BUCHLA 200e

 
Jim Aikin
 
 

Don Buchla has a legendary reputation as a modular synth pioneer, having started building instruments in the mid-1960s at about the same time as Bob Moog. In recent years he has concentrated on alternate MIDI controllers, but with the 200e series he returns to his roots, while also re-envisioning modular synthesis for the 21st century. My very first synthesizer was a modular, though it wasn’t a Buchla. So when I picked up a 200e at Buchla’s Berkeley studio and brought it home, my fingers were twitching. I couldn’t wait to see what sorts of sounds I could coax from it.

System Overview

Buchla has always had his own unique view of electronic instrument design. You could make a case that the System 200e is as much a work of art as it is a musical instrument. But then, so is a grand piano. And while the 200e may not require quite as many years of practice to master as a piano, it’s not quite plug-and-play. Some dedication and commitment will be required to get great music from it.

The front panel is colorful, thanks to the red, yellow, green, and blue LEDs, and is densely packed with knobs, jacks, and buttons. While you have to plug in at least a few patch cords to get a peep out of the 200e, it would be a mistake to think of it as a purely analog synth. As we’ll see, this is very much a modern analog/digital hybrid, not a throwback to those glorious days of yesteryear. The designs of many of the modules are based on Buchla’s 200 series, but all of them have been enhanced.

Control voltages are routed using banana plug cables, so several plugs can conveniently be stacked on a single output jack. Audio signals are routed using 1/8" mini-plug cables, which are shielded for low noise. As a result, the system makes an ironclad distinction between the two types of signals. You can’t run a control voltage through a filter or VCA, for example.

Modules are 7" high, and a panel chassis (which Buchla calls a “boat”) is a little over 25-1/2" wide. A typical three-boat system opens up as shown in the photo on page 72, but is hinged to fold neatly into a carryable shape, somewhat thicker and longer than a briefcase but with a handle. This is a brilliant bit of hardware design. The folded-up synth is light enough to carry one-handed, which is emphatically not the case with a lot of modular instruments this complex. The lump-in-the-middle external 12V power supply runs hot, but I had no problems with it.

Even the names of Buchla’s modules are unusual. What the rest of the world calls an oscillator (or, more accurately, an oscillator/LFO pair), Buchla calls a Complex Waveform Generator. In place of envelope generators, he offers Quad Function Generators. The Dual Arbitrary Function Generator might more familiarly be called a step sequencer.

Most of the knobs rotate smoothly, but a few are stepped to allow entry of precise values. Flat pushbuttons are used for stepping through various switchable options (such as the sawtooth, square, and triangle waves in the Mod Osc section). The filter, MIDI, and 249e sequencer modules have small two-line blue-on-black LCDs for displaying parameter values and other essentials.

Tying the system together are the MIDI/USB Decoder, which allows up to eight-note polyphony via MIDI (though you’ll need extra oscillators to go beyond four notes at a time), and the Preset Manager, which is physically part of the same module. Each module in the system is capable of storing its own knob and switch settings in up to 30 user presets. Even patch cord routings can be stored, to a very limited extent, using the Control and Signal Router module.

Presets can be saved and loaded using a small memory-card slot on the rear panel. This slot can also be used for firmware updates. While I was working on the review, Buchla sent me a new software rev that fixed a couple of bugs and added new features to the 259e oscillator. Can’t do that on any other modular.

As usual with an instrument this complex, you shouldn’t expect to find every detail discussed in these pages. Overall, I found the 200e system stimulating to work with. The initial release, which contained four 259e oscillators, was less than entirely satisfying sonically, for reasons that are discussed below. But just before press time I drove back into Berkeley for a preview of the 261e oscillator, which sounds much, much better. I also spent an hour patching and listening to the 256e Quad Control Voltage Processor ($1,000) and 250e Arbitrary Function Generator ($1,700), both of which are discussed below.

