EPISODE 1909

[INTRODUCTION]
	
[0:00:00] ANNOUNCER: Bennett Foddy is a legendary game designer known for creating wholly distinctive games such as QWOP, Getting Over It with Bennett Foddy, and the recently released Baby Steps. He's also a former professor at the NYU Game Center, where he taught game design alongside developing his own experimental work. 

In this episode, Bennett joins Joan Nash to discuss his systems-driven approach to game design, why frustration and difficulty are often misunderstood, how streaming and speedrunning have reshaped how games are played and experienced, and what makes his games stand out. 

Joe Nash is a developer, educator, and award-winning community builder who has worked at companies including GitHub, Twilio, Unity, and PayPal. Joe got a start in software development by creating mods and running servers for Garry's Mod. And development remains his favorite way to experience and explore new technologies and concepts.

[INTERVIEW]

[0:01:07] JN: Welcome to Software Engineering Daily. I'm your host for today's episode, Joe Nash. And today I'm joined by Bennett Foddy. Bennett is a game designer infamous for titles like QWOP, GIRP, and recently Baby Steps, as well as an educator at the NYU Game Center. Bennett, welcome to the show. 

[0:01:21] BF: Thanks for having me. 

[0:01:22] JN: So, let's get started with what your game dev journey was. How did you get into making what we'll cover is a very frustrating selection of games? 

[0:01:29] BF: It started in around 2005. I started to become aware that something was happening in the world of video games. I'm really kind of a lifelong passionate gamer. But one of the things that started to come through were there were free games. There were games that were made by very small teams that were starting to kind of gain attention, and I think we started to call them indie games around that period of time. 

But some of them were coming out of the doujin scene from Japan that was a little older. Some of them were coming out of a kind of independent practice that flowed from shareware in the '90s and so on. But there was definitely something feeling like it was happening around then. And I guess Adobe Flash started to become a platform that you could make games on at that time. 

I started in a period of my life where I was trying a lot of different things. I'd been in a band. I was making TV commercials for mobile phone ringtones. And really, I was supposed to be writing a dissertation for my PhD in philosophy and procrastinating like crazy. But I started to get into writing Flash Games and sharing them online. And it was a period of time where you could knock something together on your laptop, pop it on a free shared web host, and it would get featured on major gaming blogs, and people would be playing it in the sort of tens of thousands. That's how I got into it. But it took me quite some years to be able to get into it full-time. 

I think probably the first game I released was 2006 or 2007. Made QWOP in 2008 when I was in America. I was doing a post-doctoral fellowship in philosophy at Princeton University. But that started to kind of gain attention, not right away. I think I released QWOP in 2008, and it made a reasonably medium to large splash on enthusiast gaming blogs. 

But in Christmas of 2010, I'd moved to England for a second post-doc in philosophy. Suddenly, my web server was melting, and my traffic had gone up by a hundredfold. And QWOP had kind of exploded in interest and popularity. From that point, that was kind of what planted the idea in my head that maybe I should be doing this stuff full-time. Suddenly, WIRED Magazine wants an interview, and I'm being invited to conferences to give talks and stuff. It started to feel like something that was just worth pursuing with my full energy. 

I suppose the very short version is that in 2013, I moved back to New York, to NYU, where I took a job at the new NYU game center. At the time, it's like the games department in the art school at NYU. And I taught there and made games alongside for 8 years. And then we started working on a game that was big enough that it really needed my full attention. And so I quit my job and went indie, as we put it in indie game circles. We worked on that for basically 5 years, and it came out in October. It's called Baby Steps. And so that sort of brings us up to the present. 

[0:04:29] JN: Amazing. I know that obviously you mentioned QWOP, it kind of went viral pre the concept of reality. I definitely was aware of it, and I know many of our listeners will be. But folks who haven't encountered, how do you describe your games to people? 

[0:04:46] BF: I think a lot of them are kind of absurdist. A lot of them are grounded in simulated physics, physics engines. I don't know. I mean, I think simulated physics, that's kind of a mouthful. But I guess in the early 2000s, they started to be real-time physics engines that had fully simulated dynamics. And then they started to become available as open-source engines. Box2D was the one that I made QWOP in, for example. 

So, a lot of my work has been around kind of playing with what you can do with those engines. A lot of them, people think that they're difficult. People think that they're maybe indifferent to the player's experience. I don't know. There's a certain aesthetic of that. I resist the idea that they're difficult. I think maybe difficulty is kind of a complicated and overloaded idea. 

But people think my games are difficult anyway. And a lot of them concern movement, I suppose. The kind of movement of the player, and making that kind of problematizing in certain ways. I have some things that are kind of outside of that through line, but that's definitely the most kind of recognizable through line of the three most recognizable games in my - which would be QWOP, Getting Over It with Bennett Foddy, and the new one, Baby Steps, which I made with two other people. I made it with Gabe Cuzzillo and Maxi Boch. 

[0:06:04] BF: Perfect. Yeah, we'll definitely come back to some of those games out of the box, like Zipper, I want to talk about. So, we'll definitely come back to three of those, as well as your collaborators, because I believe at least one of them is a former student as well, right? 

[0:06:13] BF: Yeah, that's right. 

[0:06:14] JN: Yeah. Awesome. Cool. Perfect. Yeah, I guess to stick with a couple of things I want to talk about from what you just mentioned. Obviously, difficulty. We'll come back to the difficulty concept, difficulty. But physics. So, you describe I think it was to Polygon. You said that like a lot of your games are essentially the interaction between a physics rig and level design. And I thought that was a really interesting I guess kind of like basic place to start from designing a game. Can you tell us more about that? 

