The Bitmap Brothers: Style as Substance

Every account of the Bitmap Brothers opens with the photographs, so let’s get them out of the way. Three men in black, arms folded, sunglasses on, lit like a record sleeve. Studio portraits shot the way you’d shoot a band that had just signed to a label. The press ate it, The Face ran them, and the phrase that stuck to Mike Montgomery, Eric Matthews and Steve Kelly for the next thirty-five years was style over substance.

I bought Xenon 2 on the strength of the box when I was fourteen and I would have told you at the time that the sneer was fair. I’ve since come round to thinking the accusation gets the causality backwards. The chrome was doing design work. The way a Bitmap Brothers game looked was downstream of how they’d decided it should feel, and that decision was made earlier and held harder than almost anyone else in British development was managing at the time.

A record label that shipped floppies

They formed in 1987 in Wapping, which is a detail worth keeping because it explains the company they kept. Their early publishing went through Image Works, a Mirrorsoft label. By 1991 they’d set up Renegade Software with Tom Watkins — the pop manager behind the Pet Shop Boys, Bros and Erasure — and Martin Heath of Rhythm King Records.

Read that sentence again. A games developer founded its own publisher in partnership with a pop manager and an independent dance label. That is a structural decision about what kind of thing they thought they were making, and you can hear the consequences directly in the products: Bomb the Bass supplying the Xenon 2 theme, Betty Boo on Magic Pockets, Nation 12 on Gods. Those were Rhythm King artists. The studio had a route to real musicians because it had built one on purpose.

Compare the alternative model. Most British studios of the era treated audio as a job you gave to whoever on staff could drive a tracker, and the results were frequently extraordinary — the C64 scene proves it, and I’ve argued elsewhere that the SID chip made composers into system designers. The Bitmaps went the other way and licensed pop, which is a considerably less romantic answer and produced a distinct house sound: heavier, more produced, closer to a club than a bedroom.

Xenon and the arrival of the surface

Xenon (1988) is a vertically scrolling shooter where your ship transforms into a tank. As shooter design it’s fine. Slightly stiff, honestly, and the transform is more of a gimmick than a system — ground targets need the tank, air targets need the ship, and the decision rarely gets more interesting than that.

What Xenon actually announced was a look. Everything metal. Everything riveted. Backgrounds in near-black so the sprites read hot against them, hard specular highlights on every edge, a font that looked stamped out of plate steel. On an Amiga in 1988, next to the prevailing fashion for airbrushed fantasy and primary colours, it landed like a different industry had shown up.

Xenon 2: Megablast (1989) is the one people remember, and it’s the one with the complication: the Bitmaps designed it and The Assembly Line coded it. It’s slower than Xenon, deliberately — a heavy, dragging pace, an organic bio-mechanical world of teeth and tubes, and a shop between levels where you spend for upgrades. The Bomb the Bass theme runs underneath, an arrangement of a track that had been a genuine chart record.

The shop is where you can see the studio’s actual instinct. Xenon 2’s upgrades are expensive, permanent within a life, and lost on death, which sounds like standard punishment until you notice the game is priced so that you can’t buy everything and has to be planned. It’s a shooter with a build. Gradius had a build too, and I’ve written about how Delta pushed the same idea somewhere stranger by putting the shop on a moving bar. Xenon 2’s version is the least elegant of the three and the most legible, which turns out to be the Bitmap Brothers pattern in miniature.

Gods, and a machine that watches you

Gods (1991) is the studio’s most interesting design and its least played, and I’d guess those two facts are related.

It’s a platform-and-puzzle game with a slow, heavy hero who throws things. The animation has weight — you can see the wind-up, the release, the recovery — and the cost of that weight is that you cannot flick him around the way a Mario or a Sonic lets you flick. That frustrates people. It’s also the entire point. Gods is a game about commitment: you decide, you throw, you live with the throw.