259e and 261e Complex Waveform Generators

The 259e and 261e each contain two sub-modules — an audio oscillator and a simpler modulation oscillator. You can modulate the main oscillator with the mod osc without using a patch cord, but the mod osc has its own output jacks (one each of audio and CV), so it can modulate other things as well. While the 259e and 261e share this basic design, they diverge in many particulars. Since the review system contained four 259e modules, we’ll start by discussing them, and then turn to the 261e, which is priced identically.

The 259e mod osc has three basic waveforms (triangle, square, and sawtooth). The main 259e osc has a more complex timbre control section consisting of a dual waveform selector and knobs labelled “morph” and “warp.” In simple terms, this oscillator uses a sine wave to sweep two digital wavetables. Each table contains a single-cycle waveform. The warp knob determines how much of the wavetable is swept — higher warp values produce higher overtones — and the morph knob controls the blend of the two wavetables. Each side of the morph has a choice of eight tables. And of course, both warp and morph can be voltage-controlled for vivid waveform animation.

Did he say “digital?” Yes, the 259e is a digital oscillator. Analog purists may not be pleased about this. I’m happy to have lots of waveforms at my fingertips, but I found that many of the wavetable combinations produced aliasing. Pure sine waves, which don’t alias, are available among the waveforms, and it was possible to coax some sweet tones out of the 259e, especially at low warp settings and at lower pitches. However, the higher-pitched tones of most of the waveforms alias spectacularly, making them ideal for edgy, glitchy mixes but not desirable for traditional definitions of “beautiful.” According to Buchla, this design was intentional.

I found that feeding a voltage into the oscillator’s CV in made the pitch slightly unstable. With pitch sweeps, the instability won’t be apparent, but when I used the 249e sequencer to play a melodic pattern that had sustained pitches, the oscillators jittered in an unpleasant, nervous-making way. The jitter was so slight that it was perceived as timbral changes rather than pitch changes — imagine a smooth surface on a painting with tiny flecks of sand embedded in the pigment. The only way to make the sand go away is not to use the CV input, which makes the sequencer and MIDI input less useful.

Buchla tells me he’s aware of this problem. A fix has been implemented in the 261e, and it will shortly (perhaps even by press time) make its way back into the 259e. He views the entire 200e system as a “work-in-progress.”

Both the main and mod oscillators in the 259e have continuous full-range tuning knobs, plus a fine-tune knob, but no octave or semitone switching. The mod osc will operate in low-frequency, high (normal) frequency, or pitch track mode, and can sync to the main osc for waveform sync sweeps. When I switched it to hard sync and listened to the output, the tone was harmonically related to the main osc, but I didn’t hear the classic analog hard-sync effect used for lead patches. The tone was more grainy.

Personally, I don’t feel the low-frequency range goes as deep as it should: The slowest this LFO will go is .25Hz (one cycle every four seconds). Sweeps lasting ten times that long are musically interesting. The 281e Quad Function Generator will slow down to a 20 second period, or even slower with voltage control.

Separate CV inputs with attenuators are provided for main osc pitch, mod osc pitch, warp, morph, and mod amount. In addition, both the main and mod oscillators have audio FM inputs with trim knobs. The FM has a more gurgly character than is typical for actual FM synths, making it suitable for brash effects but not mellow bell-like tones.

Missing from the 259e oscillator is a finely calibrated CV input for reliable pitch tracking. I was told I could get calibrated pitch tracking by turning the CV amount knob up all the way, but when I programmed diatonic scales or arpeggios over several octaves with the 249e sequencer and applied them to the oscillators, the scale was a bit stretched, making tonal music somewhat problematical even in the absence of pitch jitter.

I had a much more enjoyable experience with the new 261e. Like the 259e, the 261e can produce a wide variety of waveforms, and several dimensions of waveshape are voltage-controllable. However, the 261e is anti-aliased and uses analog waveshaping. Some may prefer the more untamed sound of the 259e, but if I were ordering a 200e system, I would probably specify four 261e’s.

