if you want to use sounds

off more than two cards plus the internal tones. There is no way to copy the PCM ROM samples, so the way they come arranged on the cards is what you are stuck with. Due to the size of good samples, a PCM card can have as few as eight to ten "tones" on it. It has a fairly good support for sound effects with some internal sounds and additional sound effects on PCM cards available.

SUMMARY: The Roland U-20 offers very good "realistic" sounds with adequate control over their performance.

It offers good external controllers and has stereo and mono out,

both before and after effects processing. There is a limited but adequate number of additional PCM cards available.

The manual is OK, but a third party manual is recomended.  It is

a hard unit to intially understand. It does not seem workable with a wind controller.

It does work very well with an external drum pad unit

like the Roland PAD-5 Controller or with BiaB. The rhythm section is particularly well done.

It has good layering and split capability, including key velocity

switching and mixing.

NOTE: Roland has perhaps the most widely hated and poorly regarded customer support in the United States. There are people who refuse to consider Roland equipment for this reason alone, regardless of how good the hardware is.

Mike Burger

# Roland U-20

The U-20 review was very nice; however I would like you to add one thing to it, namely, that the aftertouch requires an enormous amount of pressure to trigger adequately. I found this to be a mojor drawback, since I difinitely bought the unit because of that feature, among others.

There is a mod that can be done, I have heard, which consists of replacing a resistor, but I sure didn't want to try it, but a repair shop could probably do it.


# Waldorf MicroWave

DISCLAIMER 1: Well, this isn't REALLY a full review of the Waldorf MicroWave, but it's the best anyone is going to get for a while from ME. Trying to learn new gear while singlehandedly keeping a research group going and writing one's PhD thesis is at best a nontrivial task. In the month-plus that I've owned the MW, I've worked with it for barely ten hours. But those ten hours were enough for me to formulate some initial impressions to pass along, as I had promised. (Thanks to Neil Weinstock for booting me in the ass on this.)

DISCLAIMER 2: This is a Metlay review. You should know by now what that means. No flames, people: I honestly don't care if you agree with my opinion of digital, sample-based Black Boxes (TM). So stuff it, okay?

Now then:


When you open the box, you find the MW itself, a set of optional stick-on rubber footsies if you aren't going to rackmount it, and no fewer than THREE manuals. One is the Performance manual, which explains the architecture of Single and Multi Patch organization, MIDI setup, and the things you need to know to get started. The second is the Programming Manual, which takes you inside the voice structure itself. The last and thinnest manual is the update for software rev 1.2x, whatever it is you get. Mine was 1.20; I understand that 1.23 is now shipping, which allows WaveSlave compatibility.

The manuals themselves are a joy to use. Their only weak point is that they were obviously typeset on a Mac with very primitive software– they look UGLY. However, Roland manuals are uniformly gorgeous, and aren't worth the paper they're printed on, so this is a minor point. The manuals are clearly organized, well-indexed, and loaded with useful data; the translation from the German was obviously done by either an American with German as a second language, or by a German who'd been speaking English for decades AND living in America for part of that time– there are a number of jokes sprinkled through the manual, obviously aimed at an American audience (references to covering food before putting it inside wouldn't mean much to Germans). Each type of modulation has at least two examples given, that are musically sensible and clear to understand. I would instantly recommend this box, along with the Xpander (if you also had the MAtrix-12's manual, which is much better than the Xpander's), as a GREAT first subtractive synthesizer for the learner, just from the ability of the manual to communicate worthwhile knowledge.

Having perused the manuals once through (I recommend this), you hook up your MIDI controller and turn on the MW. It CAN give stereo output from one of the two audio jacks, but the level is fairly low: it sounds much better through a sound system rather than headphones.

The first thing you'll notice, as has been mentioned by Nick Rothwell and various other people, is that the presets are almost uniformly LOUSY. I mean, God AWFUL! They communicate NONE of the capabilities of this machine, and whoever programmed them should be shot in the throat and left to die. Playing one of these in a music store, with the presets the only available avenue of listening and something like an SY99 right next to you, is NO way to sell MicroWaves– if I had heard it before I bought it, I would have saved my money. So, learn to initialize voices, and DO SO. Clear EVERYTHING and start from scratch, you won't be sorry you did.


The Micro's front panel is a bit sparse for my tastes, but the pages are organized in a fairly reasonable manner and the logic of the Micro's tree structure is a lot more snesible than most other synths out there. Basically you start with a 4x4 grid of pages, each with varying depths. They're grouped into four rows based on the sorts of activities you'll want to switch between (i.e. all the voice editing parameters are in one row, all the fast-edit parameters are in one row, etc.), so that you can jump around with minimum hassle. You choose your row with a big button marked MODE that cycles around the rows. Within each row, you pick a column (and hence a page) by punching one of four buttons. If the page you're looking at has several subpages (some do), then you punch the column button repeatedly to cycle through them until you find the one you want.

Within each page or subpage, there are the parameters themselves, each with its own value. You access and change these with the red button and Big Red Knob (TM) over at the left side of the machine, by the display. The cursor will be either under the parameter or its value– you use the button to move the cursor betwen the two fields, and the knob to dial in what you want.

So, as an example, an edit to Filter Cutoff would be:

Punch MODE until you're in the Voice Editing row (Row #3) Hit the second column button; if the display says VOLUME, hit it again Put the cursor under the parameter in the window, dial until CUTOFF appears Punch the red button, dial in the value you want.

Sounds nasty, doesn't it? Well, I've got news for you: compared to trying to figure out where the up-arrow is going to put the cursor on the Wavestation's Mix-Envelope page, it's a CINCH. The interaction of the controls is logical and works exactly the same way for all actions on the synth (with some exceptions that make a lot of sense themselves); in less than a half hour, you'll be FLYING. Especially since when you're doing something like setting up Multi Patches or editing sounds, the number of keystrokes to get somewhere drops precipitously. I do not LIKE this setup– I prefer a full front panel– but it does work, and work well, as opposed to something like the SY77, ugh.


Each voice (there are eight) has two Oscillators driving two Wavetables running into a Filter and then to a Pan unit.

The oscillators each have octave, pitch and detune (five octaves-plus range on each), pitch keyboard-track defeat, and two pitch modulation sources, one with a sidechain control and one straight in. The second modulator can be quantized for glissandi. Note that this is all just PITCH control; waveform control is in the Wave modules, that come next.

The Waves have a common Wavetable (the unit comes with 32 and has room for 12 more) and individual Start Wave, Startsample, hardwired modulation settings for Wave Envelope, keyboard tracking, and velocity scaling, and two more modulators, one with a sidechain input. The stepping mode can be set smooth or stepped. Since the Waves are the most unique part of the Micro's sound, I'll talk more about them later.

The mixer lets you blend both Osc/Waves with a white noise source; the mixer can be overdriven if desired. The VOlume page also has modulations for Volume envelope, velocity, keyboard tracking, and two general mods, one with sidechain. MIDI Controller 7 always controls overall volume.

