The Eurogamer Xmas game!

I saw the games other people were writing on the machine, and wanted to write my own so decided to teach myself to program. Articles in magazines helped, but learning to program the Spectrum mostly involved trial and error.

Back when I first made the decision it was really a choice between the ZX81, the Spectrum and the Vic 20. Everyone I knew had a Spectrum, and although I had heard of the Commodore and newly released Amstrad they hadn't become established at that point. Somehow the Spectrum managed to remain the most popular one in the UK as other machines came and went. Its biggest selling point was probably its simplicity.

My early games were developed on a combination of a rubber-keyed 48K attached to an Opus Discovery disc drive, and a +2. Unfortunately, the disc drive gave up before too long and I was left developing games on the +2 with the tape drive. I don't know how I managed, thinking about it now. The source would be written using LERM's Z80 Toolkit, and the graphics drawn with an art package such as Artist II. Nowadays of course all development is done on a PC which is a lot easier.

Yes, the engine, in-game graphics and sound effects are usually mine. 8-bit programmers were almost always one-man-bands back in the day. In the past

I've had a go at music too, but these days usually ask a dedicated musician like Yerzmyey or Matthew Westcott to do the job for me, as they're far better than I'll ever be at that sort of thing. Sometimes I'll ask for help drawing a loading screen too.

If I'm working with someone else the game is developed first, in isolation.

Once a good portion is done it's sent off to the musician or artist so he can produce something which will suit that particular game.

Very good usually, but I'm paranoid about every new release. I look to do something different with each new game, whether it's a subtle minor improvement on a game style I've seen before or a completely experimental mix of bizarre gameplay elements. Some of my more unusual games have left members of the community wondering if I'm on mind-bending substances at times. Actually, I do drink rather too much homebrew wine but that's another story.

The simplicity of the machine also means that games sink or swim based on how strong their designs are and how good the gameplay is; you can't rely on fancy visuals to get yourself out of a corner and that forces the developer to concentrate on the really important things. Above all it's fun, and I can experiment to my heart's content. Because there's no development budget there are no financial risks involved, and I can fool around with bizarre ideas without worrying about how many units are going to sell. Not even Jeff Minter has that degree of freedom.

When I'm designing something unusual like Quantum Gardening or Loco Bingo it's always approached from the gameplay side, deciding what gameplay mechanics to put together, and then spending a good deal of time trying to work out how to connect them up in a way that will work and be fun to play. By the time this is done the theme will usually have suggested itself, and I'm left at the end with a game that can seldom be described by any sort of meaningful back-story. That's probably why the plots to my games sound so strange.

Cronosoft sells tape versions of my software for those who prefer the genuine article and prefer to play their games on the hardware itself rather than emulating it. They don't just sell software for the Spectrum though; they have games for other 8-bit machines, written by a number of authors. Cronosoft makes no profit as commercially these machines have long been dead, but it's important to keep supporting old hardware.

Typically, it is about 2 months. It can vary though, I've knocked them out in 3-4 weeks before, whereas Egghead 5 took 7 months because I had to design 140 unique and challenging screens.

Finishing it!

Generally, it is the data that takes most time to organise. I can knock out a game engine in almost no time at all as I've built up a library of routines over the past couple of decades. Drawing the graphics and generally designing the layout of levels and that sort of thing isn't as quick, as a game needs a sensible learning curve and plenty of variety as the player progresses to later levels. I like to give the player a reason for continuing, and occasionally leave an undocumented surprise in for the player to discover.

Apart from handheld consoles I don't really "get" modern games. They have great graphics and atmosphere, but don't give quite as much of a gameplay hit as older games. I wouldn't rule out developing something for a newer machine at some point, but it would have to be something of my own design.

My absolute favourite would have to be Halls of the Things - there's no sound and the graphics are terrible but without these the gameplay seems to be enhanced. Target Renegade, Chuckie Egg and Skool Daze are just perfect. Just to be different I'll name Sheepwalk as the fifth. Actually, it's a terrible implementation and plays like a dog but the idea behind it is fantastic and I just love the mischief the player can cause.