Magazine Archive

Home -> Magazines -> Issues -> Articles in this issue -> View

Back To The Future

Cheetah MS6 Multitimbral Analogue Synthesizer

Seems like everyone has been waiting to get their hands on this low-cost 6-voice, multitimbral, analogue synth expander. Martin Russ looks at the affordable face of technology.

Martin Russ looks at the affordable face of technology.

Cheetah have diversified from the computer accessories for which they are best known and established a reputation as an innovative manufacturer of low-cost products which everyone else seems to have overlooked. Their MK range of master keyboards have become virtually the only choice for people on a tight budget, and their MD8 drum machine shows that 8-bit sounds and cassette storage can still provide a versatile and usable unit at a very reasonable cost.

The latest addition to this stable of 'affordable' products also goes slightly against the trend; the MS6 is, in fact, a rack-mounting, 6-voice, multitimbral analogue synthesizer for under £300. On the face of it, there would appear to be similarities between the MS6 and the Oberheim Matrix 1000 reviewed in September's magazine. However, the price, sound and features of the Cheetah MS6 make it an entirely different proposition, and only superficially similar to the Oberheim. (The unit under review is a pre-release production model, with Version 5.0 of the System EPROM. Final release models will have System 6.0.)


The MS6 uses the latest in both analogue and digital technology to produce a multitimbral, 6-voice, subtractive synthesis expander. For the technically minded, an 8MHz Hitachi 63B03 microcontroller (as found in virtually all Japanese synths) and a 27256 system ROM act as the system controller for six Curtis CEM 3396 synthesizer chips (the sound generators). A large number of Toshiba 4051 analogue multiplexer chips are used to determine signal and control routings, and there are even discrete transistors! All the major components are mounted on a single large PCB which fills almost the entire box.

The front panel is served by an additional PCB used for the switches and LED display. Two transformers are used in the power supply, presumably to isolate the analogue and digital rails. There is a single trimming potentiometer associated with the two 8-bit DACs. Overall the inside is neat, tidy and professional - remarkably similar to the Japanese-made Oberheim Matrix 1000, both in layout and in the quality of construction.


The exterior of the MS6 is much more likely to be the focus of attention, so let's examine what there is.

The MS6 is a 1U high, 19" rackmounting unit, nowhere near as deep as many rack units (200mm), and with a sturdy all-steel construction. The front panel has a single pole mains switch, four bright red 7-segment LED displays (visible from a considerable distance) and twelve positive click switches - arranged in two rows of six. Everything is clearly labelled in light blue or red, and only four switches have more than one function, which is a welcome change from the norm. Notable by its absence is a headphone socket (although you can drive headphones from the audio output socket via a mono-to-stereo jack convertor). The top of the MS6 case features the traditional silk screened chart of the parameters - Cheetah told me that they intend to supply a 'Roland' style programming card with all the parameters listed.

The rear panel has a voltage selector, a rather short 4-foot permanently attached mains lead (an IEC socket would have scored more points, but would have probably increased the price), the full complement of MIDI In, Thru and Out sockets, and a mono audio output jack socket. The fuse for the unit is internally mounted, and not accessible from outside.


The MS6 has only two modes of operation: Play mode (the power-up default) and Edit mode. In Play mode the 1 to 8 numeric buttons are used to select the sounds, using the 11 to 88 method, providing access to 64 sounds. The Bank button increments the bank number from 1 to 8, and then back to 1 in a cyclic fashion. The LED display shows the bank number and the memory location (or 'tone number' in Cheetah-speak).

Banks 1 to 5 are ROM banks and contain the factory preset sounds. Bank 6 is a RAM bank and is used for storing sounds which you will use in the performance memories. Bank 7 is another RAM bank, and the only one with 32 memory locations (all the rest have 64) - it is used to store your own 6-voice polyphonic sounds. Bank 8 is used to store performance memories.

Two buttons are permanently assigned to controlling the volume when you are in Play mode, which explains the absence of a volume control knob. They temporarily increment or decrement the VCA volume parameter from 0 to 63, and thus control the overall output level of the selected voice whilst you are playing it (permanent changes are made by editing the parameter as usual).