The filter is a four-pole lowpass analog filter with cutoff, resonance, modulation of cutoff by the Filter envelope, velocity, keyboard track, and two modulators, one with sidechain. The resonance also has its own mod source (As the manual says, "Yes, it can be done."), and goes into self-oscillation very nicely, thank you.

The panning module sets a voice's stero position and has its own modulator. Normally voices are set to center in Single mode, but in Multis you have the option of spreading different Singles out a bit. MIDI COntroller 10 always controls voice pan, as an added modulation to existing ones.

There are two LFOs. Each has rate control, shape and symmetry controls that provide many different waveforms (over 300 of them, actually), a rate-randomizer, rate and level modulators, switchable LFO sync, and a simple three stage envelope especially for LFO amount.

There are three envelopes, labeled Volume, Filter and Wave. They are hardwired but defeatable to the pages named, and can also be sent anywhere else if desired. The Volume envelope is ADSR, the Filter envelope is DADSR, and the Wave Envelope is eight-time/eight-level with looping and variable sustain point. It can do either sustain or release loops, and has modulator inputs for times and levels. The Wave envelope has global modulation to all times and levels at once; the other envelopes have individual modulators and amounts for each stage.

The unit also has Glide (equal rate or equal time, choosable), several temperaments including user-defined tuning tables, and can store names of up to 16 ASCII characters.

There are also Macro and Fast-Edit pages. I won't describe them in detail, but they allow you to call up several preset envelope shapes and/or modulation types, enter them instantaneously, and modify them or existing envelopes/modulations with special screens (an exception to the normal editing modes) that let you fly in new values in a flash. If you're not interested in exactitude, and want to hack out a useful sound in a hurry, these modes make it easy. (And if you're a perfectionist, you always have the option of going in and tweaking the parameters in the regular edit mode.)


The MicroWave has nine edit buffers. If you edit a patch and leave it to go to another patch, you do NOT lose your edits; they;re waiting there when you come back. This is LOVELY; I wish the Xpander did this. Even the VS's Review buffer must be manually saved before hunting around. So you can be editing up to eight patches at once! The column buttons double as a shift key and a set of three memory-management tools (Store, Compare, REcall); this assures that nothing involving memory can be done in the course of normal keypresses, and that a save or deletion is carefully premeditated.


Each Multi PAtch (like SIngles, there are 64) has one to eight Instruments. Each Instrument has its own Single Patch, MIDI Channel, key and velocity windows, velocity response curve, transpose/detune, temperament, volume and pan position (overlaying that of the Single patch itself).Each Instrument can be routed to the stereo outs or to one of the four individual outs. Each Instrument has its own MIDI input filters, and so on, and so on. Voice allocation is dynamic. There are four generic Controllers called CTRL W, X, Y and Z. These have different effects on each Instrument, but are assigned globally. In addition, the Micro understands Velocity, Release Velocity, Pressure (Mono AND/OR poly), Pitch Bend, and MIDI Controllers 1,2,7,10,64,65, and the numbers you pick for W,X,Y,Z. A few of these are hardwired: 7 to volume, 10 to pan, 64 to sustain, 65 to portamento switching (defeatable). Oh, I almost forgot. Glide can be in half-steps (glissando) if desired. My Micro currently understands everything in my rack, including the joystick on my VS and (if I get one) the four front-panel sliders on the Roland D-70. And the controllers can be routed anywhere….


A bit more detail is needed on the Wave modules. Here's how they work, sorta. Each Wavetable (there are 32 preset, 12 blank that can be loaded via MIDI by the user) has 64 Waves in it. OK so far? 32 Tables, 64 Waves. Now, of those 64 waves, the last three are always the same: 61 is a triangle wave, 62 is a square wave, and 63 is a sawtooth. But Waves 0 through 60 are DIFFERENT for each Table. Where these waves come from isn't important: some are stored in ROM, others are interpolated

from the ROM Waves by an algorithm kept in a mayonnaise jar under Wolfgang

Palm's bed and guarded by attack dogs. What IS important is that the MW is capable of shifting from Wave to Wave in real time, under tha control of any modulation source you wish! This can be a smooth shift or a stepped one: unlike the Wavestation, smoothing doesn't halve polyphony. Each Wave has its own unique character: there are smooth waves, grimy waves, ringing waves, blah waves, formants, transients, jarring noises. And they get even MORE character when swept around in Wavetables. Now, some sounds can benefit from using only one wave at a time; others might use a small range of Waves; using an etire Table isn't always the best thing to do, as the jumps from sound to sound can be jarring.

What are the Tables like? Well, my notes aren't much help:

"1 Hollow filter sweep sorta 3 Buzzy Inverted ring hollow !!!!! 4 Saw to Hollow and back? 5 Rectified? 10 high Feedback 12 Phonemish? 13 Evolution– Animated! 14 Buzzy Rattly Bright!!! 15 Yanking Drawbars on the L-100 17 Multilooped Swell! 21 LEAPing OCTAVE resonances!!!!! 23 Hiccupping Organ twisty loop!! 24 Analogish filter Twist 27 Ragged S/H!! 28 Glittery Arpeg Filter with nasty tip 30 Beware of BIG THUNK at end!!"

…and so on. You have to hear them to understand.

I could spend a year just trying different Wave setups in the same initialized patch, never touching anything else in the architecture.


The unit is very forgiving of mistakes, with one small but important exception: if you switch to Multi mode in midsession, you CAN lose your edits if you're not careful. There's no volume control, except as an (admittedly fast-to-access) parameter in the grid. MIDI operation appears flawless, but Nick is hammering on his much harder than I do on mine, so it's not for me to say. Sustained notes don't quit when you switch programs, but on rare occasions they can hiccup momentarily. The buttons activate NOT when you push them, but when you LET GO of them –weird until you get used to it. The Micro's built like a tank, solid and heavy.


The MicroWave is not the ultimate synthesizer, or the ultimate anything. If it had four oscillators per voice with a vector joystick and an arpeggiator like the Prophet VS's, wave sequencing like the Wavestation's, and the modulation matrix and added features of the Xpander, then we'd REALLY have something. But let's not get ridiculous here. It does give you a hell of a lot of control, especially via MIDI. It's relatively bugfree – I haven't hit a bug yet, and Nick has only reported a couple. And it sounds


Hm. This is the part of the review I was dreading. How DOES it sound?


Let me put it this way– I have not yet (emphasis on the YET) come up with a sound that has sat me back in my chair, saying, "Oh, baby, THAT'S why I bought this machine!" Such an immediate, visceral response to the machine is important to me; in the past, I have gotten them from the modular drones and analog brasses and wild effects on my Xpander, the silky strings and hammering pads on my VS, and the gorgeous electric pianosynth on my EX-8000. But from the MicroWave– nothing. YET!

