inkle’s Overboard!, a thrilling “whodunnit where you’re the one whodunnit,” received rave reviews upon its recent launch on the Nintendo Switch, PC, and iOS. Narrative Director Jon Ingold joined me via email to discuss the game for The Avocado.
Please tell us about your history in the game industry.
I’ve been making games independently since I was a teenager. I started out making parser-based text adventures in the 1990s, and didn’t get a job in the games industry proper until I was 27, when I started work at Sony’s Cambridge studio as a designer. I work on bunch of cancelled games and one not very good released game, largely doing systems and game balancing, before leaving to co-found inkle with Joseph Humfrey, who I met at the studio.
We’ve been running inkle for the last ten years, and it’s been a lot of fun!
For those unfamiliar with it, could you please give a brief description of Overboard!?
Overboard! is a classic 1930s murder mystery, except that you’re not the detective – you’re the murderer. In the opening sequence of the game, failed actress Veronica Villensey pushes her unlikeable husband overboard. Unfortunately, she’s left a lot of evidence and witnesses behind. It’s up to you to get her out of trouble, and – if you’re skilful – to pin the blame on somebody else.
The game plays somewhere between a comic book and an adventure game. You can move freely around the boat, talking to the other characters, who all move around freely themselves, and remember everything they see, and do. You have eight hours of game time to hide the evidence, convince the witnesses they didn’t see what they thought they saw, frame someone else, and not contradict yourself!
How long did it take to develop the game and what were the greatest or most surprising hurdles?
The game was done really fast. In fact, we began the project as an experiment to see how quickly we could put a game together, and in total development took around 100 days from initial idea to finished game. To make that work we re-used a lot of old code, took design decisions we could be really confident with, and prototyped as little as possible.
So the most surprising things were things that turned out to be easier than we’d expected: for example, we weren’t originally committed to having characters move around mid-scene, rather than only moving at the same time you moved – on the map screen. But we had an idea how to implement it, tried it, and it worked well enough that we put it in the game.
You’ve now made titles with 3D movement (Heaven’s Vault) and those with more constrained exploration (80 Days, Overboard!). Which of those two styles is more challenging to design for and why did you opt to return to a more purely dialogue-oriented experience with Overboard!?
3D is about a million times more difficult to build for than anything 2D. One can lose entire weeks trying to get the camera to go somewhere sensible, and to stop the player walking through cracks in the floor. It also limits what you can allow the player to do because everything happens has to happen on-screen, in visible detail.
We’re really proud of what we made in Heaven’s Vault, but it was often a very frustrating experience. Making more constrained games like Overboard! allows us to focus on the parts we really enjoy – complex stories, fun characters, unexpected player freedom.
What specific works influenced Overboard!‘s colorful Art Deco visual style?
Mostly travel posters! The balance and limited number of colours was key to the style, though our artists, Anastasia Wyatt, modernised the lines and shapes a little to give it a fresh feel. She also did a fair bit of research into costume to make sure the characters each had a distinctive look.
What is your methodology for mapping out the complex dialogue and event trees for a game with as many unique outcomes as Overboard!? Did you have to heavily modify the ink scripting language or was that a natural fit for this type of project?
ink is an incredibly flexible tool, and we’re very accustomed to using it now, so generally whatever we want to do with it, we can do. And the secret to using ink is not map complex dialogue trees, but rather to build everything as atomically as you can. That means every line of dialogue checks for itself if it makes sense in the current state of the game (rather than relying on where it comes within the tree); and every event checks for itself that the triggers it requires are all in place. The advantage of doing things this way its easy to scale up and build outwards, so a simple scene can become layered and complex simply by adding more detail.
Obviously, you have to test a lot to make sure all the paths work and there are no nasty surprises – and there always are! But usually fixing issues is as simply as adding a missing condition, or writing in a quick extra line of dialogue to cover a particular circumstance.
What is your favorite character or area in the game?
That’s really difficult. I love Lady Armstrong because she’s so poised, so sophisticated, and so brazenly horrible. Everything she says is insulting someone, and so she’s really fun to write. But my favourite character has to Veronica herself: she’s willing to do absolutely anything to get her way, but she’s still sympathetic and quite human, and is caught somewhere between being a manipulative genius and a accident-prone walking disaster. Finding that balance was a lot of fun.
Are there any gameplay mechanics or story sequences that didn’t make it into the finished game? If so, what caused them to be cut?
When I was writing the game I made myself a list of all the things I wanted the player to be able to do – things that I felt this sort of story ought to include. (Things like “seduce the Commander”, “smother someone with a pillow”, “steal from the Chapel”, “play cards”, “get hit in the face by a cricket bat”.) I was pleased to get almost everything on that list in!
The only feature we didn’t get was stealing the metal cross from the Chapel and walking around the boat, being able to bludgeon people with it. It was too complex to implement well – too many ways to end character’s storylines, too many bodies to deal with, and if you’ve played the game you’ll know the metal cross has another feature attached to it that would have required a lot of dialogue writing. So that one would have doubled the development time of the game, easily (and I’m not sure it would have improved the game.)
What’s next for inkle?
We’re currently mid-project on a larger, longer game, about walking in the Scottish Highlands. We’ve been blogging about development here (https://www.inklestudios.com/highland/) but there’s no word yet on when the game will be ready – there’s still a lot to do!
Follow inkle on Twitter to keep up with all the latest on this game and future projects. If you’d like to see more from me, you can find me on Twitter under the handle @SinginBrakeman. Thanks for reading, and be sure to let everyone know what you think about Overboard! in the discussion below.