Back in 1978 computers in their very basic form were beginning to appear, mainly supplied by the microprocessor manufacturers as test/evaluation boards for their CPUs.

A "SYM 1" made by Synertek Systems one of the suppliers of the 6502 8 bit CPU was the first purchase. The CPU was clocked at 1MHz and had 1K of on board RAM, expandable to 4K should finances permit, the operating system - then referred to as a monitor program - resided in a 4K ROM.

The user interface was an extended hex keypad with associated 6 x 7 segment displays, power was via a bench PSU - not supplied. Programs were entered in machine code e.g. A 9 0 0 would represent load accumulator A with 0x00.

This very simple "computer" was pressed into service with my first VTR program.

The story was that a TV game show was being recorded in a studio but unfortunately the digital effects device used to zoom the scores in/out of frame was not working, so the recording was made without it.

The problem then was how can the digital effects be reconstructed and edited in at the correct times, well the many digital I/O lines of the SYM 1 were used and attached to the parallel output of a timecode reader. A program was written to compare the values read in with a list of prepared timecode values embedded in the program. When a match was detected an output pin was pulsed to trigger the digital effects.

It was very satisfying to watch the digital effects added on to the half hour edited program all without manual intervention.

Next progression was to a BBC Microcomputer Model B with 32K of RAM this had a full sized keyboard, storage options were initially cassette tape, then later 40 and 80 track 5.25" floppy discs.

Still continuing with the TV theme, a TV Sync Pulse Generator was designed (it provides all the timing pulses to keep cameras and other equipment in synchronisation). One of the requirements is that the output pulses are of a well defined shape and this is commonly achieved by using analogue filters built of discrete inductors and capacitors.

Analysing such filters is a complex mathematical exercise just the sort of job a microcomputer should be good at - if there was a program to perform the task.

Hence the "AC Linear Circuit Analysis" program (such a catchy title - well it told you exactly what it did...), this analysed circuits and produced gain/frequency plots. This then showed how existing filters performed but designing them was still another matter.

Having now a taste for filter design could only mean one thing - a Filter Design program, this calculated the coefficients and component values for Elliptical filters. Previously this data was only available from books of tables, these were the sort of books that cost a couple of weeks wages.

It was interesting to see how many errors there were in the published tables mainly due to columns being swapped over, but then is was not easy to solve 11th order polynomial equations, making the proof readers job rather difficult.

These were all analogue filters but digital filters were soon to come along, however this was a bit much for the BBC Microcomputer, so one of these new-fangled PC compatible computers was purchased. It boasted unbelievable specifications - 80286 CPU, 640K RAM, high resolution colour screen (640x480), 10MB hard disc and an 8MHz clock speed with a Turbo switch taking it all the way up to 12MHz.

The DOS OS was all you could have wanted and it allowed a Digital Filter design program to be written, culminating in 400 pole design (for those that are interested the cut off slope is 3920dB/decade). The filter was duly constructed and performed to the design parameters.

A lot of people were happy writing and using DOS text based programs remember WordStar, but there were other offerings with a new Graphical User Interface - Windows. This did make the existing programs rather unattractive in comparison, they were still streets faster but that didn't seem to matter.

So the transition to Windows 3.11 was made and VTR control and monitoring programs took over from heavy mathematics.