Sensible Software: The Small Sprite Doctrine

There’s a fight in game design that gets argued out on a screen every time someone decides how big a person should be. Make the character large and you can animate his face, sell his weight, give him a personality the player reads without a line of dialogue. Make him small and you can show what’s around him.

Sensible Software picked small, over and over, for fifteen years, and the consistency is what makes them worth writing about. Jon Hare and Chris Yates had a thesis: in a game where the interesting thing is the relationship between actors, the actors should be small enough that you can see all of them at once.

Wizball, and a studio finding its hands

They formed in 1986, two teenagers out of Essex. Parallax came first that year, and Wizball landed in 1987 on the C64.

Wizball is the strangest good game on that machine’s shooter shelf and I’ve written about it at length. The premise: a world has lost its colour, you’re a bouncing ball, and you shoot your way through drab levels collecting droplets of pigment to restore red, green and blue — mixing them in a cauldron at the side of the screen. Your ship can’t be steered so much as persuaded; it rolls, it has momentum, and taming it is the first skill the game asks for.

The design tell here is already visible. Wizball’s real subject is the relationship between your ball, your cat and the paint you’re collecting. The cat — Nifta — is a second entity you can control or leave to the AI or hand to a second player, and the game is fundamentally about coordinating two things that move differently. Small sprites, wide view, two actors, one problem. That’s the whole studio in 1987.

They also shipped the Shoot-Em-Up Construction Kit in 1987, which handed the tools to everyone else and produced a decade of covertape-quality shooters. As a business decision it’s odd. As a statement about what they thought design was — a set of legible rules anyone could recombine — it’s completely on brand.

The doctrine, stated plainly

MicroProse Soccer arrived in 1988, and Sensible Soccer in 1992, and between them the thesis gets its cleanest expression.

A footballer in Sensible Soccer is a handful of pixels with a two-frame run cycle. He has no face. He has no kit detail beyond a colour. Zoom out that far and something happens that no amount of animation budget can buy: you can see the shape of the play. You can see the overlapping run, the gap opening at the back post, the defender committing early. The game gives you the information a manager on a touchline has, and then asks you to act on it at the speed a player does.

The ball is the other half. Sensible Soccer’s after-touch — curving the ball after you’ve struck it by holding a direction — is the mechanic everyone remembers, and it works because of the zoom. You can see the whole flight of the ball, so you can shape the whole flight of the ball. A close camera would make after-touch a guess.

I’ve argued that 16-bit sports games were better systems than their descendants, and Sensible Soccer is exhibit A. The modern football game has photorealistic faces and a camera pinned to the ball, and the consequence is that you’re playing a series of local decisions with no view of the structure. FIFA can render a blade of grass. It cannot show you a pitch.

Sensible World of Soccer (1994) then did the other thing Sensible did, which was to bury an absurd amount of data under a game that looked like a toy: something like 1,500 clubs and 27,000 players, and a career mode that let you manage for twenty seasons. I’ve covered how that combination works as a system. The relevant career point is the pattern — a simple, legible surface with real depth beneath it, which is the same trick as Speedball 2 and a very different trick from a game that looks deep and plays shallow.

Mega-lo-Mania, and the doctrine at strategy scale

Mega-lo-Mania (1991) is where the small sprite goes furthest. You’re a god allocating men to sectors on a map across epochs, from rocks to nuclear weapons, and your people are dots that mine, build and fight while you watch from far enough away to see all four rival gods at once.

The system underneath is a tech tree tied to attrition — you research by assigning men to a lab, and men in a lab aren’t mining or fighting. Every advance costs presence somewhere else. That’s a genuinely tight economic loop, and it’s readable only because you can see the entire island at a glance.

It arrived two years after Populous had opened the god game and it’s the more mechanical of the two — Populous is about terrain, Mega-lo-Mania is about allocation. The lineage runs forward into the real-time strategy games that Dune II would codify the following year, and Mega-lo-Mania gets there by a different road: from the top down, from a god’s chair, with the population as a resource you spend.

Cannon Fodder, and the doctrine turning on you

Cannon Fodder (1993) is the one that matters most, and it’s the one where Sensible discovered what their own design philosophy was capable of.

The mechanics are Sensible Soccer’s, redirected. Tiny men. Wide view. Mouse-driven movement and a shoot button. You control a squad of up to four, you split them, you flank, you clear a map. It plays beautifully — the squad-splitting is genuinely tactical and the maps are built to reward it.

And then the small sprite does something no large one could. Your men die instantly. One bullet. There’s no health bar, no death animation worth the name, just a soldier stopping. Between missions you get a hill with gravestones on it, one per man lost, filling up over the campaign. Recruits queue at the bottom waiting to be promoted into the gaps. The title screen has a song — Hare and Richard Joseph’s “War Has Never Been So Much Fun” — sung cheerfully over the whole thing.

The zoom is what makes the argument land. You are far enough away that a man’s death is a sprite blinking out, which is exactly the distance a general has, and the gravestone hill then insists you count. The game gives you the callousness and then hands you the ledger. That’s a real piece of formal argument, achieved with pixels the size of a fingernail clipping, and I’ve made the fuller case elsewhere.

The tabloid row that followed — the poppy on the cover, the Royal British Legion’s objection, the Daily Star front page — is the part that gets retold, and it’s the least interesting thing about the game. Cannon Fodder’s anti-war content is in the systems. Anyone who played past the third mission knew that.

The long decline

After Cannon Fodder the studio’s story gets ordinary, which is to say commercial. Sensible Golf (1994) applied the doctrine to a sport that doesn’t need a wide view and demonstrated the limits of the idea. Sex, Drugs and Rock ’n’ Roll (1995) was a band management sim that landed badly. Cannon Fodder 2 was more of the first with less of the point.

Codemasters bought Sensible in 1999. Hare has spent the years since talking, teaching and occasionally trying to get a Sensible Soccer successor funded, and the honest position is that nobody has replaced Sensible World of Soccer’s career mode in thirty years despite several people trying, including Hare.

The reason is structural. The doctrine requires a publisher willing to ship a game that looks like it cost nothing. Sensible’s games were always going to lose the shelf to something with bigger sprites, and by the mid-90s the shelf was all that mattered — the same squeeze that ended the budget label’s democratic little economy. A studio whose entire competitive advantage is legibility has nothing to show in a magazine screenshot.

What they were right about

Zoom is a design parameter and almost nobody treats it as one. It decides what the player can know, and what the player can know decides what decisions are available. Sensible worked that out in 1987 and spent fifteen years being right about it while the industry ran the other way.

The evidence is that their games remain playable and their contemporaries’ mostly don’t. Sensible Soccer’s after-touch still feels like a skill you can improve at. Cannon Fodder’s squad-splitting is still a real tactical vocabulary. The sprites were the instrument all along.

Where to start

Sensible World of Soccer, Amiga, and commit to a career. The first three seasons are the tutorial.

Then Cannon Fodder, mouse in hand, and go far enough to watch the hill fill up.

Wizball if you want to see two teenagers being cleverer than the market they were selling into. It’s the hardest of the three to love and the easiest to admire.