The difference was, that the other synths I've named were programmed by people rather than chimps, and were acquired back in the dim days before my doctoral thesis, when I could actually sit and work with my gear for days at a time without being distracted or feeling guilty. The Micro has so much depth and so much power that if I had some decent presets to at least HINT at where to start, and some time to really work with the machine, I KNOW I could get it to scream for me. But the patches I have created so far are heavy-handed and wearing, by and large. I need to learn to paint with smaller brushstrokes. I have created about a dozen patches so far. Some are tutorial in nature, designed to remind me of how the Wave Envelope looping works (by the way, Nick, I just got the bill for that panicky phone call to Scotland last month. You DON'T want to know!) or where waves are in the tables. Others are unimpressive; the strings are no good, and the wavescanned bass is rather slapdash. But there is also a burbling drone pad that sounds like the background to Edgar Froese's "Aqua," and a Moog Taurus that rattles the fillings….

I'm getting closer. Slowly, painfully closer, at a speed that is frustrating as hell. But my frustration, ultimately, is at MYSELF, for having gotten rusty. The Xpander and VS are like riding a bike at this point– I sit down, I turn some knobs, I get the sound I heard in my head at the start. And to an extent, I can get the Micro to sound like an Xpander. But what's the point of that? I am trying to push BEYOND what my other gear can do, and THAT'S what is so time-consuming and agonizing. There IS something there; getting to it hurts me.

So. My recommendations? They're fairly simple. The MicroWave costs about a thousand dollars. Basically it's an eight-voice analog synth with some digital tricks no other instrument can match and a great multitimbral MIDI implementation. Its presets stink, and should be cleared at once. It has no effects processing; I don't believe it needs it.

If you want an analog synth with sonic and programming power, and you're not afraid of starting from scratch– BUY THIS MACHINE. In these days of $1700 Xpanders, $1000 Prophets and $1200 MIDIed Minimoogs, it's an absolutely unbeatable buy. For your money, you will not find more power (whether that power is audio or MIDI).

If you need gooey effects and nice realistic presets, buy something else. Now if you'll excuse me, I'm not going to be able to sleep tonight if I can't get this (^@&#%$#*! string patch to work…. | –sound advice from the Nickmeister

# Yamaha CS-50

The Yamaha CS series of analog synths ranged from the CS-5, a remarkably powerful little one-voice one-VCO synth somewhat like the Korg MS-10 or the Micromoog, up through the CS-15, CS-20, CS-30 (beloved instrument of Mark Shreeve, the UK's second greatest electronic musician ever), and up and up to the CS-80 and the later CS-70M, which was Yamaha's last analog monster. I may have missed or miswritten a couple of numbers, having only worked with a 5 before, but the basic trend, as it was with the CP electronic pianos and the SK organ/string machines, was that the higher the number, the greater the power. The CS-30 was a very, VERY capable instrument, and I would expect the CS-50 to be as good or better, with a quite respectable sound. How much does he want for it?

BTW, it may be rare like Kawai SX240's are rare, but it's not really RARE <nudge nudge wink wink> like Minimoogs or TR808's are RARE <eh eh>. | (n. planer)

Subject: Re: Comments on Yamaha CS-50

Sorry, but you've got the chronology (and technology) wrong: Around 1975/6 the CS-50,60 & 80 came out. They all share common electronics.

CS-50:  4 octave, velocity sensative keyboard, 4 note poly,
1 changable preset (front panel knobs). Seen lately for about $100
CS-60:   5 octave, velocity sensative keyboard, 8 note poly,
      2 changable presets (1 front panel knobs + 1 mini knobs). 
Adds the wonderful pitch-bend ribbon!!!  Originally $2500,
got mine in '81 for $1000, Seen lately for $200 to $400
CS-80:   5 octave, velocity & POLYPHONIC PRESSURE keyboard,
8 note polyphonic, 2 sounds at once (basically 2 CS-60 from
one keyboard- layered/not split), 6 changable presets
(2 front panel knobs + 4 mini knobs).  Comes with wheels
that insert in back for rolling around.   Originally $7000,
got mine in '85 for $1400, Seen lately for $1000 to $1500
All of these are monsters (100-220 lbs).

Around '77/78 the CS-5, 10, 15, 20M & 40M came out. These were all Mini-Moogish lead synths ranging from the CS-5 (2 octave, 1 note, no memory) to the CS-40M (2 note, RAM based programable presets).

The CS-70M (M is for memory) came out in 80/81 to replace the CS-80. It has very little to do, in the way of sound generation and user interface, with the CS-80. It was basically Yamaha's Prophet 5. Pretty useless- haven't seen one since '82. Right around this time the GS-1 & GS-2 came out which lead to the CE-20, DX-1 (nice machine!), the DX-7 and therefore the end of useful synthesizers from Yamaha.

If anyone wants more info just ask.


Subject: Yamaha CS-30…

The CS-30 ? I have one of them…. It looks impressive and weighs a lot. The underside is made of wood !!!! It has a weird quasi-user-hostile interface when it comes to making sense of the signal path and routing, but, eventually you get used to it.

However, there are at least four major drawbacks: 1. It is WAY unstable when it comes to tuning.

 It can start drifting anytime and gets annoying.

2. The cute little sequencer has no CLOCK INPUT (of

 any polarity), just CLOCK OUT....

3. The RESONANCE cannot be set to self-oscillate. 4. The CV INPUT & OUTPUT is of the linear Hz/Volt

 variety, which means I haven't yet   figured a way
 to drive the thing off my MC-8 (or MPU-101 if you
 happen to have one...) 

Peace. – Evan Makris Boston, Mass.

Subject: Yamaha CS40M problem

I have recently aquired a used Yamaha CS-40M. It's one of the few monosynths with memories (20 in all) and a versatile beast with rich analogue sounds.

But there's one major problem: once I have stored the pathces into memory they can't be re-edited when recalled. The knobs work only in panel mode. Now I wonder if this is done on purpose or is this some malfunction or misuse by my side. I have no user's or service manual so I can't check this out. And if Yamaha haven't provided this synth with edit capability could it be modified to have it?

My experiences of other CS-series synths are following:

CS5/CS15 - Good basic monosynths with singing sound, CS5 perhaps too limited. CS50/60/80 - Great analogue polys, especially CS80. Plenty of knobs and thick

           sound. Heavy and generate a lot of heat so there are tuning
           stability problems and component aging when older.

CS70M - Electronically quite different, more reliable. Good versatile

      polysynth although it has no useable link to the world outside: no
      MIDI and the sounds can be stored only in magnetic cards. No touch

Janne Lappalainen Turku, Finland

Subject: CS30 Architecture

It is a monosynth with basically two independent VCO→VCF→VCA lines. But there are lots of opportunities for mixing between the lines. There is a lot of semi-patching ability using input switches. It has FM (no-envelope), ring modulation, noise input on one of the filters, and processing of external sounds from mic etc. The FM is a control on ocs 1, RM is a control on one of the VCA's. It has a sequencer, which is an 8-stage variable step-speed analogue, I would classify it more as a second, programmable-step LFO. It has three envelop generators (ADSR) with a lot of choice how you modulate what with which envelope generators. You have five virtual envelope generators, but two of these are inverted versions of the other ones. You can switch the input of the second filter to be the high-pass of the first filter, giving complex filtering abilities.

