Background history

The late 60`s and early 70's were the advent years of the Voltage Controlled Synthesiser and almost every band and solo musician was using one at one time or another. Tangerine Dream and Kraftwerk used little else. The most popular and affordable synths were the VCS 3 and Synthi AKS by EMS, the MiniMoog, the Arp Odyssey, and later the Prophet V by Sequential Circuits, although The Who used the more expensive ARP model, the 2600 (as featured on their hit "Won't get Fooled Again").

I just happened to be lucky to have the chance to play around with these amazing machines, and being a programmer, occasionally even impress others with some original patches.

These synths were so much fun that in '78 I decided to build my own. It was codenamed the "4600" (an older version of the 5600) and sold by Maplin as a DIY kit. It was a full featured synth with ADSR as well as two Transient Generator modules, but the 22x22 pin patchboard, although very versatile, proved hard to use, so it was all rebuilt as the 3800, with a digital keyboard controller and routing via dial switches on each module. The digital keyboard controller allowed me to add a little 128-note sequencer, which I had found in a magazine, and similar to the usual note-sequence entry found on most TB-303 emulators. The end result turned out to be so good that a musician acquaintance offered me a large enough sum to convince me to part with it...

At that time I was working in a Data 100 small mainframe installation, designing and writing various accounting and statistics applications in RPG II. It wasn't a well paid job, nor was the work too demanding for my capabilities, but I was always too much of a dreamer for those glitzy career jobs in the city that pay well but demand a sharp and ruthless attitude. One day the boss rolled in a box on castors - there was a small printer on top and some LEDs and switches on panels on the side. What excitement - I could actually imagine it sitting in a corner of my flat.!!!

"Do you know how to use one of these..?" he wondered.

"Sure - a mini - we need some manuals for it..." I had no exact knowledge on the processor inside that box, nor about what additional capabilities it had, but I had heard it was supposed to be fast. At least by the standards of that time, anyway.

Nothing came from it in the end, but just imagining the speed of the machine started me off thinking about a way to generate sounds via the processor in real-time. Then the Sinclair Spectrum with its Z80 arrived and I went into business as 'microdot software' with a generic relocatable printer driver and screendump utility, followed by a chequebook accounting program. At the same time I started thinking about a navigation system for cars, but as sat-nav wasn't yet commercially available as it is now, positioning of a car on a map was still a problem. The 'realities' of life...

Annie Nightingale was so enthusiastic about this event that it remained engraved at the front of my memory. I was running a basic A2k half/half, OS1.3 and a 20MB peecee harddrive across janus. By now I had some asm practice on the Amiga and I had a 'feel' for the speed of it, so I was getting more confident to at least start, if not finish, the impossible. The Amiga already had an excellent little synth program, known as Aegis Sonix 2, which may have been a bit limited - but if that could do what it did on a basic 68k, on an '060 it should record in real time too...

As I only had Devpac to program with, it was not too easy - coding Intuition gadgets in asm was pretty time consuming. But a start had to be made, especially since I knew that there was soon more speed to be had than even the '060 I was hoping to get one day - the PPC 604e.

Eventually I had one oscillator running with a cycler gadget to select octave, and a slider to tune within that range. It ran a single hardcoded triangular wave sample - but I could at least change frequency fairly smoothly with that slider. Almost, but not yet...

The first version of the chillmachine:

Still, time won't stand still, and I went through a long upgrade spree which eventually led to me getting on-line. SAS C was selling quite cheaply by then, therefore I lost my last excuse of not being able to afford a decent C compiler, as neither Sozobon nor PDC were quite in the same league.

This time it was actually generating sine, triangular, ramp and square waves, the latter even phase adjustable, and the results were sent to either AHI or straight to Paula, via an asm routine. For the GUI I was using gadtools, but since I wasn't happy with sliders for some of the controls, I had to design a round button of my own, which I also decided to convert into a BOOPSI gadget, so it could be useable with BOOPSI or Reaktor.

This program was started many years ago, it isn't 'written to spec', rather it evolves - IOW, I make it up as I go along. Nor am I doing it because I must, or expect to end up rich from it, but because - well, some call it "hack value", - I just found it to be a fun thing to do. Probably once it`s finished, I lose interest and rather listen to music made by others using it.

Now there are some DSP routines running on the Delfina's 56002 as the engine, written by none other than Michael Henke of DelfMPEG fame, and for a while it almost seemed to all drop into place. A real analogue synthesiser for the Amiga - and now, after a 3 year downtime due to my A4000 dying on me, I will have to restart it by getting into DSP coding myself. As well as making it open source. This year, 2011, started well, with an A4000 being offered in csa.marketplace (never was lucky with ebay auctions), just as I had enough saved up for it.

Unfortunately this did not work out as I had hoped.

Click on the picture to see why.