[0:06:36] BF: Here's the thing. It was at the Centre for the Moving Image in Melbourne a couple weeks ago, and I was trying to kind of express it in this particular way, which is that when you're designing games, there's basically two ways that we get mechanics, or sort of genres, or types of play. Some of them, or maybe the majority of them, the designer, maybe not starting from zero, maybe starting from a bunch of genre assumptions, has an idea. And the idea is like a full kind of game idea, and they implement it as they imagined it. Maybe they kind of change course a little bit along the way as they kind of learn more about it, but it's kind of created from whole cloth. 

But the kind of games that I've been making, especially with physics engines, it's not so much like that. It's more like I'll create a way of interacting with the game world and kind of play with it. And the gameplay loop is sort of found rather than invented. Right? For QWOP, for example, I had a hope that it would be kind of a fully controllable, easy running system that would have some kind of interesting implications or interesting things that kind of fell out of that system. 

What it actually is, is something that's much more absurd, and silly, and funny, and difficult, right? And sort of, in a way, it wasn't at all what I was looking for, but I liked what I found. So, it feels as though it was a little more discovered. And I think finding things, discovering gameplay in a system, is easier when you start with a very rich and deep system. 

What's a physics rig is a physics rig is a relatively small number of entities in a physical simulation and a set of constraints. And the constraints might be the level of the world that it's in, or it might be the joints, it might be the muscles, it might be some other kind of constraint, right? But it's very, very hard, I think, as a designer to get more than a sort of intuition about what sorts of constraints will be interesting or what will happen with the constraints, because physics is complicated, and it's unpredictable. 

And so the way that I approach doing it, or that I have approached doing it through my career, is to kind of set up some constraints and see what falls out of that. Maybe I have an idea about sort of roughly what it will be like. Very frequently, there are things in there that aren't exactly what I predicted. And then trying to kind of follow what's actually there in the playing with the prototype. 

I think relative to a lot of other game designers, I probably make more prototypes and throw more away. Partly because of this approach. If you're sitting down to make a grand strategy game or a puzzle game, you probably don't throw as much away. You probably sort of sit there, and you iterate until you get it working exactly the way you want. But when you're working in the kind of games that I've been making, I think a lot of the time, what seems like maybe a good idea, you realize actually, for some sort of deep reason relating to geometry, or relating to momentum, or something like that, there's not actually a gameplay there. 

That's why I think it's playing with rigs. Sometimes it's playing with level design idea. It can play be playing with any kind of constraint really, but it just has to be a rich enough set of constraints.


[0:09:50] JN: Right. Interesting. I guess a lot of your physics rigs tend to be human form or close enough. Obviously, you've had unicorns but still four limb, etc. Is that an artistic inclination, commenting on human body, etc., and it's usage? Or is that because eight-legged alien monstrosities don't make satisfying physics rigs in your gameplay design? 

[0:10:10] BF: I think it's two things. I think, first of all, there is a relatability thing. I think human bodies make us feel a certain way. Games that are about movement are really about embodiment. They're about feeling like you are the avatar that's represented on the screen. It's a kind of easier move for your brain to make if it's humanoid or at least it's that kind of animal. It's not impossible with a robot, obviously. It's more of a heavy lift. And so, I sort of feel like that kind of emotional response, and that kind of psychological response is easier for a human. I mean, I think that's the first thing. 

The second thing is that with video games, I really feel like how much I enjoy playing a video game and how people enjoy the kinds of video games that I enjoy playing is kind of opposite reading explanation, or doing a tutorial, or something like that. There are certain types of games where that's totally unavoidable. You can't learn to play Catan without being told the rules. There's no way to intuit it. And Catan is a wonderful game, right? Really, I'm not trying to run those sorts of games down. 

But what I love in playing games, especially real-time games, video games myself, is a sense of being able to kind of explore and discover what the rules are, and to be able to jump right in and for kind of learning how to do it being a lot of the fun. And that being the case, the players need help to understand, first of all, what are the goals, what are the limitations, what's going on, what can happen, what should their expectations be. There's all this stuff that has to be conveyed by the game designer. 

We try to kind of pick settings for games where there's just less to do on that front, right? So, if I give you a human body right away, you know, it can't fly, it's fragile, it takes steps of about this size, maybe it gets fatigued, maybe not, but weighs about this much. There's all of these intuitions that we have. There's all this stuff that we know, all this common sense that we bring to bear when there is a human body pictured, right? 

The same thing goes for - in QWOP, the situation is he's in an athletic outfit, right? And he's facing left to right. And right away, you have a set of whole thick bundle of expectations and knowledge that comes from the fact that you've watched the Olympics. You know what his basic task is. In fact, the humor of that game comes from subverting what you're expecting. You see a very athletic athlete dressed up for a race inside of a stadium, and you expect competent running. And then it's the players. It's because of the players' incompetence that he is falling over. That's the joke of the game. But you're just getting a lot for it being the human body. 

And I think as a game designer, I'm just always looking for these sorts of shorthands because that is really what allows us to remove explaining and to remove tutorialization, and difficulty curves, and all of that stuff that has its place. But it's not really the - if I'm doing anything as a designer, it's to kind of try to skip that stuff, to try to eliminate that flavor of it in order to be able to give the games a different kind of personality that's maybe a little more adversarial, a little more resistant to the player than a regular video game. So, it's more important to my work than it is to other designers as well. I just really need everything to kind of be self-explanatory. 

[0:13:36] JN: Yeah, that's really interesting. Yeah, I want to stick a pin in progression. But I guess before we come back down to that and difficulty curves, obviously, there's a lot of similarities thematically between QWOP and Baby Steps. I guess I want to start by asking how did you return to the idea of manual walking, I guess, for lack of a shorthand, in your first - as you said, going indie. Why come back to that idea? 