I have had some marvelous sounds out of the FM, but the oscillators (on my one anyway) do not track evenly. So, my wonderful sound is out of tune within a few keys. I plan to have a sampler as my major purchase this year, so maybe I should have the CS30 sent out to me (it's in New Zealand), as I can sample the sound. A cheaper alternative would be a Korg MS20. (I've seen them going really cheap, e.g. 80 pounds, anyone know what these machines are like? Having a patch-cord machine would be dead sexy). What about a Roland SH-2? One of the local shops claim they're going to be getting one soon.

Unfortunately the sound is IMHO a bit thin compared to Minimoogs, Monopolys etc. One of the VCA's has an additional input for the sine wave from VCO1, which can fatten the sound, but it still doesn't have "IT" (whatever "IT" is).


# Yamaha DX100

This was a 4-operator FM synth that appeared in about 1985. I suspect that it should have better been seen as a rack mount module as it is a tacky little thing with mini-keys, no velocity and a nasty, plastic case.

Like most 4-op FM synths, it has a lot of trouble with low-end response. You just can't produce any growly sounds down low. While many basses sound OK, you can never really produce a good, ambient droning sound. The presets are fair to middling. There is a good variety with the only really obvious thing missing being synth pads (although this is very difficult to do with 4-op sine only FM with no effects). Particularly excellent are the bells and one or two of the basses.

MIDI implementation is sparse but adequate. You can select omni on/off, send/ receive channel, channel info on/off (for patch changes), sys info on/off (for sys-ex dumps). Yes, there are bulk dump commands and this offers the chance to get hold of lots of DX21 patches that are available in the public domain.

The synthesis engine is 4-operator, sine-only FM and the sound is all that that implies. You have a choice of 8 algorithms (operator/carrier combinations. If you don't know what I'm talking about then you should find a good book on FM synthesis - there are hundreds available) with feedback on operator 4 only. Each operator has its own EG but, for some bizarre reason (perhaps price), Yamaha chose to have only one LFO per voice which can be routed to operator output level, voice pitch and that's about all. I think that this is also a problem with the DX7. I'd really like to see one LFO per operator, one for pitch modulation and one for overall amplitude modulation, even if this meant hardwiring them. The other problem is that there is no fixed frequency option on operators. This causes problems when it comes to programming subtle pads (which can be done on DX7s and even TX81Zs) and when I discovered the feature on other machines I was in seventh heaven! You can set the rate and level scaling on each individual operator which is a real boon when it comes down to programming acoustic instrument emulations. Controller programming is excellent with velocity response (over MIDI) for each individual operator being controllable. Breath controller and hold/portamento pedals are also supported.

It is an 8 voice polyphonic, monotimbral synth. On the back panel, there are MIDI sockets (IN/OUT/THRU), Output and Headphone jacks, footswitch jack, breath controller input and adapter input (9V). The DX100 was designed as a strap on unit and hence there is the option of powering the unit using batteries.

I like my DX100 a lot. It has an important place in my synth setup and I would never part with it, despite my frustrations with its synthesis engine. It does some superb digital sounds (the hard-edged digital sound that was so prominent in the mid 1980s) and a few passable acoustic instrument emulations (the Sitar's an absolute killer and I programmed a reasonable pan-pipes patch). If you can get one cheaply then it's most worthwhile.

Jonathan Elliott | Why is an 01/W like a piece of string? | Every yo-yo wants one.


Well, I finally got my SY77 and had a chance to play around with it. For those of you DX users who are wondering what's the scoop with the SY77 here is a mini-review of the instrument:

Voice architecture:

 Up to 16 FM sounds and 16 digital samples may be played
 simultaneously.  A "voice" may consist 1, 2, or 4 "elements"
 offering several combinations of FM and sampled sounds.
 Each element in a voice can be zoned by both note range and
 velocity range.  Each element has a wide selection of stereo
 panning capabilities.

FM unit:

 It uses 6 operators like the DX7.  However, some interesting
 enhancements have been added.  There are 48 algorithms instead
 of 32.  Unlike the DX7, an algorithm is extremely flexible.
 The programmer may specify up to three feedback loops from
 any point to any point in the algorithm.  In fact, a
 feedback loop consists of source and any number of destinations
 so there are really more than 3 feedback loops available.
 Each operator in an algorithm will accept up to 2 modulators
 which can be any of the following:
    a) another operator (determined by the algorithm or feedback)
    b) a digital sample
    c) noise
 Also, each operator may have any of 16 different waveshapes.  This
 let's you do things with 1 or 2 ops that used to take 3 or more ops.
 There are 2 LFO's instead of 1.  Pitch sensitivity is independant
 for each operator (the DX7 had global pitch sensitivity).
 Envelope generators have more segments including a delay segment,
 2 release segments, and a programmable loop point.  Key velocity
 can modify the attack time.  The pitch EG may be disabled for
 individual operators.
 Several parameters have negative ranges where the DX7 only had
 positive ranges allowing for cross-fade effects.

Sample playback unit:

 At first glance, the ROM samples did not impress me very much.  They
 were good but not great.  They were clean but not "crisp".  However,
 once I reminded myself that the SY77 is not intended to be a sample 
 player I appreciated the samples a lot more.  These samples are intended 
 to be combined with FM sounds - not merely layered on top of them but 
 fully integrated into FM synthesis process.
 There are the usual multi-sampled instruments such as piano, strings, 
 horns, choir, etc.  Additionally, there is a rather interesting palette
 of strange samples that are not musically interesting by themselves but 
 do wonders when combined with the FM sounds.  (For you LA fans there
 are the obligatory "airy" sounds).  There is the usual compliment of
 drum sounds.


 Each of the 4 elements in a voice has its own pair of filters.
 (giving a total of 8 filters in a single voice).
 The filters have a 12db/oct slope and may be set for lowpass or
 highpass.  The pair of filters can be combined as a bandpass
 (low+high) or as a 24db/oct lowpass.  Filter resonance
 may also be set all the way to self oscillation.  An obscure feature
 (buried in the manual) allows one element of a voice to "borrow" the 
 filters from another element giving up to 4 filters on a single element.
 (48db/oct anyone?)
 Now...I grew up with MOOG filters so I was going to be a tough
 critic here.  I was surprised by how good these filters sounded.
 They're not quite up to the MOOG filters but they are as good,
 if not better, than some other analog filters I've heard.
 They are clearly more than just a marketing gimmick.  The
 resonance adds that certain quality I've always enjoyed from
 a good analog filter.  Unfortunately, none of the factory sounds
 on the SY77 really show what the filters can do.  Naturally,
 there's a ton of controller routings to play with the filters
 including independent EG's for each filter pair.

Signal processing:

 Well, the SY77 instantly made me a believer in on-board effects.
 Each voice has its own signal processing parameters.  The
 signal path can go through 2 "modulation" units and 2 "reverb"
 units.  These units can be combined in various configurations.
 There are a handful of modulation options such as chorus and flange
 (all stereo).  There are dozens of reverb options including distortion
 (also in stereo).
 I'm use to using SPX-90's and the SY77 signal processing did not
 strike me as any better or worse than an SPX-90 (although perhaps
 quieter since the A/D and D/A stages are missing).

