The Battle of Rauk’s Drift

In 1998, I purchased my first PC so I could write up my PhD thesis Developing a Practical Approach to Water Wave Scattering Problems at night while holding down a full-time job during the day. Wow, history really does repeat, except no one is going to call me Dr. Games because I’ve been writing Electron Dance during the night.

It was also my introduction to PC gaming. Of course there was DOOM and Quake yadda yadda yadda but I often neglect to mention a small, quiet executable called Slay (Sean O’Connor, 1995) in my PC history. It arrived on a floppy I had inherited, that’s all I remember. I know I didn’t buy it and certainly didn’t pirate it.

Believe me when I tell you Slay added a few months to the thesis preparation time. Curious as to whether it still had power, I installed a mobile version in April.

Holy shit, it still has power.

What You Need To Know

Slay is a turn-based strategy game. You are one of six kingdoms on an island of hexes. Each kingdom has its own colour and when the game starts, the map resembles a pixelated Jackson Pollack artwork.

The aim of Slay is to assert dominion over the entire island and turn every hex to your colour. Each set of connected hexes of the same colour makes up a territory and will possess a capital; the capital earns one coin for each hex under its control, as long as it doesn’t have a tree growing on it. War is Hell but trees… trees are worse. Uh, I’ll explain this later. I’ll need to neck a few shots first for the Dutch courage.

A capital can spend 15 coins on a castle, a permanent structure that protects adjacent hexes, or 10 coins to press a peasant into military service. A peasant can do one of two things: cut down a tree or take over an enemy hex. The peasant is a weak combatant, though, and cannot take any hexes adjacent to a capital, castle or another peasant. But if you combine two peasants, that makes a piece of higher rank: the spearman. A spearman can defeat enemy peasants and also their capitals.

Blending three peasants together will make the fearsome knight, who can stamp on castles and spearmen like they were nothing. And if you’re having trouble with knights, you can always step up your game by forging four peasants into a dreaded baron.

Okay, you get the idea: earn cash from hexes, expand territories, earn more cash, make stronger pieces. Except there’s wages. Once you’ve conjured up a peasant, you have to pay them 2 coins per turn. And cost rises exponentially with rank: the baron requires an eye watering 54 coins every turn. An alternative way to understand this – a territory with a baron must own 54 hexes just to keep him happy, to pay for his daily wine showers.

What happens if you can’t pay? Ha ha, I’ll tell you what happens. The technical term is a motherfucking disaster. If the capital cannot afford wages then every single piece is abruptly replaced by a gravestone. This can be a game-ending incident if this happens to a large territory. But it also becomes your best offensive tactic: if you keep eroding its income base, eventually you’ll bankrupt an enemy territory and its defences will collapse. Well, apart from its castles.

If you connect up two territories, their capitals will merge creating a unified territory that can churn out peasants more rapidly because of the joint income. This combination of spreading your colour and joining up territories gives Slay that addictive cleaning aspect that underpins the most satisfying games. Linking up embryonic territories after a few turns is like an epipen of dopamine to the skull. But be wary: joining territories can cause all sorts of defensive problems, especially if you rush to connect them up and drag peasants away from watching your borders. That dopamine can all too quickly be replaced with adrenaline. And gravestones. Lots of gravestones.

Cometh the Chaos

The flip side of Slay’s crude addiction loop is how emotionally wounding it can be when an AI starts chewing up your hexes and murdering territories. Seeing a monstrous AI leviathan ravage its enemies at whim can be a sobering experience. It becomes personal. You cultivate a hatred for the enemy AI. I’m not joking. These are very real emotions.

Like that AI who, for some reason, has decided you are the perfect target: I hate you bigtime. And the other AI who should help to mount a united defence but, however, decides it’s a lot easier to pick on the human who is being invaded: you’re getting a shitslice of hate too. Their personalities are identical. Aggressive bullies that exploit weakness without a second thought. If these guys were human, they’d be the kind of people who solve trolley problems with “let the market decide” as the Soylent Green factories reap record profits. You know. CEOs.

Listen up, I’m the biggest hypocrite because my Slay play is just as ruthless. There’s no space for compassion, no option for diplomatic overtures and peace accords. Haters gonna hate. There’s no chance of “good game, sport” over after-match drinks.

That Jackson Pollack opening, though, promises so much. Vandalise me with your colour, the unblank canvas sings. It’s a siren song, though. Watch out for those clusters of AI colours that bode ill. Forget all that exciting potential and just crank up the anxiety because there’s trouble brewing. Turn two is where you have a legit freak out, after the AIs have taken their first baby steps. That’s where the dead end reveals happen. An island of six-sided cliffhanger endings.