[0:14:00] BF: We had spent 2019 making prototypes. We made about 10 different prototypes, and none of them were about that, really. We tried a lot of different things. We had a twin-stick shooter. We had a competitive Sim City. We had a lot of different things going on and none of them were really working. And it was actually at Gabe's suggestion. He felt like we could, in essence, do what I'd been trying to do with QWOP in 2008, which is to make a version of that where you actually can walk. 

But I think the kind of core of his idea was more like he's very interested in systems that are driven by control and where the physics is following that. He'd made this game called Ape Out, which I contributed artwork to. And that's an arcade game. It actually does use a physics engine. But when you are moving your analog stick on your controller to the right, immediately, it's not waiting for - it's not applying forces to a physics object. It's setting the position as we used to do back in the 1970s and 80s. We're just setting the position of the controlled object. You could think of that as being like one-to-one control or something like that. We could call it arcade controls. 

And I think the kind of core idea that he had was that we could drive a 3D physics rig of a body to make a kind of a QWOP-inspired game where the feet were the kind of like the arcade object, right? The feet wouldn't be like dynamics objects. They would be canonically controlled by your controller. And I tried to discourage him from that because I thought that was a bad idea. I thought it was too hard. But he made a prototype very, very quickly, like a small one that seemed a little promising. 

And then we hooked it up to a physics engine, and there was some middleware that made a rag doll follow an object. We hooked up an actual 3D character to it, and I made a level for it, and it was like, "Yeah, we're doing this." We just knew, I think, right out the gate after having sort of 10 failed prototypes. This is quite early on. I said I have a record of a message to him where I said I think this is the one. This has got legs is the way we often talk about it. But yeah, that's what you're looking for if you're going to spend a significant chunk of time on something, is something that will be worth exploring that has enough depth to be able to make it worth your while, to do polish, to do kind of a detailed world, and all of that stuff. 

I think we thought we were going to spend about one year, not sort of five or six years. But yeah, he just knew. Yeah, it was his inspiration. But yeah, I think he wanted to do something that was kind of like a fusion of the procedural animation that he had done for Ape Out. And then the leg placement from QWOP worked really well, I think. 

[0:16:49] JN: That's great. And obviously, a huge departure between those two games is one's kind of a self-contained experience. Now, Baby Steps is this open world narrative journey. And going back to what you said about it needing to feel intuitive and trying to skip tutorials, I'm fascinated. But how do you deal with progression and skill building in a game like Baby Steps, where you've got this fundamental mechanic that's so like, "Hey, we're taking something easy and making it hard." And now you need it to continue getting harder throughout. 

[0:17:15] BF: Yeah. I mean, I think our orientation to that changed along the way and in the same sort of way that QWOP becomes about failing and falling over, because that was all I was able to achieve. I think the early prototypes of Baby Steps were much more falling-down-centric. And we had much lower expectations for how good people would get. 

I mean, the truth is, is just years of design iteration on the walking model on the control scheme. Just little things. What happens when you let go of the stick? And how is it conveyed that he's at the edge of his kind of polygon of balance, or whatever it is. There's just little changes. And Gabe was in charge of that stuff. He would sometimes have a shower thought that maybe we'd been doing it all wrong, and then come in and make some kind of change. 

And then little by little over time, it started to be that people, when they got to the end of this little cave that we set them up in at the beginning, most of the players were kind of trucking around almost automatically, right? It went from being a game that was going to be more failure-oriented like QWOP, to being something which is like, "No, you're walking around actually." Competently for the most part. And you're falling over when you're trying to go up a very steep slope or on a very slippery substance, or do some kind of incredibly technical climbing maneuver. 

And it enabled us to make the world around the rig more and more kind of complicated and more and more challenging to the point where I just don't think at the beginning we dreamed that it was going to have some of the sort of jumping between objects or riding on moving objects, all of that sort of stuff. It would have seemed impossible in the first instance. But yeah, it's just iteration, iteration, iteration. 

It's the longest I've ever worked on a video game. I never really had done that kind of degree of iterating before. Yeah, it sort of brings out a different flavor. But these systems are very, very complicated. And the sorts of things that make it hard or easy aren't what you would expect all the time. I mean, we really had a strong aesthetic against having too much in the way of imaginary forces propping the character up. We wanted it to be as kind of plausible as it could reasonably be while still being stable. 

But the sorts of things that make it go from difficult to easy are concerning things like suppose you're pressing the button to raise the trailing foot a little bit too early. What should happen then, right? Or suppose you do it a little too late. How should that be handled? How should we handle letting you know when the foot is up or down? Where should the camera be pointed? 

A really big breakthrough we had is to realize the camera should always be focused on his feet, especially when the camera starts to kind of bump into things in kind of difficult terrain. It always zooms in on his feet, right? If it has to move at all. That made it massively easier to play. So, we're just constantly looking for these little things. 

And what you end up in a game of this sort of scope, you end up with a titanic pile of minor affordances, right? And that all kind of blends together into a system that seems simple, but you have, in fact, added this colossal complexity and nuance, really. More so than complexity. But it's now, instead of 10% of players finding it easy to walk around, it's now 90%, right? Really kind of inverted that way. 

[0:20:35] JN: Yeah, absolutely. I guess that reminds me of the is it Super Meat Boy or Celeste? One of them has the famous 2000 line file of your statements for what happens if you land on a platform and have to deal with it? One of those. Yeah, absolutely. 

So I guess get to the topic of difficulty, which I think broadly your fan base is in agreement with you that difficulty is not the right word. You are one of two guests we've now had that has a genre named after them, which is the Foddy and rage game in your case. You've got this really great blog post out, the 11 kinds of frustration, or I think it was 11 kinds of difficulty, that I think captures - I don't know if still captures your current feelings on difficulty. But I think you explore different ones of those different games. Obviously, Getting Over It is very much, I think, one that would put your name on the map quite intentionally. It has your name of it. Yeah, I guess talk to us about your feelings on difficulty and your approach to it in games, or not difficulty as you say. 