User interface:

 Considering how complex this instrument is, the user interface is
 not bad at all.  A large, backlit, LCD can show entire menus,
 EG and scaling curves.  There are usually several different ways
 to input data at any given time so you can reach for whatever
 input device that feels the most comfortable.  However, it's still a
 bit overwhelming and discourages casual experimentation.   A full screen
 editor would probably be a good investment.


 16 tracks, 1 song, 16000 notes.   No comment.  I don't use it.


 The MIDI implementation is certainly flexible but nothing like
 a Kurzweil PX1000.  The SY77 can be put into "Multi" mode.
 This will provide 16-channel multi-timbral tone generation
 with dynamic voice allocation.  However, there is a major
 disappointment here.  It appears that there is only one real
 signal processor in the instrument.  In multi mode, the same
 signal processing is used for all voices on all channels (overiding
 the signal processing parameters for each individual voice).
 So, if you have a killer guitar sound that relies heavily
 on the signal processing, it will lose something in multi mode.


 There are 128 preset voices + 64 internal user voices + 64 ram
 card voices.  There is also a slot for additional wave data cards.
 A 720k floppy serves the usual purposes (MS-DOS format).  One strange
 quirk - voices that consist of 4 elements can only be stored in
 certain memory banks because they take up more room than voices
 with 1 or 2 elements.  (That is the sort of design trade-off that
 a user shouldn't have to worry about IMHO).


  The SY77 has a major hurdle to overcome.  It is an enormously
  powerful and complex instrument being sold in a market where
  the "quick fix" is king.  I must also say that the SY77 is not
  instantly endearing.  This is mainly because the factory sounds
  on the instrument are generally mediocre - rehashes of earlier DX7 
  sounds with a some sample playback added.  However, some of the
  presets are truly astounding giving us a hint of what it could do 
  in the right hands.  I can't say if the SY77 will change the minds 
  of die-hard FM haters but it puts forward a good show.  I suspect 
  that some of the third-party patch libraries that may be 
  forthcoming will make the SY77 a "must have" for many musicians.

– Scott Amspoker!bbx!bbxsda!scott

Subject: Yamaha SY77 (LONG)

Gee, my local dealer asked for $1200 for SY55, even though I didn't


anything yet. Considering that they asked so much for SY55, the price they
quoted for SY77 may be absurdly high. Can somebody tell me how much is
the reasonable price for SY77, and is it worth the price ? Thanks in


I just today noticed earlier postings re: SY22 & SY55. Well, here's a bit about the SY77, because I just bought one from the source in Tokyo this summer…

Credit Cards, anyone? I paid 250,000 for mine. Ok, OK, 250 thousand YEN or $1,840 and about $125 more for the industrial-strength 'flight case' (there's also a 'hard case' but it's not as sturdy). I bought from friends in Tokyo who own a duty-free shop and sold the 77 at cost, duty-free. I had to special order a U.S. spec model, though (took 1 week). Was it worth it? Definitely.

Current retail in Japan is 275,000 yen for synth, 45,000 for flight case– about $2,330 for both.

-BUT- and here's the bad news, reputable U.S. MIDI-shops are retailing the '77 for about $2,995, flight case for $400. This price is 'reasonable'- if you're not planning a trip to Tokyo soon. Is it worth it? Well… If you think you want one but haven't REALLY STUDIED it, or wanna know what your friend just blew 3400 bucks on, read on.

It's worth it if you really NEED an SY77. It's a unique and very capable machine, everything highest quality, down to the paper used in the documentation (which is superabundant). It's a refined musical instrument, not just a rack synth. It is well finished, feels solid, and works as it should. 1st rate stuff.

ARE YOU READY FOR THIS? Basic SPECS: It has 61 keys, C-C, with aftertouch, velocity sensitivity, polyphonic capability that can exceed your fingers, toes, and nose combined. It has 128 preset sounds (incl. 2 61-drum drumsets and DAT-quality samples), a good-size wave table, (w.t. & presets = 4 megabytes of ROM data) 45 algorithms (15 more than DX-7), built in memory for 64 user-built voices and 16 user-made 16-voice "multi's" (more on multi's later), a 3.5" 800k floppy drive, 2 64k card slots (1 wave-form and 1 wave data) 16-track (15 song, 1 99-pattern rhythm track) sequencer with switchable quantization up to 1/96 of 16th note (real-time, punch-in, note-by-note entry; full song editing: after- the-fact note/velocity/pitch-bend/etc correction of course; ea. song tr. 15,000 note capacity; rhythm patterns up to 32 measures ea., songtrack& pattern copy/ paste/chaining of course). MIDI IN,THRU,OUT, 6 breath/pedal/footswitch-control assignable inputs, a pitch-bend wheel, and 2 modulation wheels (assignable, 1 min-to-max, 1 center-flat). It weighs as much as a real piano (well, almost! 35 pounds, 45 in case) These specs are NOT exhaustive. There IS more.

On Synthesizing Power- Yamaha's "Realtime Convolution & Modulation" The SY77 can serve as 16 independent synthesizers simultaneously when using the sequencer- i.e., a 16-track piece can use 16 (one track's a rhythm slave) different voices, each with its own envelope, LFM, pitchbends, digital variable LP/HP/Bandpass filters, modulation/reverb effects (the 77 has 40 reverb effects- i.e., ping-pong, bathroom, church, reverse gate feedback etc- and 4 modulation effects- symph.,flange, etc), stereo pan (the 77 can do all sorts of L-C-R effects incl. crossovers, delays, variable-speed effects), static pan, portamento, detuning, pitch-randomness, ETC– ALL simultaneously (!!)

On "Elements" (parts of a voice) What's more, each voice can consist of up to 4 'elements,' -basically complete sound forms w/ independent envelope/effects/ETC- which come in 2 types. AWM ("Advanced Wave Memory") elements are from the SY77's built-in 48kHz (DAT quality, vs. 44.1 kHz CD quality) digital sampled wave table- pianos, brass, strings, sound effects, etc. Or you can use your Cards/Disks as wave-table source data. These can be edited completely with filters, LFO, etc. The other type of element, AFM II ("Advanced Frequency Modulation II"), are synthesized with the 77's 45-algorithm synthesizer by– You. Of course, AFM elements also have their own envelopes, volume, static pan, LFO, etc too. Yes, you can save/load them to/from disk/card.

On Voices Where the SY77 gets interesting is in its ability to mix the different (wave and synthesized) elements to create unique voices, then change the parameters for each of the 1,2, or 4 AFM/AWM elements in a voice. Thus, you can make an 'Electric Piano' with a bass that kicks in with high-velocity notes, a sax riff with slow crescendo on notes continuing past, say, 3 seconds, and a warm synthesized rumble that is heard only on soft notes, or only on notes higher than, say, C3, or only on notes lower than, say, B3 and higher than D5. (Sounds like a weird voice, but you get the idea). Of course, you could then make the sax riff swing

from left to right to left very slowly, make the bass sit at 10 o'clock

left, center the electric piano, and make the bass jump from far right to mid left. You could filter out subsonic rumble from the synth element's lower registers, cut the highs from the sax riff on high notes. You could give the bass element a little vibrato. Then you could put the whole thing in an acoustic 'bathroom,' give it early echoes, or feedback distortion with a delay. And so on.

