Defender of the Crown: The Cinematic Bluff

The story of Defender of the Crown is a story about a shop window.

In 1986 Commodore had a machine that nobody could explain. The Amiga did things the salesman could not describe to a person who had only ever seen a Spectrum — a colour palette that made no sense, hardware that moved sprites without asking the processor, sound that had actual instruments in it. You could recite the specification at a customer for ten minutes and lose them at the word “coprocessor”.

Or you could put Defender of the Crown on the screen and let them look at it.

Cinemaware’s first release did more to sell that computer than any advert Commodore ever paid for, and it did it in about four seconds per passer-by. That is a real achievement and I want to give it its full weight before I take the game apart, because taking the game apart is easy and has been done for thirty-five years by people who missed what was actually invented here.

What is actually in the box

The frame is a fictionalised England with the king dead and the Saxons and Normans carving the place up. You pick one of four Saxon knights — Wilfred of Ivanhoe and company — each with a rating in leadership and a rating in jousting, which between them are most of the numbers in the game.

Then you play the campaign, which works like this. England is a handful of territories. Territories produce gold. Gold buys soldiers, knights and catapults. You march an army at a neighbouring territory. The battle resolves as a brief animated clash whose outcome is driven overwhelmingly by how many men you brought multiplied by your leadership rating. You take the territory. It produces gold. Repeat until you hold England.

Around that spine sit the set pieces. A jousting tournament, where you win land by hitting a man with a stick. A catapult siege, where you lob rocks at a castle wall on a trajectory arc. A raid, where you creep into a rival’s castle and fight a swordsman for his gold. A rescue, where you fight your way to a captive noblewoman. And Robin Hood, who occasionally turns up and does a job for you for nothing.

That is the game. Read it back cold and you have a territory-grab with four minigames and combat resolved by multiplication. There is no supply, no attrition worth the name, no diplomacy, no terrain, no fog of war, no economy beyond a single number going up. The strategic layer has roughly the depth of a pub quiz round about medieval England.

The catapult section is genuinely decent — an honest little ballistics problem with a satisfying arc. The joust is fine. The sword fights are bad, in the specific way where you find the one move that wins and then use it forever, and the game never notices.

The C64 conversion is the control experiment

Here is where the argument gets settled, and it gets settled by an accident of publishing history.

The Amiga version came out in 1986. The 8-bit conversions followed in 1987, and I met it on a C64, which is the version most British teenagers actually had in their hands. Same rules. Same territories. Same leadership multiplication. Same four knights, same catapult, same Robin Hood.

It is a nothing of a game.

That is the control experiment, run for free by the market. Take Defender of the Crown, remove Jim Sachs’s artwork, keep every line of design intact, and what remains is something nobody has ever written a fond retrospective about. The C64 version has no defenders. The conversion is competent — it is a faithful port of the thing that was designed. The thing that was designed was simply never the point.

Sachs is the reason the Amiga original exists in the culture at all. He was a self-taught artist who understood, earlier than nearly anyone, that the Amiga’s palette could be used for painting instead of for colour. The result was screens that looked like book illustration on a home computer at a moment when home computers looked like teletext. Put those images in a shop window and you sell hardware.

So the honest description of the Amiga version is: a slideshow of extraordinary paintings, with a mediocre board game between them to justify the price.

What the paintings are actually doing

Look closely at how the art is deployed and there is craft in the placement that the design never matches.

The screens arrive at the exact moments when the underlying system has nothing to say. You commit an army — a decision with no texture, since it is a number against a number — and the game answers with a painted plain and men in ranks. You raid a castle, and before the sword-fight starts you get a night scene with a wall and a torch. The art is patching the gaps in the mechanics, and it is timed to patch them so precisely that you never notice a gap exists.

The other trick is the frame. Cinemaware built the whole thing to look like a film — the title sequence, the scene transitions, the way an event announces itself with a picture and a caption. The C64 had been running strategy games for years, and every one of them told you what happened with a line of text. Defender of the Crown told you what happened with an image, and an image does not invite you to think about the rules that produced it. Text says “you won, 400 to 250”. A painting says “your men are on that field”. One of those is auditable and one is not.

That is what makes the word bluff the right one. The presentation actively discourages the kind of attention that would find the design out. It works for about six hours, which in 1986 was a full-price game’s worth.

The bluff was a real invention

Now the part where I stop being clever about it.

That structure — a thin spine of decisions, punctuated by expensive set pieces, arranged so the set pieces carry the emotional load — was a genuinely new thing in 1986, and it is the dominant architecture of the mainstream games industry today.

Cinemaware called their thesis “interactive movies” and got laughed at for a decade, mostly because the FMV era took the label and produced landfill with it. The laughing missed that Bob Jacob’s outfit had noticed something true: presentation is a system in its own right, doing work the mechanics cannot do. The reason you care about a territory in Defender of the Crown is that a beautiful painting told you it was worth caring about. Strip the painting and the territory is a number. The art is generating the motivation that the design has no way of generating on its own.

Every blockbuster since has run this play. The set-piece pacing of a modern cinematic action game is Defender of the Crown’s pacing with a nine-figure budget: a corridor of light decisions between showpieces, where the showpieces are where the money and the meaning live. Cinemaware got there first, with 512K and one painter.

The criticism that fairly lands is that Cinemaware never went back and put a real strategy game underneath. Thinness on its own is forgivable; leaving it thin for four years was a decision. Kellyn Beck’s design was sufficient for the trick and nothing more, and the company spent the next four years repeating the trick — It Came from the Desert is the same architecture pointed at a fifties B-movie, and it is a better game because the frame gave it a real structure of investigation to hang the set pieces on. When Cinemaware’s design caught up with its art, they made something that holds. Defender of the Crown is the version where the art arrived alone and won anyway.

And it did win. There is a companion piece to this on the shelf: Shadow of the Beast, which pulled precisely the same con three years later with parallax instead of paint, and is remembered with precisely the same mix of awe and embarrassment. The Amiga’s two most famous games are both demos that got a price tag. That says something about the machine, and something less flattering about all of us who bought one.

What the design should have been

The frustrating part is how close it sits to a real game.

Give the territories terrain that affects a battle. Give armies supply, so a long campaign costs something. Give the four Norman lords personalities and grudges, so the map has politics. Give leadership a downside. Any one of those turns the spine into a system with decisions in it, and the set pieces would then be punctuating something instead of carrying it.

Populous shipped two years later, on the same machine, with a fraction of the art budget and a landscape that was itself the game. Pirates! had appeared the year before Defender of the Crown on a machine with a tenth of the power, and had six real systems and an ending that meant something. The tools were there. Cinemaware were solving a different problem, and they solved it completely.

Where to play it

The Amiga version is the only one worth your time, and it is preserved and emulated well. Come to it as an art object with a game attached and you will have a lovely ninety minutes. Come to it as a strategy game and you will have finished England before you have finished a cup of tea.

The 8-bit conversions are historically interesting and nothing else. There is a later NES version with some added content, and a 2003 remake that added systems nobody wanted and lost the paintings, which is about the most complete way to misunderstand a game that I can think of.

Sachs did the thing the Amiga was for, before anybody had worked out what the Amiga was for. Four decades on, the pictures still work and the game still does not, and the industry chose the pictures. It was paying attention.