[0:21:20] BF: The idea of the blog post is that we have very limited language, I think, for this stuff. Let's just put a pin in difficulty for a second and talk about frustration, because it's like a thing that at least Getting Over It and QWOP are associated with being frustrated. And it's definitely a flavor in my game. But I feel like that's such a blunt over-encompassing word, because actually there's all kinds of different feelings that we call being frustrated or being annoyed. And we don't really distinguish between them very much in English anyway. I don't know if it's different in other languages. 

Yeah. I mean, as a player, I feel like different senses of being frustrated have very, very different emotional consequences to me. And one of the interesting things about playing video games, not just ones that set out to be hard, but all video games really, is they all have a little bit of a mix of different ways of kind of provoking you and feeling like the game is pushing back in some way or offering friction. There are some frictionless games. They're in the minority, I think. Some visual novels, for example. Even walking simulators, as we call them. It's a genre I'm very fond of, where you just walk around and look at things. Even those have friction, right? So, it's definitely something that's universal in games. 

But a big part of design is about controlling what exact feelings that you're provoking. And I just got interested in introspecting on that. I mean, when I play games, whether they're single-player games or when I'm playing competitively with friends, one of the main sources of joy for me is to savor the feelings that I'm having. And actually, this goes for watching movies and listening to records as well. When somebody is trying wine, when they're tasting wine, if they're an enophile, they smell it, and then they put a little bit into their mouth and they kind of taste it on the different parts of their palate. And then they try to kind of put into words what exactly it is that they're tasting. I'm not a wine person, but that's what they're doing, right? 

And I think it's a little strange that people don't really do this by and large with other art forms, particularly video games, which are really kind of good at exposing, I think, to you what your feelings are while you're having them. So, when I'm playing a game, and I'm starting to get on tilt, that's a thing that we can sometimes say, "I'm starting to choke. I'm starting to become overly self-conscious of my actions." Maybe I'm becoming a little heated. I start to feel heat rising in my face. I might feel a little kind of pit of disappointment in my stomach. I might start to become a little bit unsteady or trembly if I become like a little - I get a little bit of a fight or flight reflex or something like that. Those are all things that I think - I'm thinking about that in the same sort of way of sort of rolling that around in my mouth and tasting it on different parts of my palate. 

And games have such a rich ability to do different things. That's one of the things that I really am fascinated about with games, what I really am pursuing. In my work, what I'm trying to do is figure out what the colors are on the palette and then to try to paint a picture with those colors in whatever kind of way. 

Difficulty, I don't know. I mean, I think we're sort of stepping out one kind of layer of abstraction, right? What is difficulty? Well, difficulty is that you expected the game to give you a reward or to say that you're the winner when you did a set of actions and you did those actions and it didn't reward you with what you expected to be success. But it's really built around what your expectations are. 

I think you see that very easily in a simple game like QWOP. Your expectation is you can run to the end of the track because you're an athlete. And the fact that you can only go 3 m makes it difficult. But is it difficult? I mean, I don't know. I mean, if your expectation is that you can go 3 meters and that something funny will happen, then it's very easy. Everybody can access the full range of experience in QWOP. And it's not like you've missed anything if you didn't get to the end of the track. You had the full experience if you spent 20 minutes trying to walk and kind of going upside down. It's really, yes, about what people want out of the game. And it's about what they expect to get. And it's about how much effort they expect to have to put in. 

And then I think there are situations where games expect a huge amount of work, but they don't get called difficult, right? If you sit down to play Europa Universalis, it's just this mind-boggling complexity you can't possibly hope to overcome without watching YouTube videos, or having a friend to guide you through, or something like that. But that game's not known as difficult, right? For some reason, we would only talk about the difficulty of that game in terms of how the AI plays against you, and what the resource constraints are, and all of that stuff. It's a difficult grand strategy if the AI is really fierce and is constantly kind of attacking you. 

But there's just as much friction in that game or way, way more than there is in a lot of my games, right? I think the same thing goes for accessibility, right? There's a lot of talk about accessibility. Usually, accessibility features for a game are ones that enable you to see the ending, to see all the content regardless of how good your reflexes are, or how good your vision is, or whatever. Whatever other sorts of abilities your body has. 

But we don't often talk about accessibility in terms of how easily and quickly does the game get you to the kind of core of the experience, right? It's never deemed that a game is inaccessible because you have to do 8 hours of tutorial before you're playing the real game. If you're playing Final Fantasy XII, you're not really playing the real game until you're in about hour 20, 25, 30. And nobody calls that inaccessible because people's expectations and orientations are toward the storyline and the visual content. And we get that really easily. 

Yeah, I mean I don't really have like a kind of unified theory about this. But I would just say that just like with frustration, the kind of concept of difficulty seems a little bit overly broad and blunt, and not really kind of fit for the sorts of debates and discussions that we're trying to have with them. Maybe just a little more language or a little more theoretical constructs are needed to really be able to kind of crack these open. 

[0:27:34] JN: Yeah, it's interesting. Maybe it came across this way just because you know we're talking about your games. But the things you listed on one side and then the things you list on the other, Europa Universalis as an example, it really does to a certain extent. The way we talk about difficulty does talk a lot about losing progress of some form, which is obviously something you play with a lot. I know you've been heads down on Baby Steps for the last 5 years. But I was wondering if you're aware of the growing trend in games that are resetting progress, like hardcore World of Warcraft, hardcore Minecraft, and the fact that those are becoming super popular. And if that hadn't given you any thoughts. Because, obviously, that's site you've been playing with for years now. 

[0:28:08] BF: Yeah. And those modes have been around for a while. But I have noticed they've started to get more popular. The one that I love is RPG maps in Trackmania, especially. I mean, there was one called Bennett Foddy ate my checkpoints, which is you have to climb up a building. Very, very difficult things. And you fall back down, there's no checkpoints, right? 