Wow! Music with one note Fans, rejoice! But there's MORE.

On 'Multi's' Once you've made your incredible one-note symphony, you can make 15 more totally different voices and stick them in a 'Multi.' A Multi is a set of 16 voices (any of the 16 can be changed/edited at any time via EDIT function) for recording a 16-track piece. Before you can record 16 track pieces, you must define which 16 voices will compose the multi (like getting your band together, guys, you can't record without your musicians!). Or you can use the SONG mode for recording one-voice solos. Of course, you can record anything (MULTI or SONG) in any voice(s) and then change any voice(s) for playback at anytime (great for quick ideas on whether a flute or fat strings or raunchy guitar suit a piece/track best.

That's all I've got time to divulge. If you want to know more, write me.. or buy an SY77. –Konrad O. Solomon Mr. Net-Policeman says he's: "Wasting too many people's time with stuff they don't need to know but think they want to!" Perry says: "If I screwed up here… Boo me, don't sue me!" (OOH, that's BAD)

ks3o ("three-oh")

Subject: Yamaha SY77 Update

Hi there!

Just wanted to end some common misconceptions about SY77 and clarify any incorrect implications I left in my posting Yamaha SY77 (LONG), as well as respond in general to recent discussion of the machine. I even had an idea or two!

A) The SY77 is NOT a MIDI-sample-managing keyboard. You CANNOT load samples into it via MIDI. It is NOT a sampler a la Ensoniq EPS or Casio FZ-10M. I believe Yamaha was banking on the idea that most potential SY77 owners would ALREADY possess one or more other keyboards or boxes which perform such MIDI-sample- tasks.

B) The SY77's integral wave-table is NOT editable. It's ROM. You cannot, as I seem to have implied, addend to this wave-table data from disk or card (or MIDI). Sorry! You CAN, however, save any voices or multi's you've synth'd on SY77 (and their constituent elements) to disk or card or internal RAM (from which you can then retrieve) or transmit them via MIDI-OUT.

C) You also CAN obtain card/disk sets which provide SY77 voices- for example the Yamaha 2-card Saxophone set and 2-card Percussion set or 1-card Japan Synth Programmers' Assoc. voice-sets and multi-disk Sound Sources Unlimited voice collections (reviewed in Keyboard).

Now: Some of the voices used by SY77 (such as the ones in the 'Percussion' cardset from Yamaha) are called 'Drum Set' voices. I FORGOT TO MENTION THESE IN MY EARLIER POST! Each "Drum" type voice consists of 61 individual AWM-type wave-forms, ONE PER KEY. Think about it! These are not MIDI-portable samples you load into the SY77, but they could serve virtually the same purpose if you had the voice data. Someone (Yamaha? Sound Source? 3rd party?) just needs to produce "Drum Sets" more like traditional MIDI (multi-)samples: a "Drum" type multi- sampled 61-key piano voice, for instance, is a very real possibility! Or, IF Yamaha sold a sampler box which samples for "Drum" type voices, voila! You say: "That's stupid. MIDI is better."

Maybe. Because on SY77 you can switch between AWM/AFM (1,2, or 4 element) voices and "Drum" voices with no effort and blinding speed. Think about it! No more time-consuming, patch-juggling MIDI-dumps for frequent multi-sample-switching. Fewer keyboards, boxes, and cables in live performance; less junk to haul on the road! This is Genius!! I hope Yamaha is listening. AND on SY77 you have the power to move a "Drum" set's 61 AWM's around the keyboard freely, edit their static pan, volume, pitch, etc.,individually- all with a turn of the 77's data-wheel or a few presses on its 10-key pad. All that's missing is more "Drum" type voice-sets (I don't want to call them 'samples'), or that imaginary box! It would evolutionize sample-tasking!!

  Got that, Yamaha?
  Am I seeing things, guys? (flame me at

D) If you can get an SY77 for $2000-$2500 (N.R.M.Synth readers, is this only true in New Mexico?) vs. the $2995 I was quoted on the phone by the Authorized Yamaha Dealer in my home town (Honolulu, Hawaii), more power to you! Of course, the dollar is way down from this summer, (I got 154 yen to the dollar this summer when I bought my 77 in Tokyo, now it's like 137 to the dollar), so prices may be rising- on speculation.

E) Used SY77's? Used keyboards abound! There are as many used keyboards as there are dealers who put them on display (they may call them demos but they're used all the same and it's to your advantage if you can make them admit it). Used may be the way to go for equipment you like but can't afford new (if it's in good shape). I won't mention names, but

from recent discussions on this board and elsewhere, Yamaha's have

nowhere near the breadth and depth of problems to be found in other equipmnt. The number of 1st generation DX7's still in use speaks for itself. Other Yamaha users, how is your stuff holding up (SY77 users?) I have had zero problems with SY77 to date, except it weighs a lot and attracts too many curious finger(print)s wherever I take it!

That's all, folks. Konrad Comments, Questions, and Better Informed Opinions welcome… via e-mail

Subject: Re: SY77

I've worked with the SY77 for about a year, and am quite impressed overall. It will in fact take much longer to fully fathom the voice programming possibiliti es. A good way to get an overview is to check out the reviews in both Electronic Musician (they loved it) and Keyboard (Jim Aikin is hard to please), and also Howard Massey's owners guide sent by Yamaha or available from Mix Bookshelf (as well as his recent article in Keyboard).

Contrary to what Jim Aikin implies, this is much more than two synthesis techniques side-by-side. The FM component is truly a new generation, which addresses many of the shortcomings of the DX series, as well as adding some nice features. I especially like the operator looping feature, which can be used to create some real nice evolving and modulating voices. The pitch EG can be individually assigned to operators, which also leads to great extended, evolving textures. The sample waveform section - AWM - is somewhat less impressive. The samplesI find to be of mixed quality, but the filters (with resonance) are wonderful.


Can someone summarize the technical aspects of Yamaha's RCM
synthesis? I saw an SY99 demo at the keyboard show in LA over the
weekend and was pretty impressed. The sales people were pushing this
"amazing RCM technology" which generates more "expressive" sounds. When
I asked what type of expressions they were talking about they said,
"Well, you can EXPRESS yourself better musically."
Well, at least the demos were good. So anyway, can someone explain
to me what RCM is? And what exactly is the "vector synthesis" they use
for the TG33 and SY22? What does the TG55 use?


I'll try and help you with a description of Yamaha RCM.

I was also at the keyboard show and saw the Yamaha SY99 demo. As an owner of an SY77, I couldn't believe the glowing description of RCM by the Yamaha demo guy.

RCM (Realtime Convolution Modulation) allows the output of an AWM (sample) element to be used as a modulation input to any of the 6 operators of its paired AFM element. A 2-AWM and 2-AFM voice allows 2 such pairings. However it does not allow an AWM voice (or sample waves) to be used as a carrier source for an AFM operator (something I'd still like to hear). The AWM/AFM pairing allows the original AWM element, plus its RCM'ed version to be mixed together to form a voice.

