Sounds of the '80s Part 2: The Yamaha DX1 & Its Successors (Retro)
Reviews : Keyboard
 |
In the first part of this history of FM synthesis, last issue, I mentioned that 1982 saw Yamaha demonstrating a prototype keyboard with "six-operator equation generators" that you could edit. Appearing a year before the launch of the first DX7 (covered last month), the prototype I mention was nonetheless a true 'DX', with all the functions and attributes we now associate with this famous family of synthesizers.
I know of only one mention of it to appear in print. In the International Musician and Recording World Electronic Musical Instrument Guide 1983/4, there's a picture of an otherwise unknown synth. It featured nine programming sliders and an intriguing flip-up panel to the right of its enormous control surface but, in all other respects, this huge, six–octave keyboard looked identical to the instrument it would eventually become. Despite sporting Yamaha's famous, analogue 'CS' series prefix, it would evolve into the biggest and best DX of them all. The Yamaha CSDX was destined to become the Yamaha DX1.
The DX1
I remember the first time I saw a DX1. (You just knew I was going to say that, didn't you?) It was March 1984, and I was walking past a music store in a neon-infested street in Osaka, Japan. I even remember thinking that, if Yamaha's GX1 analogue megasynth (retro–reviewed in the February 2000 SOS) looked li
ke
the helm from the original starship Enterprise, the DX1 looked like...
well, a more modern version of the helm from the original starship
Enterprise. (Remember, this was five years before Jean-Luc Picard first
pointed his index finger at the camera and said "Engage".) Anyway, this
DX1 was large, flat, beautiful, and had enough red LEDs to compete with
the street outside. It also had a price to match -- ¥3,500,000, which
was somewhat over £10,000 in 1984 (somewhere between £30,000 and £40,000
at today's values).Everything about the DX1 spoke of quality: its physical size, its weight, its 73-note wooden keyboard, and, of course, its price. And, although in theory a DX1 should have sounded no better than two DX7s costing a mere £3000, most players privileged enough to play one agreed that this wasn't the case -- it sounded better. You might be forgiven for suspecting that the reasons for this were as much psychological as sonic, but the DX1 genuinely did sound better because Yamaha hand-picked the components for its flagship instrument, leading to a noticeably quieter and cleaner-sounding output. But even this was to miss the point of the DX1. Its ostentatious control surface was more than mere decoration; it was the breakthrough that finally made FM accessible.
Since we're here to praise Caesar rather than bury him, let's take a good look at what made the DX1 special. We'll start at the top left of the panel. It's here that, under a long perspex window, you'll find the graphic representations of the 32 algorithms that the DX1 shares with its little brother, the DX7. To the immediate right of this, there are no fewer than 13 LED windows that -- depending upon the algorithm chos
|
But this is far from the end of the story. Moving to the right again, the next section has further LED windows that show the frequency ratio (or, if selected, the fixed frequency) and the detune amount of the currently chosen operator. Then there's the biggie... the collection of eight numeric LEDs (00 to 99) and four 16-segment LEDs that depict the amplitude rates and levels (or, if selected, the pitch-change rates and levels) of the currently selected operator. If you've ever torn your hair out when trying to puzzle out a sound on a DX7, you'll appreciate just how marvellous this on-board representation of the envelopes can be.
Next, there's the Keyboard Scaling panel, which offers more LED windows showing the break point, depths, and selected curves for the chosen operator. Finally, yet more LED windows display the velocity sensitivity, amplitude modulation, and output level for that operator.
As you can imagine, all of this takes up a significant chunk of real estate, but there's no question that it's space well used. Indeed, the only way to make it more useful would be to display all six operators simultaneously. Unfortunately, this would make the DX1 about five feet deep!