And that one went on to be the - as I understand it, went on to inspire Deep Dip, which is a collaboration between a number of RPG map makers from Trackmania, which is this giant colossal tower that takes hours and hours. And the whole community was involved in a kind of collaborative effort to see who could be the first to beat this checkpointless tower. And then Deep Dip 2, which was even bigger. Again, Deep Dip 2, they had a community pot. It was like $30,000 for whoever could finish that first on stream. 

I mean, I can't really take credit for this except maybe in just a small piece of it. Because I feel like this kind of sense of hardcoreness, the sense of really being prepared to send people back a lot is something that started to really pop off around the time that Dark Souls came out. Demon Souls really sort of started that, these FromSoft games, which were like we're going to - in the clothes of a genre, action RPG, that had become very kind of progress-oriented, very number-go-up oriented, which is really what we mean when we say RPG systems. We mean you can't lose progress. That you're always ratcheting up progress, right? They were like, "No, that's no fun. Let's go back to feeling some fear because you can lose a lot of your effort." 

And it was received, I think, a little coldly by journalists and by a lot of players when Demon Souls came out, but it started to kind of catch fire. Some people found it had a very kind of passionately positive response to it. And so I think you can kind of see we're almost at like kind of peak sending you back now. And then we'll get another backlash. We'll have another 20 years of checkpointing, and quick saving, and all of that stuff. 

These things come in waves. They're fashions. And the reason for that is that there are trade-offs, right? I mean, I think what you lose when everything is sort of safe, and checkpointed, and quicks saved, and so on, you lose a sense of stakes. Obviously, if you can't lose progress, then you're never really afraid, and you're never really stressed. You can't really go on tilt. You're just kind of oriented. And it orients you more and more to the kind of micro successes and failures you have. 

If I'm playing, for example, Quake 1 with my fingers on F5 and F7 to quick save and quick load, I'm just trying to optimize every little kind of 1-second stretch of the game. And then I'll save if it was good enough. And I won't save if it wasn't good enough. And that's a way to play. And there's a whole aesthetic of that. But what you don't get is any kind of feeling of being out on a high wire. 

And by the same token, when you have a game that sends you back a long way, you have a lot of stakes. You have a lot of stressful feelings or whatever it is, excitement and feeling of it mattering how you play, but you never get to kind of really focus down on the micro. I think it makes sense that it's something that comes in waves and in fashions. But yeah, I think we're right now probably cresting the peak of sending you back culture. 

[0:31:35] JN: Yeah, it's really interesting. I always find it interesting because I have very much my game-playing friends very much fall hard on either line. There's a group of us who only play games that hurt us and are incapable of playing games that don't. And there's a group that just like do not want any of it, whatsoever. And never let each other cross in any way. We can't even come close to the axis. And it's just so fascinating. 

Changing tack a little bit. I think that explanation was a great example of I guess something not strictly game related. But I want to talk about your dual nature as an academic, as a game designer. Because aside from giving us wonderful views and insights into game design, and you also have amazing fun with it, I think your IGN reaction to the getting over at speedrun is art. I just think it's perfect. It's a perfect artifact. But I think you've used that to examine some things that are really interesting. 

One that you kind of touched on there when you were talking about Trackmania, you've been incredibly aware of some really interesting gaming subcultures and really spoken about them interestingly. But I think one that I want to chat to you about is - well, two I want to chat to you about, speedrunning and streaming. Because, obviously, I'm a big Ludwig fan. The Getting Over It generation was like a big thing for me. And I know that your journey very much started with like MoistCr1TiKal playing QWOP and everything as well. How has seeing streamers basically have their careers been made by your games? How has that influenced your practice at all? 

[0:32:51] BF: If I'm honest, I didn't really understand. Even having had that experience with MoistCr1TiKal and QWOP. And I have watched the occasional LetsPlay. I mean, I have to just be honest and say when I'm looking for a YouTube of gameplay, I'm always typing no commentary into the search. I didn't want that experience of seeing somebody reacting for me. I felt like it would kind of spoil things for me. 

And I definitely once started to see streaming take off. Streaming has lots of different forms. I have a lot of friends who watch League of Legends streams, which are really not reaction videos. It's more about watching people who are at the top of something that's very competitive play it in a way that they can't and to kind of learn from it and admire it the way that we admire athletes. 

[0:33:36] JN: The peak of a frustrating experience watching someone grind League of Legends. That is games that hurt you. No one's done better than that. 

[0:33:44] BF: I don't think they're all the same. But the kinds of streaming content that I've been most interfaced with is the stuff that my games have been supporting, right? Stuff where so-called rage game, somebody's playing a game. Because it provides emotion-provoking, relatable situations that they can have an outsized reaction to on stream. And that their fans can kind of enjoy the entertainment value of seeing their favorite streamer react in a certain way. 

And when I was making Getting Over It, I think it's clear in the script that I was sort of oriented against that, right? My orientation was it annoys me that people aren't going to play this game, that they're just going to watch it on stream. And here is this thing where the kind of meaning of it comes from the fact that you're going through an ordeal and you're sort of in contact with the designer. 

And I think that game is that for some proportion of the audience, especially people playing it the first time. I didn't have a deep understanding of what it is that people are doing when they're watching streams or anything like that. I was just kind of against it because I knew it was sort of stopping people from actually playing the games themselves. 

As it turned out, Getting Over It came out, and I think probably the great majority of people have experienced it in the format of watching people play it. And that certainly hasn't been bad for my sales. And it would be churlish to complain about that situation. So I've tried to have more sympathy for that experience, right? And to kind of understand that it's playing a different role. 

