The inspiration for my new game, A Perfectly Normal House!, came on the Monday before Halloween, when I was alerted to a buck-wild 3D real estate walkthrough that was making the rounds across the more online parts of Twitter. What struck me is that it was scary, not because it put together a horrifying narrative, but precisely because of the opposite – it was the surreal nature of the house from the eBay DVD warehouse adjoining a baptismal font to the bizarre architectural layout. I’m very fond of liminal spaces and areas that feel like they don’t belong without being overtly supernatural, so I knew I wanted to create a game that captured the experience of exploring this house.
Then I realized, hey, Halloween’s coming up! Since this is a horror game, I should release it on Halloween! This sound line of thinking had one flaw: I had to make an entire game, start to finish, in 5 days. Not being the kind of person who usually does things like game jams, this was a way tighter deadline than anything else I had made before. This is the story behind the 5 days it took to build A Perfectly Normal House!
When creating a top-down design – that is, a design inspired from a genre or real-life concept – the main goal is to capture the “feel” of something, whether it’s Camp Grizzly‘s tribute to sleepaway camp slasher films or Twilight Struggle‘s recreation of the Cold War. In this case, what I wanted to capture was the sense of shock and dread you discovered when looking at the rooms of this house that shouldn’t exist.
My first pass on APNH! was pretty close to the final version, where you randomly generate rooms with things that could make you lose composure points. I used “composure” instead of the more popular “sanity” because the concept of “losing sanity” is harmful to a lot of people, and everyone has times where they’ve lost their composure, so it was a more relatable experience. Originally, I had a press-your-luck element, where you could leave the house at any time and calculate its value based on the rooms you discovered. I didn’t even need to playtest this without realizing the issue: The player was incentivized to not experience the weirdest parts of the game, which was where all the enjoyment was.
The solution I came up with in the time I had was to make the game fully deterministic, and have the player’s only function to be rolling for room contents and arranging them as they see fit. This flies against common game design knowledge, but I ultimately went with it because APNH! is an experiential game, and the golden rule of good experiential design is “create the right experience.” This way, I could force the player to get really weird and use Composure as a timer to make sure it wouldn’t go on so long it would become boring.
So if the game’s deterministic, what’s the value in requiring the player to go through it at all? I could, with some aid or research, create a weird real estate generator that you could just load all at once. While there were time constraints here, I felt that wouldn’t give the same effect because the one-by-one production of each item in the room created a building sense of…”dread” isn’t the right word, but maybe “bafflement” – that you can’t get with an all-at-once approach.
Wednesday: Creating Lists
I used to think I was good at math until I made a bunch of dice result tables for A Perfectly Normal House!. It turns out that, when I want a result to occur about 5 percent of the time, and it starts from 05, I am very likely to commit a fencepost error and make it go from 05-10. I wish there were an automated tool I could use for something like this – there’s a table generator online but it doesn’t let you input ranges, unfortunately,
Beyond my inability to count, the biggest challenge was just coming up with enough stuff to fill out the tables. Much like party games, which I cover in The Arty of Party, the bulk of the design work on APNH! was finding lots and lots of items that fit the spirit of the story I wanted players to tell through the game. As mentioned, I didn’t want to get overtly supernatural, but I had a few other limits too. I didn’t want to get overtly scatological or sexual, both because that would be too easy and because I want Phantom Knight Games to mostly stick to a PG rating.
I also didn’t want to do anything that crossed the line from “creepy” to “scary” – no dismembered body parts or murder evidence or things like that – because that would turn the narrative into “the pieces of some slasher movie monster slowly come together as you explore the house.” I didn’t want a narrative at all, to be frank – I wanted the game to reflect the absurdities of a house that’s been lived in a little too much. There shouldn’t be any form of plot or progression, only weirdness upon weirdness until your player-avatar can’t take it anymore.
My listmaking was mostly arbitrary, but I went into it with targets for how often you should lose Composure in a given room. I absolutely didn’t want the game to end at the first cuil (level of weirdness, of which there are three), and preferably not the second either. I ultimately settled on having about 15% of the objects at cuil 1 making the player lose composure, followed by 30% at cuil 2 and 50% at cuil 3, when all hell breaks loose.
I normally prefer to playtest more than twice, but I was on a very tight schedule, so twice it would have to be. The first playtest was with an online D&D group made up of some of my college friends, and took place over Google Drawings. One of the members of this group was my friend Christopher, who took it upon himself to reorganize the Google Sheet where I had my list of objects, as well as calculate the average amount of Composure loss per cuil. He did such a service to the game that I gave him a Special Thanks credit. In any event, the playtest went very smoothly and ended with the hapless realtor discovering a room filled with unloaded submachine guns. My main takeaway from here was the starting Composure for the player should be much higher than I already had it, and I settled on 20. I also made it a bit more likely to roll bathrooms and kitchens.
The second playtest was with my dad. I tried to be more hands-off, since this would be my only chance at something approaching a “blind” playtest, and we used pencil and paper this time. I came very close to having to explain what an anime hug pillow is to my dad. THESE ARE THE SACRIFICES I MAKE FOR MY ART. Anyway, the playtest worked great, and had a weird number of bathrooms considering he didn’t roll up that many kitchens. Satisfied, I started on the final part of my project: Laying it out in InDesign.
Friday: Creating the PDF
This was mostly boring behind-the-scenes work, so I’ll spare you the details. InDesign’s ability to carry over text across multiple boxes proved invaluable, as I was constantly tweaking the font size and whatnot. Getting everything to fit on one page with no weird hanging boxes onto the next page was a real challenge, but I was eventually able to get all the content looking at least fairly nice.
I did end up with some white space in cuil 2, so I included a coupon for pizza and a poetry corner. Want to see them? Get a copy of A Perfectly Normal House! and see for yourself!
Saturday and Beyond
On Saturday, I successfully launched A Perfectly Normal House! on itch.io, and got more downloads than I expected for a company as tiny as mine.
On the whole, I think working on APNH! within the deadline I gave myself was a good idea. I haven’t been able to get a lot of work done on my games since quarantine began because it makes playtesting much more cumbersome, so it was nice to be able to shake the rust off and remind myself that I can use my skills to make people happy. Once things go back to normal, I don’t expect I’ll be making a lot more small projects like this, but if the circumstances arise I might do so.
I’ll close this design diary with one more exhortation to get the game – it’s free, and if you’re the kind of person who likes looking at a space and wondering “what the hell?”, you’ll have a great time. Good luck, and happy house hunting!