You control all of the pretty lights using a control panel that requires more than double the space occupied by the displays themselves. This features yet another innovation that comes as a very welcome surprise: real buttons. Whereas the DX7 had horrid membrane switches (sure, they kept the price dow
|
Of course, much of what I've outlined so far describes merely a sexier DX7, and it's not until you search for the patch–selector buttons that you discover the next, huge difference between the DX7 and the DX1. Where the DX7's 32 membrane switches doubled as patch selectors, the DX1 offered two sections -- 'A' and 'B' -- each with four banks of eight patches, for a total of 64 patches in all. Oh yes, and don't forget the dual ROM/RAM cartridge slots; one each for the 'A' and 'B' banks, thus making 128 patches available at any given time. Not bad for 1984!
And we're still not finished, because alongside the 24 patch-selector buttons lie a further 16 multi-function buttons and a 40-character x 2-line LCD. If you are playing (as opposed to editing) the synth, these take you from 'Single' mode to 'Dual' mode, acting as eight banks of eight memories that access the 64 dual-patch performances. In Edit mode, eight of them act as LFO editors (it's quite an LFO section in a DX1!) while two others allow you to transpose and name patches. Then there's Function mode...
This has an associated Function Job Table silk–screened into the area occupied by the flip-up panel on the CSDX, which lists no fewer than 47 additional parameters, sub-divided into Tuning, Performance Memory Parameters, Memory Management and MIDI Control sections. When you see that, for example, nine of these parameters control the poly-aftertouch control of individual operators, it all looks hideously esoteric, but I can tell you from experience that -- once mastered -- it all works remarkably quickly and simply.
Returning to the right of the panel, there are four faders and two buttons where the nine faders of the CSDX once resided. These control the master volume, the A/B balance between the two sections, portamento time, and data entry (either by up/down buttons or continuous controller). All of this sits behind the Minimoog-style pitch and modulation wheels that, like everything else on the DX1, are beautifully weighted and positive in action. Oh yes, and speaking of weighting, how does a 73-note piano-weighted keyboard with polyphonic aftertouch grab you?
At the back of the DX1 you'll find 13 inputs and outputs. These include the now ubiquitous MIDI In/Out/Thru sockets, continuous controller inputs for volume and modulation amount, and momentary pedal inputs for sustain and portamento on/off. But it's the six audio outputs that interest me most. In truth, there are just three outputs because they are presented in both balanced (XLR) and unbalanced (quarter-inch jack) format. Nonetheless, they offer individual 'A' and 'B' outputs, as well as a single, mixed output. You might ask why I put so much emphasis on these. Well, if you're wrinkly or crumbly enough, think back to 1984 and ask yourself how many other instruments of the era offered two fully–functional 16-voice MIDI synthesizers in a single (albeit very heavy) case. The Prophet 10? Sure, it had two engines, but it was merely 10-voice polyphonic and was pre-MIDI.
Likewise, the eight-voice Roland Jupiter 8 and Yamaha's own CS80. It
wasn't until much later that the two-synths-in-a-box concept really
caught on, and only Roland -- with the dual-JX8P Super JX10 -- came
anywhere close to competing with Yamaha in this area.Now imagine super-clean versions of, for example, the classic FM pianos, layered with equally excellent brass and woodwind sounds, Clavinets, or the DX's trademark percussion patches. You could even create some remarkably lush ensemble performances by directing different string and brass sounds to each output, and treating them externally with the effects units of your choice. I know it's fashionable to believe that FM was incapable of producing warm analogue pads, but just try a DX1 hooked up to a couple of Dimension Ds, a Lexicon or -- better still -- an Eventide effects processor. I guarantee that you'll question your preconceptions.
Of course, the DX1 is not what you could call a convenient synthesizer and, given its size and weight, it was inevitable that most of them would end up living in studios of one sort or another.
Consequently, it came as no surprise when, a year later, Yamaha announced the DX5. With a superb 76-note semi-weighted keyboard, this offered channel aftertouch rather than poly–aftertouch, lacked all the LED screens, and provided fewer buttons, but functionally it was an otherwise perfect substitute for the DX1.
Indeed, many players preferred the DX5 to the DX1, because the DX1's keyboard is perhaps the heaviest you'll ever encounter on a synth. To quote my mate Blue Weaver (The Strawbs, The Bee Gees, and the chap who played for the Pet Shop Boys what was later to become my DX1), "You can shred your fingers playing glissandos".