There's something that I do believe quite deeply, which is that it's something that's natural and that we should hope for as game designers is that the games are used in ways that were not originally intended. And I think a lot of the Trackmania RPG maps that I was mentioning before were never envisaged by the designers of Trackmania. That's not a usage that they would have supported. I mean, maybe they love it, but it wasn't their intention. That's quite clear. And I think that often times the kind of most beautiful aspects of games come from the sort of unexpected usages, folk usages of the games and the folk - the house rules and the different - the mods, and the modes, and all of these different ways that the kind of player base kind of runs with it. 

In the case of Getting Over It, some proportion of people had the sort of experience, as I intended it, and that's great. But there was this kind of phenomenon that was much bigger of people using it as a platform for rage performance. Now, I don't like watching rage performance. It's not a thing that works for me. It always seems inauthentic to my eye. But I certainly have enjoyed watching reaction content of one type or another in my life. I think it owes a lot to Japanese variety shows of the 80s and '90s in terms of format. Having the kind of celebrity face in the corner while something interesting happens, and you see them react. There's something that's very human about it. There's something that's kind of very basic and deep. And I'm not against the fact that Getting Over It was a kind of a good platform for that or at least a timely platform for that. 

And I guess to really answer your question, do I think about it now while we're making games? Yeah. Yeah. Now it kind of feeds in. We try to design everything both from a point of view of how this will work for a spectator and how it will work for the player simultaneously. 

And one of the nice things about that is it actually means that if you do that, if you focus on both, as we tried to do in Baby Steps, it means that games are really nice for kind of hot seat play in the living room with families. And I've heard from a lot of people who have said that playing Baby Steps, they had a wonderful experience with five people playing handing off the controller. It's certainly something we observed when we were play testing it as well. And I think that comes from a kind of an orientation to understanding that that kind of stream spectators will always be there. I mean, it seems good. I'm happy to have a kind of a broader conception of who the audience is. 

[0:37:44] JN: Yeah. I've never gotten a sense of how many of the players it is. I had a thing with Obra Dinn. I played Oradin, and I was like, "That was amazing, that was incredible." I wish I could erase my brain and play it again fresh." And I had a similar experience of Getting Over It. But my reaction to that experience in both cases was like I now want to see someone else go through this for the first time. 

And I think that those people in streaming get often, unfortunately, labeled as like the backseaters and banned and told to go away. But I do. I think that is - Getting Over It, I think a lot of the joy is seeing someone encounter the narration for the first time, seeing someone get bodied by it for the first time. Actually, quite a few people have played it. And they play it and then go seek a stream out to see someone get that bit. 

[0:38:23] BF: That's interesting. Yeah. No, I didn't think about that. But yeah, we do that with movies too, don't we? If we really love a movie, we want to be able to discuss it with people, and we want to be able to share the experience. So yeah, there is a kind of amount of experience sharing like that as well. 

[0:38:37] JN: Yeah, I guess on the subject of people doing unexpected things in the game, that brings us perfect to speedrunning. And I guess a very similar question, we've had game developers who have spoken about not intentionally putting in sequence breaks. But as they're designing a level, thinking about, "Okay, how are speedrunners going to engage with this?" Is that something that's factoring in at all? I mean, the current record for Getting Over It, I think, is disgusting. But how's Baby Steps looking? 

[0:38:58] BF: Baby steps, I think the glitchless record is something like 45 minutes now, which is crazy. It's really for a game that you just have to actually take steps in if you're playing glitchlessly. The glitchy one is something like 8 minutes. I think it's gotten down to about that little - 

[0:39:12] JN: The joy of physics-based games. 

[0:39:14] BF: Getting Over It glitchless is like, yeah, 45 seconds. Glitchy is faster. I'm not so interested in the glitchy runs. My preference would be to patch out the glitches. I don't think that you should. I think whatever you've shipped, it becomes the tech that the speedrunners are using to kind of compete for better times. You kind of have to leave those bugs in because that becomes the little platform of play that they're doing, unless it's kind of ruining anybody's experience. But you really have to search for the glitch that the 8-minute speedrunners are doing. 

Yeah. I mean, I think about this stuff a lot. I was aware of it and thinking about it when I was making Getting Over It. I think speedrun culture had already entered the mainstream in 2016, 2017. Every time I would play test it at NYU, somebody would say, "Oh, speedrunners are going to love this." And I'd be like, "Okay, well, I got to think about that a little bit and try to sort of support that." 

I don't know exactly what the requirements are. I guess to make a speedrun or be interested, it needs to not be too random of an experience. If they have to kind of reroll all the time, they don't like that too much. It's good if you leave the possibility of skips open. So, yeah. I mean, when we made Baby Steps, for example, Baby Steps has no invisible walls, give or take. There's like one or two situations in the dark area where we have to do that to stop you from getting soft locked. 

But in the open world, there's no invisible walls. Very, very rare for an open-world game. And we just assume. A network security admin assuming that every password is automatically compromised, we assume that every cliff is compromised, every difficult area, rock wall, whatever it might be. You have to assume that they're all compromised. And you make the game work allowing for that. I think that's how you support speedrunners as best you can is to make sure that the game doesn't crash and doesn't fall over if they do something that's not expected or not desired. 

I think I love that approach as a designer, really, in general. I think about one of the things that I found most inspiring was learning about some of the exploits people use in Spelunky. Spelunky is a game that has certain kind of things that look like they should be invincible. There's an invincible ghost that gets you if you take too long. There are invincible walls, etc. But it turns out everything in that game just has health. And the invincible things just have a ton of health. That's such a good way to approach design, is to never use absolutes, right? Just to always kind of leave the door or jar just a tiny bit for people to exploit things. 