For as much hype as they gave RCM at the show, its interesting to note that very few of the SY77 voices I have played (Yamaha or other) use RCM. In my 4 months with the SY77, I have had little luck in producing musical sounds using RCM. When an AWM element of any timbral complexity (piano for example) modulates an FM carrier, the FM output becomes noisy very quickly. Its main utility seems to be in adding roughness (for attack transients) or noisiness (for wind sounds) to an AWM sound. The basic truth is that most AWM sounds make lousy FM modulators. – David H. Miller Retix Internetworking Group

Subject: Mini-review of the SY99

I was able to attend a demo of the SY99 about two weeks ago and, for what it's worth here are my impressions of the beast.

 Looking at it from the front, it is exactly the same as the SY77 except
 that the keyboard has more keys. Every wheel and button that was there
 on the SY77 is there on the SY99 and there are no extras, although some
 have more functions.

Comparing the glossy brochure of the SY99 against the SY77 we find: 1. The tone generator stages are the same. 2. The keyboard has 76 velocity sensitive keys (SY77 was 61) with channel

 aftertouch which on the SY99 can also be zoned. For example, the
 aftertouch can be set such that it is only effective on the highest note
 being played so that you can 'trill' only the melody line.

3. 2 DSP effect units (as on the SY77) but with 63 effect types. 4. The sequencer has changed a bit. It still has 16 tracks but can hold 10

 songs (instead of the SY77's one). The sequencer memory has also been
 expanded to hold about 27000 notes. It still has a resolution of 96ppq
 internally and 24 ppq for MIDI sync.

5. Memory: Same 128 preset voices and 16 multis plus another 64 internal

 (i.e. editable) voices and 16 multis. The waveform memory size has been
 doubled to 8 Mb and has 267 sound samples in it (SY77 has 112 sounds).
 It has the same card and waveform slots plus the 3.5" floppy.
 But it also has an additional 512 Kb of battery backed up memory
 (expandable via 5 rear slots to a total of 3 Mb). Up to 512 Kb of the
 total possible 3Mb can be assigned by the user to Midi Data Recording
 and the remainder is used to store user samples.

6. Controllers: no changes

There are of course some interesting additions compared to the SY77: a. The SY99 can load Standard Midi Files (formats 0 and 1) from the disk

 and the sequencer data can now be written out to the disk in Standard
 Midi File format 0.

b. The SY99 can read samples into the new expanded memory but ONLY thru the

 midi port. During the demo, this was done 'live' in front of our eyes,
 where the demonstrator recorded a 2 or 3 second riff played on the guitar
 by a member of the audience. He then edited it slightly and then played
 the sample back into the SY99. The sample could then be played by pressing
 any key on the keyboard and the SY99 adjusted the sample speed so that,
 for example, playing a low note made the original sound as though it had
 been played on a bass guitar rather than a lead.
 However, transferring a few seconds of a sample via MIDI is SLOW and a
 SCSI port would have been nicer.
 BUT note that the SY99 is NOT a sampler. In order to create your own
 samples, you have to own/buy/rent/borrow/steal a sampler.

e. It recognizes loop points in the sample although this was not


d. Samples (from the expanded memory only) can be stored on disk. e. It can function as a Midi Data recorder by storing a bulk dump from

 another device into the expanded memory and from there it can be written
 to disk. You can also send bulk dumps to other MIDI devices.

f. The Jump/Mark function has been expanded so that it can remember up to

 five previously used pages instead of allowing only a single marked page
 as on the SY77.

g. A lot of the information pages have been restructured/reordered to

 provide a somewhat more consistent and easily accessible user interface.

There are probably some interesting features in there that weren't demonstrated. For example, very little was said about all the MIDI features that have been added and there was no demo of any of the new DSP effects that have been added.

As you might expect, it sounds exactly like the SY77 except that, with the ability to play arbitrary samples, some of the demos had voices and other interesting non-standard sounds (compared to the SY77). Many of the built-in samples that existed in the SY77 have been redone and are apparently much better than the originals. And then, of course, there's all the extra preset samples too.

For the current owner of an SY77, what about the upgrade path? If you've got the bucks, presumably you can swing a trade-in deal with your local purveyor of music computers. BUT what happens to all those disks you've got? Apparently, no problem. Although all the preset voices and wave samples are different on the SY99, it will map your SY77 voices into their SY99 equivalents.

Of the "complaints" about the SY77 that I have seen on the net, there are a couple that remain in the SY99: 1. The sequencer still records just one track at a time. 2. The sequencer memory is not battery backed-up so a power failure, or

 switching off the SY99 at the wrong time, can lose you a lot of blood and
 sweat that went into the performance-of-the-century that you'd almost
 finished. It would be nice even if that memory were protected by a large
 capacitor that could hold the memory for a few minutes. Long enough to
 survive most power glitches and easily long enough for you to switch it
 back on again if you lost your mind and accidently turned it off.


Subject: SY99 reviewed by an SY77 owner

A couple of months ago I posted a brief "first impressions" article about the Yamaha SY99 keyboard. Well, after some soul searching I decided to trade in my SY77 for an SY99. Now that I have it in *my* studio environment, here is a more detailed review of the instrument. Although my comments will the most meaningful to existing SY77 owners, anyone considering a purchase of an SY99 might find this article useful.


A couple of years ago, Yamaha released the SY77, the first major improvement of the DX family synths. Not only did they substantially enhance the FM capabilities bringing it into the realm of thick, analog-style sounds (with filtering); they also added ROM samples. As I pointed out in my SY77 review at the time, the quality of the samples served only to remind me that the SY77 was primarily a powerful FM synthesizer. In my opinion, the samples served in a supporting role, lending a helping hand from time to time. I never fully understood why Yamaha chose to spin off the samples to star in their own show, the SY55.


At the moment I have both a 77 and 99 setup in my studio in order to transfer patches. I have both instruments together on a keyboard stand playing them through a quality PA system and headphones. Considering how accustom my ears have become to the SY77, I was plesantly surprised by the side-by-side comparison.


The 99 offers many enhancements over the 77 - some subtle, some not so subtle. Most obvious is the longer keyboard (76 keys compared to the 61 keys of the SY77). Front panel controls are identical. Other major enhancements include an additional 2 megaword of ROM samples, a completely new effects processor, the ability to load user samples, and master controller capabilities. The factory patches are quite different.

Minor changes include more sequencer memory, up to 10 songs at a time, a midi bulk data librarian utility, and miscellaneous convenience fetures such as the ability to edit two filters together when using them as a single 24db/oct filter (an SY77 peeve of mine).


With the exception of better effects, I wasn't expecting much improvement in the fundamental sound of the instrument. (I originally demoed it in a noisy music store with tiny speakers.) At home, I compared the 99 to the 77 with identical patches with the effects turned off. At first I noticed that the FM engines sound exactly alike. However, I stumbled onto a favorite FM-only patch called "Spirits". It is a haunting, highly resonant sound which is unfortunately plagued with intermittent distortion and noise on the 77. The 99 played the patch flawlessly. Apparently, something in the FM engine has been cleaned up.