In short, the DX5 was slim, manageable, and a delight to use. And, at around £3000, it cost just 30 percent of the ten grand demanded for its almost identical bigger brother. It's true that neither the DX1 nor the DX5 were the most powerful examples of their generation of FM synths, offering just a quarter of the eight-part multitimbrality and 128-note polyphony of the TX816. But that wasn't the point. The DX1 and the DX5 were superb performance synthesizers, and they remain so to this day.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
A Large And Unruly Family
Throughout the mid-'80s, Yamaha flooded the market with a plethora of low-cost FM synths. These used the truncated four-operator/eight-algorithm FM architecture of the DX9, so they were not compatible with the DX7, the DX1 or the DX5.
The DX27 was the basic model. With just 8-note polyphony and a 61-note keyboard that was neither velocity- nor pressure- sensitive, this was, in retrospect, a particularly poor instrument. Nevertheless, the combination of 192 preset patches, 24 user memories, aggressive pricing and the magic FM cachet ensured that the DX27 was a commercial success. There was even a DX27S, with built-in speakers and a chorus effect. This was no doubt aimed at the 'home' synthesist, but I suspect that Yamaha sold very few.
Then there was the DX100. This was in many ways identical to the DX27 but was fitted with a 49–note mini-keyboard and strap pegs. Designed for Howard Jones-style posing, this was perha
ps
the most successful of all the four-op machines. Certainly, it was
instrumental in introducing a whole generation of younger players to the
joys (or not) of FM synthesis. And let's not forget the FB01, an
eight-voice, eight-part multitimbral version of the DX27 and DX100 in a
half-rack box designed to complement the home organ of your choice. You
couldn't edit it from its front panel, but that didn't stop Yamaha from
selling bucketfuls of them.Slightly better than any of these, the DX21 featured a basic split/layer capability, thus making it capable of more interesting sounds than either the DX27 or the DX100. It also offered a handful of extra voicing capabilities, such as the ability to bend a single note within a chord -- great for acoustic guitar impersonations. If you're ever tempted to buy one of these cheapo four-op instruments, this is the one to get.
The Third Generation
With around 200,000 units sold, the DXs and TXs dominated the mid-'80s music scene in a way that no other keyboards or modules had done before, nor have done since. Indeed, FM dominated all forms of electronic music, and a huge family of home keyboards, keyboard expanders, electronic pianos, multi-keyboards and console organs used one or another form of FM sound generation. There were even FM synthesizers produced by other manufacturers -- specifically Korg and Elka -- to whom Yamaha had sub-licensed the technology. But, in 1987, Yamaha replaced the second generation of FM synthesizers with a third.
Yamaha sought to capitalise on earlier successes by calling all the new models the DX7 'something or other'. The DX7/IID was the basic model, and this was to be the direct replacement for the DX7. It retaine
d
the 'pure' six-op FM programming system of earlier DXs, but addressed
many of the problems of the originals and proved to be a huge
improvement over its predecessors. For example, the hissy output of the
DX7 was eliminated by the simple expedient of replacing its 14-bit D-A
converters with CD-quality 16-bit, 44.1kHz DACs. As you will appreciate,
this also increased the frequency response and dynamic range of the
synth, and eliminated much of the aliasing that plagued higher-pitched
notes on the original. There were many other notable improvements.
Yamaha improved the hardware and programming interface, added assignable
sliders, doubled the number of patch memories, introduced micro-tuning,
added a bigger screen, and significantly extended the storage
capabilities of the RAM cartridges. The new synth's FM voicing system
was also upgraded, with subtle (but important) enhancements to the level
scaling, LFOs, and pitch envelope generators. There was even a Random
pitch parameter that introduced subtle inconsistencies to the tuning of
each note as you played. Now known as 'analogue feel', this is almost
ubiquitous in 2001, but was a revelation in 1987. Perhaps the most
significant change was the introduction of a true stereo sound
generator, which made the DX7/IID bi-timbral, and capable of two-part
splits and layers (performances). Then there was the
aftertouch-controlled pitch bending, a hugely improved MIDI
specification, and more.All in all, the DX7/IID was an excellent synth, but its enhanced sibling, the DX7II/FD, was even better, because it added a 3.5-inch DSDD floppy drive to the standard specification. At a time when only a handful of expensive synths and samplers offered drives, this was a huge benefit that made backing up, saving, and loading
|
The baby of the range, the DX7S, lost the bi-timbral split/layer capabilities of its bigger brothers, as well as the improved LCD, but in all other respects it was a true DX7/II.