Yeah, speedrunning, it's a nice way to kind of resee your work. To see people, again, reconfiguring it or reconceptualizing it in a different way. Speedrunning makes a game into a different type of game, in a way, for the most part. They're taking different pleasure from it. It immediately becomes its own type of thing. It is one of the great beauties of games that they support so many different types of play. Like Thi Nguyen, the philosopher who has a book out on games, has this idea that what's special about them is that they're unfinished, that they need to be completed by the players. 

I actually think this is true to some extent of all artwork. It's always the audience that completes things. Certainly true of architecture, for example. But it is particularly true of games, right? And I think it's important to say that it's true because we do. And one of the weird things about being a game designer is you're so in control of the world, and the rules, and all the nouns and verbs that you can get led into a power fantasy of being able to control people's experience and be able like, "As long as I polish this enough and I control it enough, I can make sure everybody's having the intended experience." And you can't, I think. And it's better to proceed with the with that assumption from the jump. 

Staying on Getting Over It for a second, then I've got another question for Baby Steps. But on Getting Over It, obviously, the full title is Getting Over it with Bennett Foddy. And you narrate it. And you bait the player a lot. And you've done a fantastic GC talk about that decision. But I guess in terms of putting your name on a game and directly engaging with the player in that way, was the reaction what you expected? Did it work as intended? 

[0:43:13] BF: I think it did. I think it did. It doesn't for everybody, especially the people who are experiencing the game through streams where the audio is muted. It's not going to work in that way. The idea, what I want, is to create an experience that feels like the player is in communion with the designer in some sort of way. Not in a way where I'm controlling their experience, but in a way where they sense me as a human being present in the work. And putting my name as part of that, but then also just have this narration that starts formal and becomes more intimate through the course of play, and try to forge that intimacy by finding points of aesthetic or emotional commonality between me and the player. And I do that by gating, right? 

Basically, if you got this far, it means that you must have been enjoying it, right? If you're enjoying it, let's talk about what that means. And I think that we have the same taste. I do think that worked. And I get a lot of email from people who felt something with that. And I'm very happy about that. So we can sometimes get people to feel a certain type of experience. 

We hear from people who played Baby Steps who had some of the kind of intended feelings about that stuff as well. We wanted people to interrogate why and how they play games, and why they persevere, and what that's really about. And to kind of question their own orientation to getting instructions, and getting help, and all these sort of different things that we could call masculine types of play, both for men and women. 

And I've heard from a lot of people playing Baby Steps that it made them reflect on stuff in those sorts of ways. So, we do at least get some people to have the experience that we want. We don't have to give up on intention completely. But we just have to have a kind of a loose grip. 

[0:44:57] JN: Yeah, absolutely. Yeah. So, back to Baby Steps. I think the post-Flash-styled games. You have an aesthetic that it really describes me of Garry's Mod. Maybe I describe too many things in this podcast as Garry's Mod. But it feels like default assets and half-naked man in a pot. And the Baby Steps guy really gives me Attack on Titan kind of vibes. What drives that aesthetic? Where does that come from? 

[0:45:19] BF: I think at first it was a practical consideration. I was making Getting Over It, and I was like, "Well, I have to control the scope of this somehow. I really love pottering around with 3D art and especially effects art. I could just spend forever rat hauling on one visual effect like Lucas Pope does, except that he's much faster and more skilled than me. So, it's maybe not such a problem for him. 

But even in his case, you can see the real pleasure for him is in making a special effect or a shader look just a particular way. But if I do that, then Getting Over It takes me 25 years to finish. So I was like, "Well, how am I going to do this? And then from that point, I was trying a couple different things. I had a early prototype that was like it was all going to be cutouts from historic furniture magazines and jewelry magazines. Trying to get like photographs from like high budget magazine ads, cut them out and then make the world out of that. Didn't really like how it looked. Wasn't quite working. 

But I think from there, I realized that if it's going to be kind of detritus in this way that, that assets would actually work fine like 3D. Especially free 3D assets, which tend to be of very mixed quality. And a lot of the ideas in the game, the narration actually flow from how that looked rather than in the other direction. Right? You might think that I picked an art style to kind of match the mountain of trash idea. But actually, it was in the other direction. Started as a scope constraint. And it kind of inspired a set of ideas in the kind of theoretical realm. 

Yeah. It's interesting to me. To me, it's something that is so - there are so many games that are made out of assets now. In fact, if you're playing the Final Fantasy remakes, you see many of the same Megascans plants that are also in the Resident Evil remake that are also in Baby Steps. There's not that many different repositories of assets out there. And they are in fact being used not just by cheap or free games on Steam or an itch, what we call an asset flip, but by things with massive budgets now. It's just kind of like universal. 

And it has its own set of cultural significance, right? Like a meaning. There's ways of meaning when you use off-the-shelf parts that it's just kind of visually suggestive of a certain set of things, right? For Getting Over It, particularly the fact that all of the assets are from different free repositories and they don't all sit together, right? They don't look like they're from the same reality. It gives a kind of a collage or kind of bricolage sense of unreality or of surrealism that is flowing out and that I started to like. 

And I guess I started to think while I was doing it, "Well, this could be beautiful actually by my own lights." Of course, every visual artist thinks their own work is good-looking. I don't feel like other people have to feel like that about Getting Over It. But when I was making it, I was like, "I can make this look nice, even though each of the individual assets is kind of horrible looking." And that became very interesting to me. 

Then we kind of kept doing it in Baby Steps just because, if nothing else, it allows you - one of the nice things about working in 3D is it's just like from thought to reality very, very quickly. You want to have somebody be like, "Well, we want a school bus in this game," which didn't actually make the cut. But you can have a school bus, pop it in the game, round trip you know 2, 3 minutes. You can see if that's a good idea. Oh, it turns out it's not a good idea. So we'll cut it. Right? It gives you fluidity. So, it does commit you to a certain type of set of aesthetics. You can't do this in a kind of a cartoon aesthetic, or a painterly aesthetic, or a 2D aesthetic, or anything like that. It commits you to kind of like low, mid-budget contract studio looking stuff. But it's still possible to make that look nice, I think. 

