To: David Bivins dbivins@xxxxxxxx.com, From: Robin Whittle rw@xxxxxxx.com.au, cc: Subject: [AH] Casio *digital* keyboards - 1980-82 Date: Fri, 26 May 2000 13:52:33 +1000 --------------------------------------------------------------------------------- Hi David and All, Casio never made and analogue synthesiser. Their initial range of large and small keyboards uses a common technology. The timeline was something like this. The dates are either from brochures or service manuals. 1980 March CT-201 Black vinyl-covered wooden case, 4 octaves, two chips. Very rare! 1980 September M-10 Mini-keyboard, white plastic, 2.5 octaves, 1 chip. 1980 October CT-301 Home keyboard, steel case, 4 octaves, 1 chip. 1980 December MT-30 Mini keyboard, white plastic, 3 octaves, 1 chip. (Brochure shows woodgrain model too!) 1981 May CT-202 Similar to CT-201 but different sounds (49) and a very swish case. 1981 July Casiotone 101 Similar to CT-301 but with a better set of sounds. 1981 December MT-31 Electronically the same as the MT-30 - but with somewhat different case. 1981 December MT-40 MT-31 with chord buttons and rhythms. I think that is all of this initial series. Perhaps there were one or two more variations. All these were based on a single chip which scanned a four octave keyboard (no velocity) and control switches and which generated internally 8 notes. The waveforms of each sound were a fixed stair-step digital waveform - though some pulsated rhythmically. It is evident from the way they did it that the waveforms were not stored as such, but the increments between the steps were stored and the waveform calculated in real time. I believe this was to get around a patent by an American organ company for any musical instrument which used a digitally stored waveform. The sample rate of this digital calculation process was very high - 567 kHz if I remember correctly. This enabled it to produce really bright sounds with these sharp-edged waveforms with almost no aliasing problems. The sounds were added together internally and the result is available as a 14 bit binary number on 14 pins of the chip. The task then is to convert this to voltages *without* creating capacitive or other glitches when the various most significant bits change place. The resulting sound was low-pass filtered, usually with two possible cut off frequencies. Note this is a simple low-pass filter, with a fixed frequency, on the entire output - all 8 notes. The various instruments used different versions of the same chip - differing only in the sounds which were built into the chip. The dual chip keyboards used two running in synch, both seeing the keyboard together. Each chip controlled its own LPF and the result was mixed to mono. All these keyboards had a single speaker in them. The dual chip keyboards achieved a more interesting tone by mixing two sounds from the two chips. There is no possibility of detuning, since they must run in synch, but there is a slight detune function available in each chip, so one CT-202 mod was to detune one chip in this way, although not all 49 pitches were affected. Stereo outputs and fading between the filtered and unfiltered sounds were other mods in addition to those listed for the M-10 below. The chips have four "memory" positions, and into each position, you can "write" a sound from the set which is built into the chip. This was done by closing one control switch and pressing a particular keyboard key. The M-10, MT-30, MT31 and MT-40 did not have such a switch, so I added one. I also improved the DAC circuitry, bypassed the filtering, added switches for two vibrato speeds and for softer vibrato (I received reports of women being *most* intrigued by the hard/soft and fast/slow vibrato concepts!), an octave drop, hold for all notes or hold until a new key is pressed, slow decay at the end of the notes and a clock divide-by-two slow-down octave drop. My initial modification bulletin was for the M-10 and there was a second one for the MT-30. The M-10 chip was the same as in the CT-301, so it has 23 or so sounds in it, not just the four which are initially set into its four voice switch positions. Then I rewrote them and had a booklet properly printed covering both of them. This features the M-10 and MT-30 sitting in a paddock with cows in the background. I sent the first bulletin to "Polyphony", the magazine edited by Craig Anderton which later transformed itself into Electronic Musician. This was to publicise the modifications - since I was selling the bulletins for a few dollars. Many months later, a payment check arrived - without explanation. Many more months later I found they had reprinted a subset of my M-10 modifications, with no technical explanation at all, in Polyphony! The best part was the colour front cover with a fine young woman in a field of green and flowers, in picnic mode, with various portable instruments, including a modified M-10. I sold about 450 copies of the bulletin - so there must be lots of modified machines out there! I heard of one chap in the US West Coast (who perhaps was associated with Serge Modular) who modified a large number of them - a hundred or more. Devo is believed to have a modified MT-30 - it shows up on the Devo Ezy Listening Muzak cassette very distinctly. Perhaps the opening bass notes of Michael Jackson's Thriller use a particular MT-30 note with slow-down octave drop and no filter. One day, I suppose I should scan the brochures, service manuals and put up my modification diagrams and text on the Web, but I can't promise when. All this was before MIDI. The VL-1 and VL-5 came later. Casio also made some home instruments (mini and normal sized) based on a chip which did 8 harmonic additive synthesis. This is the MT-70 series. The sounds are nice, but unexciting to my ears. Then there was an MT-65 - a great 4 octave mini-keyboard with an identical MT-401 full-size equivalent. This uses a separate CPU to control a special sound chip. That sound chip uses the same stair-step waveforms as the CT-201 etc. series, but the chip is entirely programmable from a 4 bit data bus. I wrote some C code in about 1984 to program its waveform and play notes. Again, it uses a very high sampling rate and makes extreme demands of the DAC - which was done with CMOS inverters and an R-2R resistor array. I figured out some low-glitch DAC approaches using 74HC174 chips to latch the bits and drive the resistor array directly. The timing and slew rates of the most significant bits is critical, since this can produce large spikes if they are not entirely symmetrical for both positive and negative transitions and synchronised exactly with all other transitions. After that, Casio seemed to produce keyboards at about one a month for several years - just for the hell of it. Most of them were crappy things without the lively sounds of the original series. I lost track of it, apart from taking an interest in the VL-1 and VL-5. Their mini battery operated SK-1 sampler was a first - but it was fatally flawed by creating a thump at the start of each note. It seems the envelope was done digitally, but they stuffed up the mathematics of it inside the chip. Consider the DAC, for instance, as a 0 to 255 converter. What should happen is that 0 volume or 0 signal should produce an output of 128. Then as the volume increases with a signal which is going positive or negative, then the output would go above and below 128. However, their envelope system causes the DAC to go to 0 when there is no volume. So even with a quiet sound, there is a huge thump at the start of each note as the DAC goes from 0 to 128 as the envelope ramps up. Dumb!! - Robin =============================================================== Robin Whittle rw@firstpr.com.au http://www.firstpr.com.au Heidelberg Heights, Melbourne, Australia First Principles Research and expression: Consulting and technical writing. Music. Internet music marketing. Telecommunications. Consumer advocacy in telecommunications, especially privacy. M-F relationships. Kinetic sculpture. Real World Electronics and software for music including: Interfaces Devil Fish mods for the TB-303, Akai sampler memory and Csound synthesis software. ===============================================================