The 261e has three waveshaping parameters, whose knobs are labelled “Timbre,” “Symmetry,” and “Order.” Turning up Order shifts the waveform energy away from the fundamental and into the upper overtones. Anything from a sine wave to a rich bell-like tone can be produced with ease. The CV input for modulating timbre has a trimpot, while the symmetry and order CV inputs don’t.

The 261e mod osc has a continuously variable waveshape. When locked with hard sync, the mod osc tracks the pitch of the main osc, which opens up even more tone color possibilities. One other detail: the 261e main osc has a recessed trimpot, allowing its CV input to be calibrated precisely to the Buchla standard of 1.2V/octave (.1 volts/semitone).

291e Triple Morphing Filter

What’s a Triple Morphing Filter? Glad you asked. This module includes three separate resonant bandpass filters (the “triple” part) and its own eight-step sequencer memory (the “morphing” part). The three filters can be used independently, or run in series by patching the output of filter A into the input of filter B and/or the output of B into C. They can also run in parallel using the “all” input and output.

The three share one set of knobs for cutoff frequency, bandwidth (resonance), amplitude, and modulation depth. To select which filter you want to edit, you press the Node button. The cutoff knob is stepped rather than smoothly rotating. It’s calibrated in half-steps, which is great when you want to run white noise through it and dial in a chord. But if you were hoping to be able to move the cutoff manually for smooth sweeps, you’ll be disappointed. The CV inputs have a smooth response, however.

Each filter within the module has its own CV inputs for cutoff, bandwidth, and amplitude control. The amplitude control affects only the “all” output, but turns out to be surprisingly useful for setting up rhythmic patterns with rests and accents. Each filter also has an audio modulation input jack. Audio-rate modulation can be applied to amplitude, bandwidth, and cutoff frequency in any combination. The level of this modulation is programmable from a knob, but the CV inputs have no trim pots. This is a real shortcoming: If you want to scale the envelope amount going to filter cutoff (a common necessity), you’ll have to attenuate the voltage using a signal path within the 210e Control And Signal Router or the 256e Quad Control Voltage Processor. The 210e can’t invert voltages, so you’ll need a 256e if you want to be able to invert filter envelopes. CV signals will modulate the cutoff, amplitude, and/or bandwidth of all of the steps when the filter is being morphed, because the CV input amount is not morphable.

There are three ways to move through the eight-step morphing sequence: You can set a time value for each step (which will be multiplied by an incoming voltage), move around in the memory space with a voltage, or pulse the steps with an external clock. The three filters within the module will all use the same step at the same time, but each of them can have its own parameter values for a given step. When using the filter’s internal time settings, you can choose whether to jump or glide from step to step, creating either smooth contours or sharp accents.

It was when I started exploring the 291e that I really started having fun with the 200e system. By setting up a master clock in a Quad Function Generator, a three-step sequence in one 291e, and a four-step sequence in the other, I was able to create a complex rhythmic drone (buchla2_filters.mp3) that would be difficult to achieve on any other synth.

281e Quad Function Generator

The 281e has four separate stages, each of which can function as an LFO or as an attack-decay or attack-release envelope generator. These three modes (which Buchla calls cyclic, transient, and sustained) are selected with pushbuttons. Each of the stages has its own attack and decay time knobs, which can be set from .001 second to 10 seconds. Both knobs are voltage-controllable.

Each stage has a pulse input for triggering it, and also a pulse output that fires at the end of the cycle. Thus two or more stages can be chained to fire in order for complex contours. Stages A and B (or stages C and D) can be linked in a “quadrature” mode to provide ADSR envelopes. However, the sustain level is not stored in the 281e’s memory.

256e Quad Control Voltage Processor

The 256e was not included in the system I had for review, but I later got a preview of it, and I recommend it highly for inclusion in a system. It has four separate and identical sub-modules, each of which can scale, offset, invert, and process control voltages.