The famous claim about Gods is the adaptive difficulty. The Bitmaps said the game watched how well you were playing and adjusted what it sent at you — a player who was cruising got harder waves, a player struggling got fewer. Their own line at the time was that the AI was there to make the game fair rather than to make it hard, and whatever the implementation actually did under the hood, the design intent is the interesting artefact. This is 1991. Dynamic difficulty adjustment as a marketed feature is at least a decade early, and the mainstream wouldn’t seriously revisit the idea until Resident Evil 4 was quietly doing it in 2005.

Gods also has secrets in it that border on hostile: hidden rooms, hidden shops, treasure that only appears if you do something unmotivated. The game is more rewarding to a player who treats the level as a system to be probed than to one who treats it as a corridor. That is a taste, and it’s the same taste that runs through the immersive sims and their tolerance for players breaking things — a belief that the reward for curiosity should be real and unadvertised.

Speedball 2 and the argument that lands

Then there’s Speedball 2: Brutal Deluxe (1990), which is the studio’s masterpiece and the place the style-over-substance charge dies loudest.

I’ve made the full case for it as a piece of sports design, so the short version: the scoring is what makes it. A goal is worth ten. Injuring an opponent badly enough to stretcher him off is worth points. Bouncing the ball off the star domes is worth points, and lighting them all doubles your score. The result is a sport where the scoreline runs into three figures and where the phrase “I’m forty behind with two minutes left” describes a solvable tactical problem rather than a lost game.

That’s a real piece of systems work. The Bitmaps built an economy where violence, position and time are all convertible into the same currency, and then let you choose your exchange rate on the fly. The chrome and the ICE CREAM sample are the surface. Underneath is a scoring model that most modern sports games would be improved by stealing.

The Chaos Engine, and the run runs out

The Chaos Engine (1993) is the last Bitmap Brothers game that feels like the studio at full strength: a top-down run-and-gun through a Victorian steampunk England broken by a machine, six hireable characters with different weight and speed, and a co-op mode where the second player is a genuine tactical asset. I’ve covered it separately. The relevant career point is that it arrived at the end of the Amiga’s commercial life, sold well, and had nowhere obvious to go next.

What followed is the ordinary shape of a 16-bit studio hitting the 3D transition sideways. The Chaos Engine 2 (1996) turned a co-op game into a competitive one and pleased almost nobody. Z (1996) was a real-time strategy game with FMV cutscenes, aimed at a DOS market the studio hadn’t grown up in, and it’s a decent RTS that arrived after Dune II had already set the template and Command & Conquer had commercialised it. Z: Steel Soldiers followed in 2001. The band photographs stopped making sense the moment the industry stopped being small enough for three men to be famous in it.

What the style was actually for

Here’s the argument. The Bitmap Brothers’ visual identity was a legibility system.

Look at what the chrome buys you mechanically. Near-black backgrounds mean the sprites read instantly — you always know where you are and what will hurt you. Hard highlights on every edge mean object boundaries are unambiguous at speed, which matters enormously in Speedball 2, where nine players and a ball are moving on a pitch that scrolls. The heavy industrial font means UI numbers stay readable while the screen is chaos. Every one of those choices is a playability decision wearing an aesthetic costume.

Contrast this with Psygnosis, the other studio famous for how its products looked, where the art was frequently the best-designed thing in the box and the game behind it could be gorgeous and shallow. The Bitmaps’ surface and their systems were pulling the same direction. That’s the distinction the sneer misses, and it took me about twenty years and a lot of re-loading to see it.

They were also, straightforwardly, good at finishing things. Xenon, Speedball, Gods, Magic Pockets, Chaos Engine — the games ship complete, run well, and don’t have the ragged third-act collapse that so much of the era’s British output has once you get past the covertape demo. Competence is unglamorous and it’s most of what separates a studio with a catalogue from a studio with an anecdote.

Where to start

Speedball 2 on the Amiga, and give it a full season rather than a friendly. The systems don’t open up until you’re managing a squad with injuries.

Then Gods, with the patience it demands — play it as a puzzle game that happens to have a jump button, and the weight stops being an obstacle around the second world.

Xenon 2 last, and mostly for the sound. It’s the weakest design of the three and the purest expression of what the studio was trying to be: a games company that arrived like a record.