The remaining button selects Edit mode. The LED display now shows the parameter being edited (it always remembers the last edit you made - a nice touch) and the current value. Some parameter values range from 0 to 63, others are 0 to 1 or 2 for switch-type functions, with some controls extending from -62 to +63 (the minus sign is shown in the display by a decimal point in front of the parameter value). You select the parameter number with the numeric buttons, 1 to 8, and increment or decrement with the two buttons used for increasing or decreasing the volume in Play mode. I found that the ergonomic design of this was very effective, since the buttons behaved in exactly the same ways in each mode - if only all front panels were as easy to use as this! The remaining switch is now the 'Write' switch, and is used to store edits into RAM memory locations.

When in bank 8 in Play mode, the locations select performances instead. In Edit mode, the parameters from 11 to 67 let you edit the performance memories, whilst those from 71 to 84 let you carry out utility functions: master tuning, calibration routine, MIDI bulk dumps and loads, etc. Overall, the front panel was easy to find your way around, unambiguous, and fast in operation - obviously the result of a lot of hard and careful thought.


The MS6 uses a conventional 2 DCO, 1 VCF, 1 VCA, 2 EGs and 2 LFOs per voice type of structure. There are 59 parameters covering the majority of the controls you would expect. I was impressed with the separate LFO for the pulse width modulation (PWM) and the amplitude and attack sensitivity of the envelopes to MIDI velocity. You can use the aftertouch or pressure sensitivity to affect the sound as well. The whole system is designed to provide comprehensive and expressive control over the sound generation process. Cheetah call the system that the MS6 uses 'Wave Mix' synthesis, and this refers to a novel way of dynamically mixing between the two VCOs, which can be used to great effect.

A deliberate element in the design is the choice of available parameters. These have been tailored to enable the production of the 'classic' analogue sounds, rather than the more esoteric special effects. This makes the MS6 easier to use by those with limited analogue experience, and it is also consistent with the idea of a multitimbral (multiple voiced) unit, where you will be using it to produce instrumental timbres. You would probably be better off with a monophonic synthesizer or the Matrix 1000 for special effects - at least, this is what I use my Sequential Pro-One for!


The 6-voice poly sounds can be used to great effect in conjunction with digital instruments, producing anything from lush, warm textures to hard, bright stabs. Most importantly, the mixture tends to lose some of the characteristics which label it as a 'DX7' or a 'D50' sound. Instead, the combination of analogue and digital can gel into a distinctive and individual sound.

The MS6's multitimbral capabilities mean that sounds can be stacked to produce even richer, more complex tones, and here the results begin to move away from the simple analogue sounds and into regions more usually associated with the digital instruments. This is also helped by the Wave Mix synthesis method.

When you really start to exploit the multitimbrality with separate parts, the MS6 can take on another sonic appearance - careful choice of the sounds can result in new and unusual textures, often because analogue synthesizers are not commonly used in this way. I found that running some of my sequencer files into the MS6 produced Larry Fast/Synergy or Walter/Wendy Carlos type sounds, without any need for large modular synthesizers or weeks of recording!


Bank 8's performance memories are the route to exploring the MS6's multitimbral features. In each location you can store 36 parameters which determine the behaviour of each of the MS6's six voices. Assignment is done on an instrument basis - each instrument is assigned a MIDI channel and a number of voices as its polyphony. Normally in Play mode you would be using a single instrument with six voices on a single MIDI channel.

In the performance memories you have control over the bank number and tone number, the number of voices assigned to the instrument, the high and low note limits which enable keyboard splits and zones to be set, the MIDI channel, and the volume of the instrument. With six voices available you can set up any combination between a single 6-voice polyphonic instrument and six different monophonic instruments.

Detune is not a performance parameter, and so to produce rich, fat detuned stacked sounds you need to store slight variations of sounds in bank 6 (performances can only access banks 1 to 6, not bank 7 - so you use bank 7 for 6-voice poly sounds in Play mode), but this tends to eat up memory locations!