Each sub-module has two CV inputs, plus a third input that crossfades between the other two. If only one of the two main inputs is connected, the third will scale it up or down. The resulting signal then passes through a transfer function, which has four knobs and an LED display to give you a rough picture of what it’s doing. The first knob controls the voltage output when the input is 0 volts, and the next knob controls the output when the input is at 10 volts. By turning the first all the way up and the second all the way down, you simply invert the CV as it passes through.

Between those two extremes is a breakpoint, which is controlled by both X and Y knobs. Setting the two outer values low and the breakpoint high (or vice-versa) effectively doubles the frequency of an LFO signal. Other curves are possible, and the sub-modules can even be patched to one another in a positive feedback loop for some quasi-chaotic modulation.

292e Quad Dynamics Manager

The 292e contains four separate VCAs, each with an audio input and two audio out jacks. There’s also an “all” out jack, so you can use the 292e as a four-in, one-out, voltage controlled mixer. You can also adjust the levels of the signals manually with knobs rather than using control voltages.

MIDI velocity from the first four MIDI busses is hardwired to the 292e’s channels, but this signal can be overridden with a voltage if desired. The velocity input does nothing unless a primary envelope is being received or the VCA is opened manually, but once a signal is passing through, the “velocity” input can vary it continuously, not just at the start of a note, so complex double-envelope rhythms are possible.

Each VCA can operate in one of three modes — standard (affecting only amplitude), lowpass (in which the cutoff frequency corresponds to the amplitude), or a blend of the two. The combination mode is quite useful for creating natural-sounding dynamic variations.

227e System Interface

The 227e has four input channels for receiving internal audio signals and routing them to the four output channels. Quarter-inch outputs are included on the rear panel, and each output channel also has a pair of jacks on the front panel. In addition, the 227e sports a mic input with a preamp and an envelope follower, allowing the system to process external audio.

Left/right and front/rear panning can be voltage-controlled for each input channel. Each input channel also has a quadrophonic “swirl” mode, with which the signal can be automatically swept around the four output channels with voltage-controlled rate and depth.

A 4-in, 1-out submixer (not programmable) is included in this module for convenience, with an “in the mix” button for routing it monophonically to the main outputs. The front and rear channels have their own non-programmable level knobs and three-band EQ.

225e MIDI/USB Decoder

With the aid of the 225e, a System 200e can respond to up to eight MIDI notes (a maximum of four per channel) and up to 12 controller message types on up to six different channels. Pitchbend and channel aftertouch are included among the controller types along with all 128 Control Change messages. The outputs are spread across 14 MIDI busses, eight for notes and six for controllers. Note busses from among each group of four can be cascaded for polyphonic operation, assigned to separate MIDI channels for multitimbral operation, or assigned to the same channel monophonically for doubling or layering.

The first four MIDI busses are hardwired internally to the four oscillators, to the four sections of one of the 281e quad function generators (which receive gate information for envelope purposes), and to the four channels of one of the 292e quad VCAs (which receive velocity information for amplitude control). This is a sensible system. In the unit I had for review, MIDI pitchbend data was not sent to the oscillators, but I’m told this has now been fixed. Likewise, Buchla reports that MIDI sustain pedal messages now sustain notes. Note-on, pitch, and velocity voltages from the second four MIDI note busses appear on front-panel jacks, where they’re available as general-purpose control sources. Ten preset velocity response curves are available, including three reversed types suited for velocity crossfading. Notes can be transposed up or down by up to 24 half-steps, and 1/10-semitone fine tuning is available for each bus.

