That first project

So I wrote that blog post that was stopping me doing any game design. I suppose I’d now better actually do some game design.

Actually doing some game design

The goal at the moment is to make something self-contained, simple and achievable, because the right now the most important thing I can do is to actually finish something, produce a complete game rather than an ambitious “demo”. As such there are a couple of easy decisions to make. My “engine” of choice will be Twine, notable user-friendly scripting language for interactive fiction, as the most I will be expected to “code” is markup and some variables. A pure text adventure means I don’t have to produce any art assets or music, which is also handy. In addition, I spent pretty much all of my youth up to my eyeballs in Fighting Fantasy, Choose Your Own Adventure and every similar gamebook. I have pretty good handle on how “if you wish to go right, go to section 29”-style text adventures work structurally.

As to the subject, that’s quite easy as well. Those of you who were here for the previous instalment will remember me mentioning that some of the first games I ever wrote in BASIC were text adventures based on the 1980s-90s children’s gameshow Knightmare. There is a distinct possibility that in the year 2017 (!) people reading this may be unfamiliar with the format, so in short: a lone adventurer was thrust into a fantasy dungeon created by the magic of greenscreen, guided by three friends watching them on a monitor as nothing was actually there they were blindfolded by the magic Helmet of Justice. They had to be steered round perilous traps, deal with riddles, engage with the dungeons’ denizens and gnerally proceed through the levels to find the object of their quest.

Or, more likely, fail and die.

The show was notable for a) its pretty impressive production values – a lot of the show still holds up visually, especially the early seasons where all the dungeon backgrounds were hand-painted, and b) absolutely not fucking around. Winning teams were notoriously rare, with nobody winning at all in the show’s first season. They were quite happy to blow up, chop, electrocute or bludgeon their (ten-year-old) contestants, or send them plummeting into bottomless pits (at least when your advisors didn’t steer you down them of their own accord). The most memorable obstacle at deeper levels was a conveyor belt in a narrow corridor with circular sawblades that would rip down the walls, forcing the normally sedate movement pace to crank up a couple of gears.

A modern viewer is likely to see some video game influence here, and that’s exactly the case. According to this excellent history of the show by creator and producer Tim Child, early 8-bit computer adventures served both to inspire and to help solve a lot of problems they were experiencing in the prototyping stage of the concept, such as basing all the rooms on a standardised grid format to make setting up the studio easier. (The famous on-screen Life Force clock, featuring the dungeoneer’s head where their helmet and eventually the flesh on their face slowly fell away piece by piece, was shamelessly lifted from Atic Atac.)

There are other elements of the show that make it hany for reverse-engineering the setup into a game I can make quickly:

• Simple premise: You are going into a dungeon. Why? To get the treasure. Why? To answer a challenge. That’ll do.
• Linearity of space: There’s no branching to the dungeon; it’s a straight path of rooms to the end. It’s also more or less entirely just that – rooms. You explicitly cannot backtrack; the Dungeon Master’s catchphrase was “The only way is onwards, there is no turning back.” There’s no escaping out of a window, wandering off into the woods or more choice to advance than “go through this door or that door”. Speaking of which…
• Linearity of critical path: There is pretty much only one right way to do things, and deviating from this path usually ends in an entertaningly grisly demise. This means that I can put in a smaller variety of viable options and keep it “in-genre”.
• Segmentation: There was little to no crossover between dungeon levels, as this allowed the scripts for the actors to be modular (no throwing away an entire dungeon design because the team fell in a hole on level 1). Consequences for decisions don’t have to persist more than a few rooms.
• No simulated combat: The player is not expected to fight any threats they encounter, only outrun or outthink them. Actually getting cornered by an armed and angry opponent with no escape method means instant death. This means I don’t have to put in any code to handle combat variables or random dice rolls.
• No “special skills”: The player character is assumed to be more or less the player themseves; a regular person with no access to elite combat training or magical powers. The dungeon has to be faced with your wits and anything you can find.
• Conversation: As the show went on it used more and more NPC encounters, as they broke up the stretches of “three people try to get a guy in a blindfold through a door”. This finally resulted in them getting a proper scenery-chewing master vellain that directed the forces of the Opposition against the Dungeonner. They could spy on his lair with scrying and interact with him before the end, and it made the whole quest seem a lot more coherent than “here is a basement, we left a shield in it somewhere.” The NPCs were all, um, pretty broad strokes, but they were memorable and functional.
• Source material: with eight seasons of Knightmare there’s a lot I can file the serial numbers off draw inspiration from.

The major risk is that a player/reader unfamiliar with the original show will see some of these are not attempts to reflect the conventions of the source but limitations in gameplay, but for my purpose here this is an acceptable risk. This project is meant to be for confidence-building, after all, not a retail release.

(Incidentally, while there were a number of Knightmare computer games, the format definitely works as interactive fiction. Each of the show’s tie-in novels, written by noted game designer Dave Morris, had an adventure gamebook section at the back, and they were frequent library loans for me.)

So what’s your structure?

The dungeons in Knightmare descended across three levels, which (after some early problems with the difficulty curve) settled out to a pretty fair beginner-intermediate-hard state. Level 1 was relatively forgiving (which meant less time introducing new teams, I guess), level 2 was tough and level 3 was genuinely no-takebacks murderous. Notmare Castle (please forgive me, I’ll need to find a better name eventually) will similarly have three levels with this kind of scaling. I want to set things up so that the player can “respawn” at the start of a level should they die; I don’t want to be quite as harsh as the show.

Each level is divided into a number of self-contained rooms, each with its own particular challenge, although consequences (such as knowledge gained or items collected) will carry across to the rest of the level. I’m going to pick five rooms as a good number here as it makes for some easy pacing (allowing a small challenge peak in the middle of each level, then a room’s respite before the final room) without the amount of content getting too intimidating to produce. We stick an antechamber at the start to do a little scene setting before you enter the dungeon proper, and teleport you back there should you win.

I obviously need a visual aid for this, and don’t worry, I’ve invested in the most high-tech setup possible: sticky notes.

It should be fairly obvious how this works: each square on the grid is a room, and I just need to decide what goes in each room. Populating the map is the next goal. So what are we going to have inside it?

There are a number of specific items from the show I would like to file the serial numbers off and use I’ve done this joke already put into this loving homage.

• A wall monster. At some point, usually in the level 1 “clue room”, a face would emerge from the wall, and the team would have to answer its riddles to proceed. Extra correct answers meant that the wall monster would give you suggestions on which of the objects from the clue room you should take with you in your limited carrying space.
• A requirement for a password to get past what in early seasons was a guard but later became a sentient wall (it’s a theme) that ate people, and then eventually an implacable murder robot that promised to tear off one of your legs if you didn’t “give it a word”. I refer you again to them not fucking around.
• A causeway over a pit where you have to walk across the tiles in a predetermined sequence (one you hopefully picked up earlier) or you plunge to your death as it collapses piece by piece.
• Creepy skull ghosts and floating disembodied weapons.
• A short quest-within-a-quest sequence where one of the dungeon’s more powerful inhabitants offers to give you aid for the later levels if you retrieve a particular object for them.
• Incompetent minions of the master villain who can be outwitted because you know in advance about their “secret plan”.
• A giant freaking snake.
• Single-use magic spells that are rewarded for doing good things, the use of which is a moment of drama.
• And of course, the final battle of wits with the master villain to win the prize.

There are likely to be more, and I need to populate 15 rooms and think about some dramatic flow, but now is probably a good idea to end this post for today and have a lie down. I shall scribble on some more sticky notes at work, most likely.