Tuesday, April 12, 2011

Using Lightswitches To Kill: An Interview With Gunpoint Creator Tom Francis

Never gets old

Tom Francis, of PC Gamer (have you read his fantastic Minecraft Diary yet?), is working on an innovative freeware platformer called Gunpoint. And it looks brilliant! So click this video below and look at it looking brilliant. And then read on for my interview with Tom.

Laser Romance: In a nutshell, this game appears to be about infiltration and rewiring, the latter of which I haven't seen a great deal of in games. Are there any additional unique mechanics?

Tom Francis: Hah, well I'd hate to claim anything was actually unique. I'm sure the idea of rewiring stuff like this is already out there somewhere if you look hard enough. The rest of Gunpoint's built from things I liked enough to steal from other games, and things I hated enough in other games that I wanted to make one of my own just to try the opposite way.

A big one is guns: I hate it when guns just do damage. Guns are exciting because they can effect a massive and sudden change to a situation, that's why in movies their mere presence is often all the drama the scene needs. But guns in games just shave off hitpoints, they don't do anything massive or sudden until an arbitrary number ticks down somewhere. They've lost their teeth.

So in Gunpoint when you shoot someone, they die. Someone shoots you, you die. It's very, very sudden, and it means the game is really about being clever enough to never run the risk of getting shot. Enemies are so scared of getting shot that they won't fire back when you have them at gunpoint. And when they have you at gunpoint, hopefully all the times you've been shot in previous lives give you an instinctive fear of it that makes the situation tense. It'll be really interesting to see if people like that kind of play.

LR: Very good! So is this your first game?

Tom: Apart from a 48-hour competition entry, yes. I made a little top-down shooter in a weekend for a challenge called Ludum Dare last year - it's pretty rough, but that ridiculous time limit really taught me to be smarter about what's essential to a game and what you can afford to cut. You can play it here if you're interested: http://www.pentadact.com/index.php/2010-12-21-a-two-hour-patch-for-my-two-day-game.

LR: At the core of a good invention is this phrase, "Why hasn't this been made before?" This is how I felt about Gunpoint when I first saw your video. Has the nature of the game evolved a great deal as you have developed it?

Tom: Cheers! Yeah, it was originally about a murderous robot in space pretending to be a private detective. That's why the main character is very covered up by a big trenchcoat and hat - originally he was trying to hide the fact that he was a robot. From space.
I still kind of like that idea, but I didn't have anything I wanted to say or do with it. I got much more excited thinking about the private detective angle, the cases he'd take on, the city he'd live in. Being a murderous space robot sort of enforced a disconnect between the player and any human stories he might get involved in - what does he care? He's a murderous space robot!

So eventually the space robot pretended to be a PI so hard he actually became one. Later I realised that 'spy' was really a better term for what you do in Gunpoint.

LR: I'm sure you were excited to create a game that brings some interesting mechanics into a platformer. Where did you get the basic idea?
Tom: The idea to let the player hook up any electrical devices to each other came when I was trying to think of a good hacking system. All I knew before then was that I wanted you to be able to subvert things, the way you can hack a turret in Deus Ex and BioShock. What I enjoy about that is not "YES AWESOME ULTIMATE FIREPOWER", it's that all I did was tweak the way something worked, and my problems neatly solved each other.

Whenever I like something about a game, I want to figure out why: what the essential appeal of it is. Now that I'm making a game, I'm trying the next step: how do I capture that appeal in a new mechanic? In the case of the Crosslink, the device I ended up with, it came from trying to make that appeal more universal, applicable to more things around you, more potentially intricate if the player wanted to be clever about it.

It might also have something to do with an electronics kit my dad gave me when I was a kid. It had inputs like switches, buttons and motion sensors down one side, and outputs like lights, buzzers and relays down the other. You could link any of them to each other with wires to create little systems and circuits. It was awesome.

LR: That is awesome! You know what else is awesome? That you started working on this idea with Game Maker. Are you still using it for development and can you talk about the process at all?

Tom: Yep! It's a pretty complete tool, all I really use is that and an image editor. Spelunky, a randomly-generated platformer that's become one of my favourite games of all time, was made in Game Maker. That's why I decided to make a game - I realised a beginner's tool like that is actually powerful enough to make something genuinely brilliant and professional. I don't think I'm smart enough to do it, but so long as the potential is there it's exciting to try.

Because I'm only working on it in my spare time, I do a lot of planning, rethinking and revising in between coding sessions. It actually works really well, because I'm always stepping back and seeing the big picture.

You know how most people have their best ideas in the bathroom or waiting for the kettle to boil? It's because you can't see the interesting connections between the reams of info bouncing around your brain when you're concentrating on something: your prefrontal cortex drowns it all out to focus on the task. So only having time to code for a couple of hours at a stretch is a good way to work, I always solve last week's problem in five minutes when I sit back down to it.