The first time I tried using the internal busses for controlling the pitch of the 259e, I heard small tuning discrepancies up and down the keyboard. This turned out to be user error: The FM amount inputs for the oscillators have to be turned down to zero for reliable MIDI pitch tracking. Even after I zeroed out these knobs, I still heard small pitch errors at the low end of the keyboard. When I played a scale in octaves, some octaves were dead on, while others had slow beats. Buchla reports that this is due to the 1Hz resolution of the MIDI-to-pitch conversion. “At 50Hz,” he said in an email, “there is an uncertainty of 0.5Hz, or about 15 cents. The pitch resolution of the ear is worse at these frequencies, so I let it pass. I am aware of no user objections, and actually doubt that users will perceive the errors.”

While USB is mentioned on the panel of this module, the system I had for review had no USB port.

249e Dual Arbitrary Function Generator and 250e Arbitrary Function Generator

If the words “step sequencer” conjure up memories of early Tangerine Dream for you, you’ll understand about one percent of what the 249e and 250e can do. Yes, either module can step robotically through a simple pitch

pattern. But both the rhythm and the looping can be controlled in complex ways. The system I had for review included the 249e, but I later got a look at the 250e, which is similar in many ways but more interactive.

The 249e consists of two independent 24-position step sequencers. Each has its own pair of CV outs, another pair of programmable pulse outputs, and a pulse output that fires each time the sequencer moves from step to step. Each step defines the value that will be sent from the two CV outs, on/off switching for the two pulse outs (suitable for triggering envelopes on certain steps, for example), a time value, and some logic functions. The decision to advance from one step to another can be handled by the sequencer’s internal clock, by an external pulse, or by a continuous voltage that sweeps through the steps.

Programming is handled by tapping a button to call up one of half a dozen menus in the little blue LCD. You choose which step to edit by rotating a knob in the center of the module’s panel. Entering the data to define a sequence is a fiddly process, not something you’ll likely want to do during a concert. (The 250e is much better for live use, as explained below.)

The two CV outs for a step can be programmed in increments of 1/100 volt — 1/10 semitone increments at 1.2 volts per octave. If you switch on interpolation for either CV at a given step, the voltage will glide at a fixed rate from the previous step. Instead of outputting a predetermined voltage, a step can output the voltage value at one of four external input jacks.

To program sequences of less than 24 steps, the tool of choice is the jump button, which can be programmed with a jump to another step. The jump can be given a probability of less than 100% or made conditional on the current value of a loop counter. This counter, in turn, can be set or decremented by another step.

Each step can be given a time value of up to 30 seconds in .01-second increments. A time scaling knob can be used to accelerate or decelerate globally between 0.5x and 2x these values. CV inputs allow you to stop or start/advance the sequence with an external command.

Instead of using the clock at all, you can select the step directly with a control voltage — or, alternatively, with two control voltages in an X/Y configuration, suitable for the output of a two-dimensional touchpad controller. Plugging separate LFOs into these two inputs is a fun way to sweep the sequencer rapidly through its 24 steps in a semi-controlled way.

The 250e has only a single 16-step sequencer. This has dual CV outputs, pulse outputs, and clocking options similar to those on the 249e. It differs in having individual pitch and time knobs for each step, making it very playable. In place of the two-line LCD is a pair of blue two-digit LEDs, which are used for programming jumps. The voltage range for each step is selectable, which makes it practical to rotate a pitch knob within a musically useful pitch range.

260e Duophonic Pitch Class Generator

If the 260e weren’t included in the system, few musicians would miss it. Its purpose is to produce Shepard tones, an acoustic illusion of a tone that rises or falls forever. The module can do this in either stepped or continuous fashion, but in continuous mode it glitches, which rather destroys the illusion. I’d recommend purchasing a 256e, which is the same panel size, instead.

266e Source of Uncertainty


Across the top of the 266e are four audio outputs supplying noise in three different colors. But most of the module is devoted to generating quasi-random control voltages. It has three sections: Fluctuating Random Voltages, Quantized Random Voltages, and Stored Random Voltages.

The fluctuating outputs (there are two) supply a sort of LFO-like contour that speeds up and slows down at a random rate. The quantized outputs (again, there are two, with separate triggers) can assume an adjustable number of repeating states, with a few choices for distribution curves. The result is a stepped pattern in which the output voltage is chosen at random from among a fixed group of values. By tuning the CV input amount in a receiving oscillator, you can step randomly through a whole-tone scale or the notes in an augmented triad (check out buchla3_uncertainty.mp3).

The Stored section comes closer to the classic sample-and-hold. It has three controls, all of them voltage-controllable: skew, degree, and chaos. When degree and chaos are low and a voltage is applied to skew, the stepped output is indeed a sample-and-hold. The degree control introduces randomness in the steps, and increasing the chaos control allows more of the random steps to be large. The two stages within the Stored section can be clocked independently, but share the same set of knobs and control inputs.

Part of the fun of a modular synth, of course, is that you can apply a stepped random voltage to other things than pitch (buchla4_uncertainty2.mp3).

210e Control and Signal Router

The 210e, a unique and very valuable module, has two separate routing matrices, one for audio signals and the other for control voltages. Each has eight inputs and five outputs. They operate similarly: After choosing an input and an output with a pair of pushbuttons (which cause the matching LEDs in a grid to light up), you adjust the level of the signal from that in to that out using a knob. The audio matrix has an Input Solo button, which is very handy.

The CV section’s level knob has two LEDs, one marked “<0” and the other “>0”. This led me to expect that the 210e was capable of inverting control voltages. This proved not to be the case.

Since the 210e is programmable, it can be used to a limited extent as a patch cord routing memory. For instance, if you always plug the eight outputs from the 281e stages into the eight CV inputs, and route the five outputs to five of the cutoff frequency CV inputs of five of the system’s six filter stages, then filter envelope depth becomes programmable. But if you should need to repatch the CV cables for any other purpose, this module’s memory won’t know what you’ve done.

Conclusions

Musicians who perform and record with hardware-based modular synthesizers are sure to embrace the 200e system. Its programmable memory puts it in a class by itself, and its unique feature set opens up a vast range of sonic possibilities.

There are, of course, less expensive analog synthesizer modules on the market. But it’s extremely difficult to make a fair across-the-board comparison between modulars, because each system, and indeed each individual module, has its own strengths, limitations, and peculiarities. To assemble a system from another manufacturer that had the same functionality as the 200e would probably cost almost as much (if you could do it at all), and the resulting instrument would undoubtedly weigh three to five times as much.

In 2005, the competition for the 200e includes not only other hardware modular synths, but also software-based instruments like Native Instruments Reaktor and Cycling ’74 Max/MSP, both of which sport far deeper

feature lists than the 200e and are extensible in ways that no hardware instrument could ever be. Given the power and sound quality of today’s software, the high price of the 200e is bound to limit its appeal. But Don Buchla had no illusions about this when he designed the 200e.

For the true hardware fanatic, though, comparisons with software are beside the point. If you get sweaty palms when you see real knobs and patch cord spaghetti, you’d be foolish not to consider a 200e system as the centerpiece of your rig.

 
 
 
 
 
 
 
 
 
 
 
Leave a Comment
Name:
Location:
Average Rating :
¿``7`7890- 07890-012.3201XZC VBNMP;[']}"A¸?
`012.3XZC VBNMP;[']}"A¸?
`XZC VBNMP;[']}\
`XZC VBNMP;[']}\
`XZC VBNMP;[']
 

Finger Independence

Chester Thompson B-3 Master Class

Ten Minute Technique - Warming Up Under the Gun

The Chord Doctor - Expand Your Chordal Command

Get Funky on the Rhodes

 






What type of product has improved the LEAST over the past 10 to 15 years?
 
Subscribe Live Bookmarks Advertise Contact Us Privacy Policy Terms & Conditions
 



 
Keybord Magazine is a trademark of New Bay Media, LLC. All material published on www.keyboardmag.com is copyrighted @2009 by New Bay Media, LLC. All rights reserved