Selecting the parameters is made easier because the instrument number is the first number you select, followed by 1 to 7 for the performance parameter - so 16 sets the MIDI channel in instrument 1, and 26 sets it for instrument 2. This forethought also applies to the tone editing parameters, where the DCO and ENV parameters are similarly organised, making the parameter numbers much easier to learn. The 36-page A4 manual gives lots of information and hints on using these parameters, as well as describing exactly what they all do. It also gives extensive details of the MIDI bulk dump formats for voice and performance data, so I am sure we will see some computer-based editor/librarian programs appearing soon!


The MS6's MIDI implementation allows you to select the basic receive channel as Omni, or any single channel for the basic 6-voice Play mode. In the multitimbral performances you have complete freedom over channel assignment. MIDI velocity and aftertouch/pressure can be assigned to several control parameters inside the MS6 - the velocity can scale the attack times of the envelopes, or alter their size, and the aftertouch can be assigned to change the DCO pulse width or pitch bend, and also the cutoff frequency of the filter.

A MIDI Overflow function can be used to increase the polyphony available by automatically sending any unassigned note messages to a second MS6 (or any other synthesizer) instead of 'stealing' existing notes. Tone and performance bulk dumps can be sent and received by MIDI.

Because the DCOs in the synth chips will not generate frequencies below 30Hz, the MS6 will not respond correctly to MIDI note messages for any notes below CO (note 24), depending upon the master tuning - this is very similar to the octave-jumping which occurs at the extremes of the scale in most FM expanders.

When I first saw the MS6 manual, I commented to Cheetah on the lack of a MIDI implementation chart. I am pleased to be able to report that this omission has now been corrected, which demonstrates that they are willing to listen to users and take appropriate action - full marks!



  • 6-voice Polyphonic
  • Multitimbral - 6 instruments
  • 320 ROM sounds
  • 96 Non-volatile RAM sounds
  • 64 Performance Memories
  • Dual DCO, LFO, ENV per voice
  • 24dB/Octave VCF per voice
  • Velocity and Pressure sensitive
  • Fully programmable from the front panel
  • MIDI In, Out and Thru, with Overflow
  • LED display
  • Easy to use controls

At £299.95 the MS6 is remarkable value for money. The combination of multitimbrality and front panel editing of sounds make it a very different instrument to the Oberheim Matrix 1000 - and in many ways superior. At this price the only competition is the more expensive and totally different Kawai K1M. Interestingly, the two products together complement each other quite well - the MS6 supplying the missing filter-based dynamic timbre changes that the K1 lacks. In fact, the combination has many of the features of the Roland D50 and other LA synths which have become so popular recently: sampled sounds with conventional analogue synthesis.

In terms of usability, the MS6 comes quite close to exactly what I would have liked to see in the Oberheim Matrix 1000, and I am delighted that it is from a much more affordable UK product instead! In this age of digital synthesis the MS6's classic analogue voices may actually sound new to many ears, and in combination with other sounds (and what else is an expander for?) can make a refreshing addition to anyone's setup. It looks like the Mach 1+ joystick I got for my Atari may not be the only thing I buy from Cheetah!


£299.95 inc VAT.

Cheetah Marketing Ltd, (Contact Details).


I was raised on analogue synthesis. I built several voltage/current controlled monstrosities and drooled over the unreachably expensive synthesizers in music shops. When I graduated to demonstrating synthesizers professionally, everything was analogue - Roland's CSQ100 digital sequencer was a strange leap forward into completely unknown territory. VCOs (Voltage Controlled Oscillators), VCFs and VCAs were the name of the game, and years of using them gave me and many other keyboard players an inbuilt familiarity with analogue synthesis and large arrays of control knobs. Today the situation is rather different: a new generation of synthesis techniques has arrived - all digital and usually parameter access based. With digital has come a new generation of synthesists who know their algorithms or wavetables backwards, but who often have less idea about what a VCF is.

