Game developers serve an interesting role in speedrunning. They create the games players speedrun, obviously crucial to the process. At the same time, most runners are trying to beat a game by any means necessary, and they often work glitches and exploits into an established route. For many runners, it can be a concern as to whether developers actually like the outputs of their hobby, or if they hate seeing their games reduced to optimization and skips.
We sat down with Kyle Pulver, an indie developer who worked on Snapshot — a sidescroller premised around storing objects in snapshots for later use. Joining him was Nightmare47, who speedran Snapshot during Awesome Games Done Quick 2016. We talked about speedrunning, game development and the crossover between the two.
What's it like watching your game in a GDQ?
Kyle Pulver: It is both terrifying and exciting at the same time. One of the things I joke about with other game developers is whenever I see our games on Twitch, it's really cool, but at the same time we're all thinking, "Oh god. Please, don't crash. Please, don't have any crazy things happen. Make sure the game goes well."
I had no idea what was going to happen for the run. I wanted to see everything happen for the first time for the GDQ run.
Nightmare, were you nervous that a developer was watching you perform the run?
Nightmare47: I was actually a little excited. The game's not extremely glitch heavy. So it's really nice to show off, "Hey, I can do this with your game." And it's really cool to look at, especially with this game. It's not something you'd normally expect just watching someone play the game.
Did you know about that glitch before you saw it in the run?
KP: That one we knew about. It was a matter of "We're finishing the game and there's only so many things we can fix with the time and budget we have," so that one stayed in there. It's cool to see aspects of the game that I didn't really know were useful.
What kinds of aspects?
KP: When I saw that bug in development, I didn't think it'd be useful in a speedrun. I don't think I was ever able to control it to hit the exit of the stage. I would just fly. Like, "Welp, that happens one out of a thousand times and I have no idea what's happening." Things you don't think are going to be purposefully executed will end up playing a big role for something like a speedrun.
Do you want developers to think of speedrunners during development?
Nightmare47: A little bit. I feel like if they focus on speedruns, it may not click with me. The games I run, I really like. The games I don't run, I don't like. Naturally, if a developer isn't focused on a speedrun they might add things like auto-scrollers and slow, counterintuitive movement. I feel like if they try too hard it might not "run" well.
"You really only notice the bad luck."
It sounds like you want a good game before a good speedrun.
Nightmare47: I do really enjoy good games over good speedgames, but either works.
The conversation shifted towards RNG, or Random Number Generators, a term often-used to describe random events that happen in all games. Speedrunners are distinctly aware of RNG, as many runs can be entirely dependent on good luck. While Snapshot had very little intentional RNG, Pulver described a common development attitude towards randomness, particularly in games that have a small amount of randomness, but aren't dependent on it.
KP: I don't know if you know about casinos and slot machines — they're pretty effective. There's this whole system of a random reward schedule, where you might anticipate getting a critical hit. When you get one, it's this really big payoff in your mind, even though it's not your doing, it's the game randomly giving you a treat. A lot of games do this kind of thing.
Kyle Pulver, indie developer
Nightmare, do you feel the casino effect when running?
Nightmare47: I think it has the complete inverse effect. You just get mad and angry when it gives you nothing. I've noticed this with a lot of people, myself included. You kind of get this confirmation bias that you get bad luck, and that's all you get. You don't take into effect that it's probably more spread out than that. You really only notice the bad luck.
Were you thinking about speeduns during the development of Snapshot?
Kyle Pulver: A little bit. I kind of mentioned this during the run. When you develop a game, part of developing it is playing it all the time. Whenever you push the compile button to test something out, you're playing the game. For a content heavy game like Snapshot, where there's a bunch of levels and chapters, you have to play through the levels over and over again. You end up starting to speedrun your own game.
Things that made my runs as a developer more fun, we leave in. We didn't plan on placing an object and jumping off it mid-air, but when we found out we could do that, it was really awesome. [Speedrunning] is something I consider a lot. I'm not as hardcore as the people at AGDQ, but I do like to run indie games sometimes or go back to Super Metroid. I think as an indie developer I think about speedruns more than other developers.
This interview has been condensed and edited for clarity. For more on AGDQ and speedrunning, check out our StoryStream. More work from Kyle Pulver can be found on his website or on Twitter; you can follow Nightmare47 on Twitter as well.