Ideally, of course, I'd love to rapidly prototype every half-formed thought I have, because you often see new ideas even in mistakes. But I'm not a programmer, so actually implementing something takes me longer than it should. I have to be efficient to get something done.

LR: Are you still looking for an artist? Do you think a July release, which you mentioned in your video, I believe, is still possible?

Tom: I've had more than enough amazing submissions from people wanting to help out with the art. I'll be making the final decision about who'll be handling the art this week - a few of the best are still working on sample pieces.

It's been a really mind-blowing and humbling experience - I just put out a crappy video, and suddenly all these talented people want to help. I feel like an asshole for turning any of them down, but at some point you have to put what's good for the game ahead of how nice you'd like to be.

From my perspective, July seems pretty doable. I can't speak for the art side of things, but just last night I finished adding the last actual features, so every system I'd planned for the game now works. All that's left is making the levels, writing the dialogue and tweaking the game.

How long it takes from here is really just a product of how big I want to make the game. Luckily there's nothing about the design that demands it be incredibly short or ultra long, so I can just keep going until it feels right.

I really didn't expect Gunpoint to really appeal to people or get any attention - it was just a fun test game at first. Now that the response to that video has shown me people like the idea of it, and the art submissions have shown me how good it could look, I'm starting to think it's worth really putting some effort into making it special.

July gives me one month to create levels and dialogue, one month to work in the art, then one to two months of testing, tweaking, fixing and rethinking. If it takes longer than that, that's fine by me - it's not like I've got a publisher breathing down my neck. My total expenses for one year of development are £15 to buy Game Maker and $12 to register a domain (gunpointgame.com), so I'm completely flexible.

LR: I hate to ask this, but not enough to not ask it! As someone who is largely known as a games journalist rather than a programmer or developer, do you think people will view your work with skepticism?

Tom: Hah. I'd understand, but no, I certainly don't think it puts me at any kind of disadvantage. Totally the opposite: the hard part when you're just starting out as an indie dev is getting anyone to pay attention, and I'm very lucky that some people already know who I am. If I make a terrible game, of course, it'll be slightly more embarrassing. I'll be picturing the whole industry sadly shaking its head and tutting, "Oh dear. Thought it looked easy, did you?" I'd probably just keep trying, though.

LR: Excellent! So is it really true that Gunpoint is going to be free? And PC only?

Tom: Yep. Some people seem surprised by that, which makes me think "Damn, I could have got away with hawking this!" But it's my first game, I just want to get it in front of as many people as possible and learn as much as I can from their reactions. If people really wanted it and the idea turned out to have legs, I could always do a very cheap second chapter afterwards. It would be cool to see how well I could take on feedback about the first release, and to find out if anyone likes it enough to buy the next one. That's pretty far-fetched, though - all that's happened so far is people didn't hate a video as much as I thought they would.

The same goes for other platforms: definitely not against the idea, but for most it'd mean starting the game again from scratch - I'm not expecting the level of interest it would take to justify that.

LR: Is Gunpoint the first of many games ideas or is this more of a one-time project?

Tom: I already have a design document written up for the game after Gunpoint, and I add to it every couple of days. In fact I have a plan for the one after that, but it's pretty vague. I'd like to make one game in every genre I have a love/hate relationship with, so Gunpoint is the platformer, the next game will be an RTS, and the one after that will probably be an RPG. I want to have a useful understanding of what makes the good ones good, and what could be done differently to make them better. Being a critic's a good start, but you never know if you're right until you try making something to test it.

LR: So is Gunpoint the Citizen Kane of games?
Tom: Heh, that was actually a milestone on my to-do list (http://www.pentadact.com/index.php/2010-07-14-gunpoint-making-the-jump) - the one before it is "narrative reduces grown men to hopeless fits of sobbing". I'm just about to start writing the dialogue choices and stuff, so I, er, have some work to do.

LR: I trust Gunpoint will make me sob in a crying-because-I'm-so-happy kind of way! Tom, thanks so much for talking with me today. I truly can't wait to electrocute a man with a lightswitch.

Tom: Thank you! After seven years of interviewing developers, it's fun to try the other side of the table. There are pillows over here.

That's all she (he) wrote, folks! If you are interested in finding out more about Gunpoint, check out the new site http://www.gunpointgame.com/. It currently redirects to the Gunpoint material on Tom's blog, but will eventually turn into its own site. Tom has also started a Twitter account for Gunpoint: http://www.twitter.com/GunpointGame.


  1. Nice interview Justin. The clip's face punching sound effects made me laugh. I love it.

  2. 'Preciate it Scott. Tom is a lot of fun to talk to!

    Yes, those punching sounds are oh-so-good, aren't they?