Conventional analogue synthesis is based around a few simple sonic building blocks, in much the same way as digital synthesizers. The only difference is in the function. There are sound sources and sound modifiers. The sound sources are the Oscillators (VCOs). These usually produce just a few basic sounds - the raw material from which the desired sound is extracted. Digital technology has made VCOs much less prone to drift by replacing them with DCOs - digitally controlled oscillators. The only other typical sound source is a noise generator, used for exactly the same purposes as in any other synthesis method: to remove some of the precision in the sound and make it more natural sounding. Unlike FM synthesis, where the possibilities for producing sounds are very large, DCOs tend to produce just a few harmonically rich sounds - almost all the processing is done by the modifiers.

The major modifier is the VCF (voltage controlled filter). This is used to control the harmonic content of the resulting sound. The sound from the DCOs is rather too harsh to be usable, and the filter is used to remove harmonics from it. As you increase the cutoff frequency control of the filter, more and more harmonics are allowed to pass through and the sound becomes brighter - much the same effect as increasing the output levels of the modulators in FM. Because the filter removes or subtracts harmonics from the raw sound, this technique is called subtractive synthesis.

The filtered sound is then released to the outside world through a VCA, an Amplifier which controls the overall volume envelope of the sound, and forms the second of the sound modifiers. The final tone is a result of the selection of a suitable harmonic structure from a rich starting point, and can be very different in sound from many digital instruments - much use is made of detuning and filter resonance effects. In general, analogue sounds are often described as 'warm', probably because of the frequency beating between DCOs which often results from the independence of the Oscillators - the exact opposite of many digital systems, in fact, where the sound sources are usually perfectly in tune with each other and any detuning can sound mechanical and lifeless.


To give you an idea of the type of sounds available, here are some personal favourites from the MS6 ROM banks. (Remember that you can easily edit these to your own taste. A useful starting point for customisation is parameter 51, the VCF cutoff frequency.)

1 37 Lush strings
1 81 Haunting strings/choir
2 65 JX Brass
3 18 Jarre 'Rendezvous' synth bass drone
4 16 Moog bass sound
4 64 'Clockwork Orange' bass
4 71 Rick Wakeman 'Rhapsodies' sync lead sound
5 67 Just like my Casio MT31!
5 74 Classic Filter Sweep

Particularly appealing is the way that aftertouch/pressure sensitivity is built into the ROM sounds - the Oberheim Matrix 1000, in comparison, virtually ignored this important means of introducing expression into playing. The ROM sounds in the MS6 show off most of the analogue repertoire nicely, although they are not perhaps as polished as the Matrix 1000 sounds. Given the talents of UK programmers, this is a situation which should soon be reversed.


No product is perfect, and so here are a few of the very minor things that normally no-one ever bothers to include in reviews:

A few of the ROM voices seemed to behave oddly. Cheetah said that the review model had an old system EPROM, and assured me that the voices would be OK in the release version.

The front panel PCB was slightly bent. Cheetah said that manufacturing tolerances could cause this but that it would not affect the unit's operation.

I also suffered the reviewer's nightmare - mains hum at a very low level in a few of the voices. Trying to track it down only convinced me that my home set-up needs rewiring. Cheetah said that studio-based testers had reported no problems.

Also featuring gear in this article

Previous Article in this issue

Trackman Sequencer

Next article in this issue

Village People

Sound On Sound - Copyright: SOS Publications Ltd.
The contents of this magazine are re-published here with the kind permission of SOS Publications Ltd.


Sound On Sound - Nov 1988

Donated & scanned by: Mike Gorman

Gear in this article:

Synthesizer Module > Cheetah > MS6

Gear Tags:

Analog Synth

Feature by Martin Russ

Previous article in this issue:

> Trackman Sequencer

Next article in this issue:

> Village People

Help Support The Things You Love

mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.

If you value this resource, you can support this project - it really helps!

Donations for January 2022
Issues donated this month: 2

New issues that have been donated or scanned for us this month.

Funds donated this month: £137.00

All donations and support are gratefully appreciated - thank you.

Magazines Needed - Can You Help?

Do you have any of these magazine issues?

> See all issues we need

If so, and you can donate, lend or scan them to help complete our archive, please get in touch via the Contribute page - thanks!

If you're enjoying the site, please consider supporting me to help build this archive...

...with a one time Donation, or a recurring Donation of just £2 a month. It really helps - thank you!

Small Print

Terms of usePrivacy