But perhaps the most attractive synthesizer of this generation wasn't a keyboard at all. The TX802 was a 2U rackmount module. Essentially an 8-part multitimbral DX7/II in a box, it was the canine's hindquarters and, at current prices, I reckon it remains a little gem, even in 2001.
Clearly, Yamaha intended for the DX7/IID, the DX7/IIFD, the DX7S and the TX802 to take over from the original DX7, and to continue its domination of the professional synthesizer market. But after four years the keyboard world was ready for something new. To be fair, the DX7/IID and FD were reasonably well received, but for some reason they simply didn't have the appeal of the original DX7. As for the 'S', it was a turkey of the first order. Yamaha's reign was well and truly over.
The Fourth Generation
By the time the tail end of 1987 rolled around, FM synthesis was beginning to look decidedly old hat. The lustre had gone, and when Roland released the D50, players everywhere snapped it up in preference to the Yamahas of the day. With its PCM capabilities, built-in effects and ethereal sounds, there seemed no doubt that the D50's was the sound of the future.
But Yamaha had a couple of throws of the FM dice left, and in 1987 it launched the TX81Z, an eight-voice, 1U rackmount module that offered a new type of FM synthesis -- one that allowed you to select from eight waveforms for each of its four operators. What's more, the TX81Z was cheap and multitimbral, so it soon proved very popular as a low-cost expander.
The following year, the DX11 appeared. This was the first truly multitimbral FM keyboard synthesizer but, because it was modelled closely on the TX81Z, it too was based on just four-op FM. However, the DX11 also offered copious memories and presets, and introduced a simplified editing system called Quick Edit FM that made FM more approachable than ever before, so it proved to be a reasonable success. The same was not true of its technological siblings, the home-keyboard style YS100 (which added an effects processor to the DX11 architecture), the YS200 (which added a basic sequencer), and the TQ5, the desktop module of the YS200. Yamaha's last 'pure' FM synth was the V50, again more of a home keyboard than a serious instrument, and with its demise in 1990, a line of great synthesizers finally died.
Conclusions
If this month's retrospective has concentrated on the DX1, there's a good reason. With the exception of the hyper-expensive rarity that is the FX1 (see 'Yamaha FX1' box on page 258) it was the king of the FM synthesizers. Even today, sitting behind 10 grand's worth of hand-finished keyboard is enough to m
|
Despite the fact that Yamaha have produced some superb instruments over the past 17 years, most synthesizer aficionados claim that FM was a blind alley, and that the CS80 was the last great Yamaha synth. I'm sure that John Chowning would think them wrong, as do I. Here's why.
The CSDX (which, if you remember the start of this retro, became the DX1) took the poly-aftertouch of the CS80, and incorporated it into a bigger, beautifully weighted keyboard. It took the dual voicing of the CS80, and adopted the same voicing philosophy for its 'performance' mode. It even took the real-time control of analogue synthesis and transformed this into a real-time digital editing system with multiple displays and relatively simple control capabilities. (That the DX1 lost the nine control faders of the prototype CSDX is a shame, but in my view it's irrelevant, because the DX1 remains to this day the only true FM synthesizer that you can programme quickly and easily from its own control panel.)
So if you're looking for a later Yamaha synth that still qualifies for the tag 'great', look no further: the DX1, a true successor to the CS80.
I would like to thank Richard Clews for alerting me to the photo of the CSDX in International Musician and Recording World magazine, and for his thought-provoking views on the history of FM synthesis.
No comments:
Post a Comment