Of course, if you’ve been around this block a few times already, you’re Bill Murray in Groundhog Day or Tom Cruise in Edge of Tomorrow or whatever new movie uses the repeating day trope. They just keep making the same movie over and over again. If you’ve seen the same Jackson Pollack a few times, frustration ferments into anger. You’re confident there’s nothing new to be wrung from the opening moves and you’re just performing your best dance steps again. You ignore cries for help from unsalvageable two-hex territories which will be overrun in turn four whatever you do. You’re already dead, fellas. The bullet has been fired.

But even as I reclaimed the Slay throne of mastery there were particular levels where it seemed impossible to win through logic alone. Episodes of mathematical chaos, where a peasant flapping its arms on the Eastern shore would cause an AI hurricane to tear up the West several turns later.

And the worst of these felt unconquerable. It became my mission to defeat them.

I needed to defeat Rauk.

The Problem

Rauk is the final level on the Slay roster, an unpredictable level where AIs can overwhelm you in a couple of turns, but a few odd steps can upset how events are supposed to play out. However, every session seemed doomed to end in disaster. Again and again, the machine won. Rauk wasn’t a battle to prove my credentials as the human who could outsmart the AI – it was a battle for the soul of the game. Was Slay about order or chaos?

Rauk offers the player just five territories. Three of these are helpless, destined to be grist for the AI mill. But in the middle, in the motherfucking middle, you have two territories which can connect up. This is all you’ve got.

The problem was it encouraged my only surviving territory to grow sideways into a thin line which is really dangerous, man. The AI to the north would swell and blister alongside me then perforate my territory before I could put sound defences in place. And if I pushed back against that AI, one of the other AI assholes would get involved instead. Christ.

Let me wheel in AI War (Arcen Games, 2009) here. AI War’s conceit was not to pit the player against a smart AI, but to pit it against an impossibly overpowered AI. One false move and you’d alert it to the threat you posed. It would click its fingers and execute Operation Fuckin’ Wipeout all through your forces.

Slay is similar in some respects. Positioning peasants on your border is the idiot move as the AI cannot resist an easy kill. Unless it’s distracted elsewhere, it will actively send any available meat grinder to mince up the peasant, knowing it is obliterating hard-earned coin. If you can’t defend, then, for the love of dog, don’t give it a reason to attack.

Sound advice, Joel.

No, really. I’m not being sarcastic. That’s sound advice.

It’s just, er, look someone has to say it. If your territory is one-dimensional… then everything is a border. It’s all idiot moves.

Sometimes I’d manage to gouge out a small nugget of territory by growing an extra hex or two every turn, but I never seemed to be moving fast enough. Some trigger-happy AI would slice through my one and only territory like a scalpel across the belly and at that point I would throw my hands up in despair. You can attempt to thread your mini-kingdom back together with stitches if you want, but the other AI smell that sweet hex blood spilling out. It sends them into a feeding frenzy. The truth is, if you let one AI have a pop at you, they’ll all want to get a piece of your action.

Angry Sex

Still, you shouldn’t just quit every time disaster knocks on your door. You may not be able to win but fighting fiercely against ridiculous odds is glorious training. When I took on the level Gunn, I got to a point where I felt I’d lost and one AI had hoovered up the rest of the map.

I tried pushing back against the AI. It was a delicate operation – I had to make sure there were no vulnerabilities on the front-line the AI could penetrate. And despite being the underdog, I put the game into a near-stalemate. I established a strong position with a completely secure rear which faced coastline so could focus entirely on forward movement.

Nerve-wracking, I kept going. A hex here, a hex there, while maintaining a firm defensive line. I kept expecting the AI to summon a baron but it never did; probably because it had too many knights on the payroll. And I kept pushing the AI – hex by hex by hex. Until eventually, I bankrupted it and victory was assured.

Similarly, my Rauk disasters slowly taught me new tactics I hadn’t needed in more forgiving levels. For example, during the opening turns, you should leverage the protection from the capital: it’s like an extra peasant you didn’t pay for and you don’t need to protect the hexes around it straight away.

Sometimes you can go a bit further and throw in a castle right at the start. This can help shore up defences for a good ten turns. The problem was that a castle hobbled my territory’s finances and kept it stagnant until the AI was powerful enough to take down the castle that protected it.

It’s important to recognise strategic stagnation. If your territories are not keeping pace with some of your fearsome rivals, this is often unrecoverable unless you can dig into geography like in Gunn. But I’m like a moth drawn to a flamethrower when it comes to stagnation scenarios. I keep playing them out to see if they’re winnable. More often than not, they aren’t. But sometimes…

…sometimes it comes together.

On May 5, 2022 at 8:57am, I knew I’d done it. I’d not only survived to turn 20 but I was also the biggest kingdom on the island. Sure, I didn’t know how I’d done it, but that wouldn’t be a problem… right?

Next: The conclusion.