Sound On Sound reader Kofi Busia recently released his own independent LP called 'Oh Africa', funded entirely by himself. To encourage others to follow his shining example, we asked Kofi to write about his experiences. In this the second of three parts, he explains how he came to terms with computers, DX7 programming, and MIDI.

Last month, if you recall, we left me having just received a pile of cardboard boxes. I had, however, done one further bit of research before this stage - choosing music software. I found this, as well as understanding MIDI, and learning synthesizer programming, a long and frustrating task.

My first piece of advice is: don't be afraid of moving sliders and pushing buttons in order to learn about equipment. It is the best way to learn, and the worst that can possibly happen is having to switch a machine off and on again because it has 'crashed'. Whatever was in memory can always be reloaded. Only taking an axe to it - generally - can do permanent damage.

My first problem was understanding MIDI. Most articles and books on the subject were way above my head. I could not understand 'MIDI protocol', 'status bytes' etc. It is getting increasingly difficult for newcomers to catch up. But one day, I had a blinding flash of insight, and it all suddenly made sense. I am sure a seasoned programmer would be appalled at my simple explanation but it works for me Maybe it will work for you too. Here goes...

Computers and processors are, actually, very stupid and can only understand one thing - whether a 'hole' is empty or full. Their usefulness comes from the fact that they can make such assessments at incredible speeds (eight million times per second for some home computers). They can also perform specific tasks as a result of such assessments. I am, at this moment of writing, using a word processing program on my computer.

When the number '84' chases round my hardware/software/interface system, it responds by putting the letter T on the computer screen, and/or arranging to print it out on a piece of paper, and/or storing it on disk. '84' written in a manner the computer understands is 01010100. Each 'hole' represents either zero or a number twice as large as its right-hand neighbour, ie. the rightmost 'hole' is either 0 or 1, the next 0 or 2, the next 0 or 4, next 0 or 8, etc. (01010100 = 0+0+4+0+16+0+64+0=84.) It is the same thing only written in a different form. 01010100 is called a byte, and each of the eight 'holes' (0 or 1) is called a bit. My interface/hardware/software combination interprets this as a letter T because there is an internationally agreed standard (known as ASCII).

If I change to a musical hardware/software/interface set-up and send the same number flashing around, I get a very different result. I am asking for a totally different kind of reaction to the same number. For example, the numbers 148, 64, 64 tell my Korg expander to play the note E above middle C, mezzo-forte.

I was very disappointed on investigating what musical software was available at that time in the UK. I would have gladly changed my computer (I had an IBM compatible I used in my business, remember), but nothing I saw made this even worth contemplating. Things have changed dramatically since then, though. An American music magazine I happened to see was a complete revelation. The UK software market seemed ten years behind! I was due to make one last business trip to the USA, and decided to investigate what was available out there...

It was mind-boggling to play a synth and see what you had done immediately displayed on-screen. Some programs even displayed things in musical notation. But - to be then able to manipulate and rearrange! Want a different key? No problem. Add another phrase a third higher? No problem. Different preset? No problem. And to be able to store on disk, retrieve, evaluate, add, modify... It was all a really exciting prospect, which I found riveting. A whole field of creativity opened up.

After much experimentation, I decided on a piece of software called Sequencer Plus. It had a friendly layout indeed. I particularly liked its full and informative Help screens which were always only one keystroke away. The clear, thorough, and masterfully written manual was important to me since I would be taking the software thousands of miles away, back to England. It showed endless possibilities. I have been very happy with it, and have absolutely no qualms about recommending it. Choosing software (as instruments), is very personal. Have a clear idea what you want to do, be patient and take your time. Check around, as there are many different good features and products.

Back in England I practiced, learned my way round the software, and learned to programme my newly-acquired DX7. (Its manual is surely the most appalling and uninformative one ever foisted on an owner!) I took a few of its preset sounds as a starting point and wrote down every digit and parameter I could find. I tweaked and manipulated every variable in existence. Gradually, I learned from the results - and my ears.

One day, quite by accident, I stumbled on what sounded suspiciously like a sine wave! Soon I had two playing side-by-side, and was able to vary their pitches, volumes, etc. Hindsight tells me that I was learning about 'operators' and 'carriers'. I realised that only some operators could be heard directly, therefore the ones that could not be heard were somehow affecting the carriers. A few shoves on the data slider made that clear! Enter 'modulators', stage left.

Having started as a complete novice in programming, I actually now find the DX7 easier to programme. You can start off with a completely clean slate (sine wave) and just build up. If you are a total novice, both forms of programming are initially completely baffling. For someone more experienced, it is easy to see why the difference of approach can be infuriating. But take heart. It is possible to programme sounds from scratch, and it is not nearly as fearsome as has been made out.