The samples were another story entirely. Most of the SY77 samples were noticeably cleaner on the SY99. The piano and strings were better. The choir had more presense and included additional samples at the female end of the keyboard (whereas the 77 stretched a sample over too many notes). Saxaphones sounded more open and natural. In most cases the SY99 samples were obviously from the same source as the SY77 but much crisper sounding. In other cases it seemed that entirely different sources were used. This was especially apparent in the drum kits which were *much* beefier on the 99 (and included a lot more drum sounds).

The additional 2 megaword of samples didn't add anything earth-shattering but many of them definately had practical uses and were welcome additions.

In their review, "Keyboard" magazine complained that a few of the samples had noticable loops. This is rather subtle and not noticable when layering or simply playing polyphonically. All in all I am quite pleased with the samples on the SY99 and can imagine a sample-only product (like the SY55) based on them.

Finally there is the effects unit. As I understand it, the effects are based on the SPX1000. They have much different architecture than the effects on the SY77 and are more versatile. Notable enhancements in the effects area are a rotating speaker simulator and an Aural Exciter (the real thing licensed from Aphex). The progammer may assign MIDI controllers to modify effects parameters - nice touch.


The ability to load user samples into the SY99 didn't register in my head as something I would care about. Samples may be loaded via MIDI sample dump or from a disk. However, since I'm one of the few people in the universe vvthat owns a Yamaha TX16W sampler, I decided to give it a whirl. I stuck one of my TX16W disks in the SY99 and loaded in some samples. I then created some user "waves" (a single "wave" is Yamaha-speak for a collection of samples mapped across a keyboard). This procedure was refreshingly easy to complete (especially compared to the TX16W's OS >from hell). My samples sounded exactly as I hoped they would sound and the loops were intact. Bravo! Also worth noting is that sample memory is not lost when the power is off - no need to keep reloading tons of sample data from disk.

User waves are *not* second-class citizens in the SY99. They have the same status as the ROM samples and can be put to any use (including modulators in the FM synthesis). By the time I was through with this little experiment I was actually quite excited about it and intend to explore this more in the future.


The SY77 was frequently bashed because its keyboard could only play on one MIDI channel at a time. Yamaha attempted to address this issue but only with lackluster results. The SY99 can be instantly placed into "master" mode. Up to 8 master configurations may be stored, each defining 4 zones with various parameters per zone. It's better than nothing I suppose but I'll hang on to my own master controller software, thank you.


Aside from the wimpy master controller feature (a moot point to me), there really isn't anything *serious* to complain about. There are some minor annoyances however. 4-element patches are still confined to bank D. That was plain dumb 2 years ago and even more so today. There is still a clumsy multi-mode implementation. The on-board sequencer could still integrate more smoothly with the multi-mode capabilities. Also, why is it that all the RAM on the instrument is preserved when the power is off except the sequencer RAM? This no longer makes sense with all that user sample RAM.

While I'm on the subject, I think that the default 512K user sample memory is rather stingy. It can be upgraded to 3 meg with Yamaha's usual overpriced memory upgrades.

There are also some incompatibilities with the SY77 discussed below.


On the surface this seems like a trivial task. The SY99 will load SY77 files from disk or read an SY77 formatted data card. (NOTE: You cannot access an SY77 data card as an on-line patch bank. You must load or save the entire card to internal memory or disk. The card may then be reformatted for the SY99.) Keeping in mind that the ROM samples have been cleaned up, SY77 patches should sound the same on the SY99.

The problem arises with the different effects units. I found that some patches sound just fine without any further intervention. Some patches have the effects mix too wet - a simple adjustment is all that is necessary to fix that. Finally, some patches loaded with the effects way out in left field. I had to go in and reprogram the effects for those patches. I noticed that the SY99 at least partially translated the effects for those patches but didn't do a complete job. For example, the chorus mod frequency would be set correctly but the EQ parameters would be ignored.

It took me about 2 hours to correct a bank of 64 patches. To be fair, once I got the hang of it I noticed a consistent pattern in how the effects conversion failed and I was doing it much faster near the end. In some cases I actually preferred the "incorrect" effects and left them alone. I'll convert my remaining patch banks without too much trouble.

Finally, one small change in the SY99 caused me one hell of a major headache. The SY77 used MIDI program change messages to switch between patch banks and modes. The SY99 uses the newer MIDI bank select message to do those things. This isn't so bad except that Yamaha chose not to carry over the SY77's method of bank switching. My MIDI controller software simply wasn't prepared to deal with this and it cost me an evening to make the necessary modifications.


I had to ask myself if I was spending too much money on what was essentially an upgrade to an instrument I was already reasonably content with. Once I got the two synths together I decided it was worth it (at least for me). There's no doubt that many existing SY77 owners may opt not to move up and that's understandable. However, I've always considered the SY77 to be arguably the most powerful synth in its class though not necessarily the best sounding. It was a solid workhorse that rolled up its sleeves and did all the difficult tasks while the D70s and the Kurzweils got all the glory. So what was wrong with the SY77? Its mediocre samples brought it down a notch.

With the SY99, I believe that Yamaha is targeting the difficult-to-please sound purists. Although its sample playback engine is not going to surpass E-mu's incredible G-chip, it has a very pleasing quality to it and I find myself taking it much more seriously than with the SY77.

It seems that for the first time since the introduction of the DX7, Yamaha has finally "arrived" someplace although I'm not entirely sure where. I can't imagine what the successor to the SY99 would be like except to have more of everything (more polyphony, more memory, more samples, etc.). Time will tell.

– Scott Amspoker | |


While this is accurate, it's a little incomplete.

The TX81Z is a 4-operator FM synth, but, unlike the older DX/TX synths, the operators can be more complex than sine waves, making it possible to make some more complex sounds. The sounds are still "thinner" than the 6-operator synths, but it's still a great-sounding machine.

It's true that the TX81Z doesn't make very realistic sounding drums and pianos, but the ones it does make are quite interesting. As with any synthesizer, it depends on what types of sounds you are after.

The performance mode is a bit more complex than just being able to play 8 notes on 8 MIDI channels. Basically, there are 8 'slots', each of which has a patch, MIDI channel, number of voices, output selection, volume, modulation, note range, transposition, and a couple of other parameters.

The effects unit is not strictly an effects unit. What it does is to use a pair of voices to simulate a single voice with a delay. It can be useful, but I find it easier to do the same thing with Megalomania or MAX, and I have more control over the process.

Overall, the TX81Z is considered to have very good price/performance. In the last two usenet synth polls, the TX81Z was by far the most popular unit (even more so when combined with the FB-01 and DX-11). I've owned one since 1989, and I owned two at one time (I replaced one with a K3m – sorry Ben).

The FB-01 is essentially a TX81Z without front-panel editing, with a different set of patches, and a few other features missing. – …David Elliott … | …!{uunet,mips}!sonyusa!dce

/data/webs/external/dokuwiki/data/pages/archive/music/synth-re.vie.txt · Last modified: 2000/08/13 06:41 by

Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki