The Crumar 'Bit' Series

I remember the first time I played a Bit ONE. It was a hot summer day in 1984, and my girlfriend and I had travelled to central London to buy a set of Chase CBP1 bass pedals. An Italian company made these for Chase (a keyboard store more often known as the London Synthesiser Centre) and they were, in many ways, the poor man’s Taurus pedals.

Soon after we arrived, Mr Pammi Singh, the store’s manager and an industry 'character' without peer, calmly informed be that I really needed the CBP2 bass pedals because they sported a second oscillator. They cost, he told me, a mere £100 more. “Hold on”, I said, “I can barely afford £199 for the standard model, let alone £299 for those.” At which point Mr Singh uttered the immortal line, "Ah, but with a girlfriend like that, how can you be poor?"

While I was battling with my better judgment and my cheque book, Mr Singh showed me a new keyboard that he had just received from Italy. It was (in his words) a digital synth that, at £799, cost just half the price of a DX7 yet blew away the equivalent Rolands and Korgs of the day. It was black, sleek and, despite a 5-octave keyboard, very compact. It was, he told me, warm, fat, flexible, expandable, and it had MIDI. It was a Bit ONE.

"Bit?” I asked, “who are they?" In retrospect, it was hardly surprising that I had never heard the name, because it turned out to be the nom de plume of a far more famous manufacturer. Like the bass pedals that I had set out to buy, the Bit ONE was one of the progeny of Crumar, a company better known for its ghastly electric pianos, weedy string machines, and early ‘multi-keyboards’.

Crumar had shot to prominence in the mid-1970s when a handful of bands such as Emerson Lake & Palmer, Greenslade, and The Enid adopted some of their quirky polyphonic ensembles. Unfortunately, these instruments seemed to have no place alongside the Moogs, ARPs and Oberheims of the day. They even suffered in comparison with other Italian imports including Logan string ensembles and Elka Rhapsodies.

It was, nonetheless, an important company, if only because it made keyboards that were cheap enough for players who would otherwise have been restricted to lusting after their heroes’ unaffordable instruments. Furthermore, the Crumar Multiman assured the brand’s place in history because it was the world’s first ‘multi-keyboard’, a concept that was subsequently copied by ARP, Moog, Yamaha, Korg, and many others. But Crumar’s name always conjured an image of cheap shoddiness, so serious players avoided them, no matter how innovative or well designed they later proved to be. (It’s interesting to speculate how successful the Spirit monosynth or the Organiser T2 might have been had they been manufactured elsewhere.)

Anyway, it was clearly a problem that Crumar understood. When the company designed its first MIDI keyboard, it took a leap in the dark and changed its name to "Bit" and called its first product the Bit ONE.

The technology

The Bit ONE was not a digital synth, although it did contain digital technology. It was an analogue/digital hybrid that conformed to the standards of the day by being 6-note polyphonic, but that was where many of the similarities ended. For example, instead of following the lead of low-cost single-oscillator-per-voice synths such as the Korg Polysix and Roland Juno 106, its dual-oscillator voice structure was more akin to that of the Prophet 5 and Oberheim OB8. And, whereas many of its competitors offered just a single ADSR envelope per voice and a single LFO per patch, the Bit ONE featured two of each, and you could assign these to many more destinations than the low-cost Rolands or Korg allowed.

What's more, the earliest Bit ONEs incorporated SSM filter chips, while later Bit ONEs, Bit 01 Expanders and Bit 99s used Curtis (CEM) chips. These undoubtedly contribute to the warmth of the synths, and they are a significant element in their desirability. Even more impressively, the Bit ONE was velocity sensitive. This was no afterthought: you could direct the velocity response independently to each of the LFOs, each of the DCOs’ pulse-widths, and to the attack times and envelope depths of both envelope generators, thus making both the VCF and the VCA touch-sensitive. This was extremely powerful stuff, and it should have made the Bit ONE the low-cost analogue synth of choice for most players. But there was more…

For example, the Bit was bi-timbral, and you could place any of its patches in Upper and Lower zones above and below a user-defined split point. There were also Upper and Lower outputs, so you could direct each zone to its own mixer channel for independent EQ and effects treatments. It also had a duophonic Unison mode, and in the standard ‘Play’ mode this stacked voices 1 to 3 under one key, and voices 4 to 6 under another, thus placing six oscillators under each finger. You could even select Unison in ‘Split’ mode, and then you got two different sounds; one created from the three voices in the Lower zone, and another from the three voices in the Upper zone.

If this wasn’t enough, a ‘Double’ mode allowed you to allocate two patches across the whole keyboard. This meant that, instead of playing one patch in the Upper zone of the keyboard and one patch in the Lower, you heard the two superimposed upon each other. As you might expect, this reduced the polyphony from six notes to three, but the depth and complexity of the sounds more than made up for this.

If this makes the Bit ONE sound like some sort of super-synth, it wasn’t. Sure, it had great strengths and, when you consider that its designer was Mario Maggi, the man responsible for the Elka Synthex, this isn’t too surprising. But some of its weaknesses were astounding. Minor complaints included the position of the modulation and pitch-bend wheels, which lurked one behind the other on the control panel itself. Far more serious was its inability to balance oscillator levels: DCO1 and DCO2 were always either ‘on’ or ‘off’.

Another unexpected failing was the lack of pulse width modulation by an LFO. Sure, you could vary the pulse width using an envelope or by velocity response, but you couldn’t sweep it for those classic, lush PWM sounds. Even worse was the omission of fine tuning and detune from the programming system. Instead of being programmable within each patch, you determined the detune between DCO1 and DCO2 by moving a slider on the top panel. So, if you wanted to jump between a thick ensemble patch and a thinner sound, you had to adjust this slider as you played.

The programming system also lacked Noise. Sure, it was present on the synth, but you could only access it using a top panel slider. This was arcane. If you wanted a patch based purely on noise, you had to program a patch with both oscillators switched off, then introduce the noise from the front panel whenever you selected that patch. Weird, or what?

Finally, the MIDI specification was a joke, even in 1984. Compare it to its contemporary, the Juno 106, which understood all 16 MIDI channels, and transmitted and received note numbers, patch numbers, performance data, and any control panel changes. In contrast, the Bit ONE transmitted just note number and velocity on MIDI channel 1. When receiving, (Omni mode only) it also recognised patch changes. That was it! So, despite In, Out and Thru sockets, you couldn’t use the Bit ONE to select patches on other keyboards or expanders, and you couldn’t dump patches via MIDI. If you wanted to create a patch library, you had to use cassette tape storage.

The Next Generation

In early 1985, Crumar addressed many of these problems by releasing an improved Bit synthesiser. Contrary to expectations, this was not a Bit TWO keyboard. It was the Bit 01 Expander module, which was also marketed in the US as the Unique DBE.

The deep, 3U high Bit 01 (which came in both black and ivory paint schemes) was far from being simply a repackaged Bit ONE. It was very much a ‘souped-up’ version of the instrument, with extra parameters and a dramatically improved MIDI specification. Additions included a parameter to control the maximum modulation depth applied by the mod wheel, programmable noise on DCO1, programmable detune on DCO2, and programmable volume for the complete patch. Unfortunately, the Bit 01 lost the Bit ONE's ability to sweep the DCOs’ frequencies using the VCA’s envelope generator, and the Unison mode disappeared. On the other hand (and unlike the Bit ONE) the Bit 01 now allowed you to store ‘splits’ and ‘doubles’ as performances. Indeed, its memory was increased from the 63 single patches found on the Bit ONE to 99 patches/performances. It was a strange system - 75 of the 01’s memories were Bit ONE style patches, and the other 24 were ‘split’ or ‘doubled’ performances - but a welcome enhancement nonetheless.

Equally important, the MIDI specification was much better than before. You could now leave Omni Mode, and the Bit 01 recognised all 16 MIDI channels. It also received controllers such as modulation and pitch-bend so, for live work, it was a better synth in almost every way. Unfortunately, the 01 lacked a MIDI Out socket, so it was still impossible to save and restore sounds over MIDI. Nevertheless, at £499, the 01 provided excellent value for money, and it should have been high on everybody’s analogue shopping list.

The Final Cut

Had Crumar released the Bit 01 as a keyboard I might have gone elsewhere when, in the summer of 1985, I bought my first MIDI keyboard, a Roland JX8P. This boasted much the same specification as the Bit 01, although it also offered oscillator sync, an oscillator mixer, Roland’s famous chorus effect, and pressure sensitivity. Three months later and (as far as I was concerned) with appalling timing, Crumar released the Bit 99. Like the 01 Expander, this came in both black and ivory versions, but in every other way it was the instrument that we had expected earlier in the year.

In retrospect, the Bit 99 was simply a Bit 01 Expander with a keyboard, performance controls, additional MIDI commands that the expander (lacking a keyboard) had not required, and the ability to chain three sets of 33 patches. This last attribute was an excellent feature, although one overlooked by most players. The JX8P allowed me to chain eight patches but, in my view, the Bit’s ability to step through 99 patches during a gig placed it in a league of its own. At its reduced recommended price of just £699, the ‘99’ should have been a winner.

Retrospective

That the Bits were not resounding successes, and that Crumar disappeared just three years later, are matters of history. Whatever the reasons were, it could not have been the sound of the instruments themselves. Their lead synth patches and synth-pads were remarkable, and they also excelled at punchy brass and polysynth patches. Indeed, many of the Bits’ analogue sounds comfortably filled the shoes of the far more expensive Prophets and Oberheims. The Bits were also capable of a wide range of DX-style electric pianos and harps.

Sure, none of them were perfect. For example, strange summing differences between DCO1 and DCO2 could give unexpected tonal responses whenever you played a key. This, depending upon your point of view, was either a benefit or a pain in the neck. Oddly, the differences did not seem to cycle as you played the six voices, so you could never be sure when it would occur, or by how much. And there were other minor annoyances such as the (sometimes) audible stepping of the LFOs. But these deficiencies should not have detracted from a warm and versatile family of synthesisers that could out-perform many more illustrious instruments.

So why did the Bits disappear so quickly? Maybe it was the lack of tweakable knobs and switches. Maybe it was the lack of software support. Maybe, in the UK, it was the mistake of trying to sell them exclusively through one dealer based in Central London.

I reckon that it comes down to this: the Bits were the right synthesisers at the right prices at the wrong time. The mid-1980s were the hey-day of FM synthesis, and the world had temporarily tired of all things analogue. But today, playing a Bit through a bit of chorus and a splash of reverb is a very rewarding experience. And, for those fortunate enough to be able to so, layering two Bits through a stereo effects unit can produce stunning results. If you ever have the good fortune to get your hands on a pair of Bits (no giggling please) you’ll soon begin to wonder whether it’s time to stop lusting after those ridiculously expensive and unreliable vintage synths and to look at some of the less fashionable alternatives available at a fraction of the price.

A note about the Bits' oscillators

The DCOs in the Bits generate their waveforms using a principle known as Additive Squarewave Synthesis. In simple terms, this means that the fundamental oscillator in the system generates a high frequency squarewave that is subsequently divided and divided again and divided again… The multiple squarewaves thus generated are then added in various ways to approximate the triangle-, sawtooth-, and pulse- waveforms that the synth offers. The dividing circuit has 16-bit resolution, so the numbers involved can be very accurate and the waveforms can be close approximations to the real things. But the Bits’ oscillators are far from analogue, and you could just as easily describe them as true digital oscillators rather than DCOs.

Copyright ©1998, Gordon Reid.

Back to vintage keyboards list