A while after the cardboard boxes had arrived at my home, I was all hooked up and ready to roll. There was so much to learn that it was important to divide up my time constructively. I vowed never to let writing/arranging/playing/practicing/producing/engineering, etc, get mixed up. I kept my eyes and ears open, and made mental or written notes, but I always made a point of only working on one element of my music at a time.

I was in Seventh Heaven! I could not believe how many possibilities the instruments I had - and most especially the computer and software combination - gave me. My only real problem was coming to terms with the continuing march of technology. A couple of great drum machines were released which would have suited my purposes excellently, and were also cheaper than what I had. Samplers, too, improved dramatically and came down in price. I salivated in shops and wondered how I was going to manage without them. I decided there was no point in lusting after things I could no longer afford; if I could not make music with what equipment I had, the fault was definitely not in my instruments! Besides, if I wrote a hit album, I would be able to afford anything!

The KX88 master keyboard and computer were invaluable. Different controls could act in different ways for different songs, and be recalled at the touch of a button. The breath controller could do pitch-bending while the foot controller introduced brightness and the mod wheel controlled volume. Other times a different arrangement was better. I could switch the computer on with the germ of an idea and just play away until I had a few solid ideas memorised. I could then add, delete, speed up, slow down, move notes around, lengthen them, shorten them, or transpose them. I could add bender info, edit the modulation wheel, change time signatures, try different patches - all without losing a single moment of time, every step clearly visible on screen at the touch of a key. What I liked, I could then store on disk.

Gradually my songs began to take shape and I soon had a whole bundle which I narrowed down to a 12-song suite. The computer then laid down a sync code on the Fostex 8-track. Using this, I sequenced the first instrument. Then I wound the tape back and laid down the next part in perfect synchrony. If something was not quite right I edited the musical data in the computer and tried again. I then concentrated on the mixing desk and effects. Before I knew where I was, I could bandy 'dBs' and 'tape returns' around with the best of them! As I learned about echo, reverb, chorusing, etc, I began to appreciate what really good signal processors might be able to do for me.

Eventually, I had a whole tape full of songs. I chose three and sent my amateurish demo around a few places to see what sort of a reaction I would get. Nobody offered me a contract. Surprise, surprise!

Being aware of my equipment's limitations, I decided to go to a 'proper' studio to make better quality tapes. I asked around, casting the net wider and wider. Eventually, someone in my wife's office came up with the name of someone who "worked in the technical side" of the music biz. He was called Oliver Hitch and sounded like a fairly OK bloke on the phone. He was an engineer. I asked a few questions and discovered that he had engineered a Billy Bragg album which had "gone gold". Now Billy Bragg I had heard of! Anyone good enough for him was OK by me! I arranged to go up to see the studio where he now worked (Keyboard Studios). The owner was called Winston Sela. It was a friendly place, and we all got on just fine. It did not have £500,000 of equipment, but all the standard sort of stuff seemed to be to hand. Its sound quality was acceptable, and so was the price. We shook hands and I booked some session time.

I figured I could finish everything in three days, which they smiled at and treated like a joke. "Have you done this kind of thing before?" they asked. Both Oliver and Winston assured me that three weeks would be more likely! But I had been practicing hard and had got a game plan. Everything had been timed and organised. The only thing over which I could get no prior control were my vocals. As long as the studio had electricity and I could plug in my trusty Canon computer, I was sure it could be done. Oliver and Winston continued to give each other 'he'll soon learn' looks.

So I turned up at the studio. Oliver, the engineer, worked really hard with me and we did it! Day 1 - instrumentals; Day 2 - lead vocals and all backings except one; Day 3 - remaining backing vocal plus mixing.

During the mixing I was pretty clear how I wanted the songs to sound, but always left Oliver to set basic levels as he saw fit. I was quite happy to rely on his experience; also, I was paying him! Then we gradually shaped the tracks how I wanted them. And so - all 12 songs were mixed in one day! Even he was impressed.

I now had a good quality master tape. Back at home, I hired a Revox for editing and ran off a whole load of cassette copies. This time I was really ready. My material would get straight to an influential record company exec, its brilliance would shine through, there would be screeching cars and blazing tyres as they tore to my home, and they would offer me a contract on their knees...

Life, though, is not like that, is it? I was again disillusioned. But I had sold my house and sacrificed my livelihood to get this far, and I was not about to give up so easily. Time for another plan...

Next month, I'll tell you how I decided that it was both worthwhile and possible to make my own way into this crazy, frustrating and sometimes depressing jungle.

The end result of Kofi Busia's efforts is an unusual album entitled 'Oh Africa', released by African Records International ((Contact Details)) and distributed by EMI/Jet Star. (Contact Details).