[0:49:18] JN: Yeah, absolutely. I mean, it's a beautiful game. Obviously, I mentioned as well your collaboration with Gabe and being a former student. I guess I just want to ask straight up, how did being a game designer alongside your game, your practice teaching game designs? How did those two things work out? It seems you've kicked game designs around with students. And obviously, students have gone on to be professional collaborators, and it's gone very well. 

[0:49:39] BF: Yeah, I mean I made a couple of games with AP Thompson. Made one for the iPhone called Stair: Slide the Blocks to Ascend, which I tried to put the tutorial in the title. And we made one called Multiball. And then, yeah, I should say I helped Gabe to finish Ape Out. And that went really well, and was really kind of happy with how that was going. We went on to make Baby Steps together. 

The thing that I miss about universities is that there is a kind of an intellectual and creative community that you kind of get for free. You get very interesting people coming to ask to do an independent study with you because they have some interesting project. And they show it to you, and you can just immediately sort of jump into a fascinating, and interesting, mutually beneficial situation. 

The class that I like the most was a class where students would make a different game every week for the whole semester. And you just get this kind of fire hose of interesting ideas. And it's not like I was stealing the students' ideas. Although I do think there is a statute of limitations on that. And if they haven't used an idea after five or six years, I feel I should be able to steal them. 

But no. I mean, I think that the sense of being part of a kind of live creative community where ideas are bubbling around, and there's kind of like an economy of ideas floating around, I really miss that. Yeah, I think it's tough. I mean, there's not too many - I can say this without sounding too shady. There's not too many academics who are actually publishing games at the same time. Of course, there are some very prolific ones. Paulo Pedercini down from Pittsburgh and CMU. And I think about Eric Zimmerman at NYU. And Frank Lantz also made Universal Paper Clips while he was at NYU. 

There is some exceptions to this rule, but the great majority of people teaching video game design are not publishing a lot of games. And I understand that because obviously teaching takes a lot of effort, and games takes a lot of effort. And it's difficult from a point of view of breaking your attention. One of the things people generally need to make video games is uninterrupted focus. And it's tough to do that when students are running into your office all the time or you have to kind of keep on putting things down to go and teach a class. 

But for me, even given that, it was still - I think the kind of balance of how it was working was that it was generative. It was kind of more inspiring and energizing than it was distracting and draining of time. But having said that, there's only certain types of things you can do. And I can do a game like Getting Over It in that context, which is small scope. It took me about a year alongside full-time teaching load. 

I couldn't do a thing like Baby Steps like that, right? There's just too, too much work. And I think one of the things people don't understand if they're not game developers is that the difference in the amount of work between those two types of projects is titanic. It's not like a 4x difference. It's like a 20x difference just in the amount of iteration that you have to do, how long it takes to kind of prop up each individual system. All of that stuff is just really kind of titanically different.

When people talk about there being Foddyan games, I don't even think that the kind of unifying thing is that they're rage games. I think it's that they are really, really small-scoped in a very particular way, right? It's like one level. It's like one system in one level. That's a way that you can keep it very small. So yeah, there's some things you can't do in the academy. That's why I'm not in there anymore. But I do miss that sense of community. And I think we will be kind of eternally looking for ways to get it back without being in the academy. 

[0:53:12] JN: Yeah, it's beautiful. Well, not that you can't have it, but the sentiment. But you did mention Multiball, which, funny enough, Multiball was the genesis of this episode. Because I only very recently heard of it and was immediately upset I couldn't play it. But obviously, I found out the reasons for that, etc. But then also, you seem to have been fairly aware of that in creating it. How was the experience, I guess, of creating this thing that would only ever be like exhibited and performed rather than released? 

[0:53:37] BF: For people who haven't played Multiball, which is most people, it's a kind of a pastiche of emulated games. I played something like 1500 different two-player arcade computer console games using MAME. MAME is an emulator that now emulates basically every system from history. So, I was really able to kind of go through just a ton of them. And I winnowed that down to about 300, 310 games, where I would capture a moment in that game as a save state, like somebody's about to score a goal, somebody's about to shoot somebody, somebody's about to win a point. 

And then with the help of AP Thompson, we wove them together into a kind of like a randomized gauntlet. So, I'm now playing against you. It's going to drop us into an exciting moment in a game. And we're going to play 10 seconds until somebody scores a point, and then it's going to drop us into another game. This is like the WarioWare of multiplayer games in a way, but it's also like a trip through history in a way that WarioWare isn't. 

And I had that idea bouncing around for years. And it wasn't until I had a conversation with AP about it, where he seemed to think it wouldn't be that hard to do the technical side, at least for him. He's a very, very gifted engineer. And MAME is actually set up to do this kind of thing because it has like built-in debugging tools. It has like automated debugging stuff built in. 

So what I need to do then was I needed to play all these games, find these save states. That's the easy part. And then go into the debugger and try to figure out which variables change when you score a goal. Which variables change when you touch the ground, when you shoot somebody, whatever it is? 

And I had never done anything like this before. But as I got more and more into it, I went into a sort of like a weird, obsessive state that I don't think I've ever been in before when I've been making games. I got just fixated every day. I'd be like, "Ah, I'm going to do five more of these. And I would be just trying desperately to kind of - sometimes if it's like a Commodore 64 game, there's variables that are hidden in the audio CPU, and instead some little pocket of RAM that's not supposed to be used. 

Or sometimes with the really old, like the Intellivision games, it's like the score is not actually stored anywhere. It's kind of like encoded in the graphics memory that's used to display the score. And there's all these little tricks that you discover. It's this kind of fascinating detective game itself, just kind of figuring out how to detect when somebody has scored a point in an emulator. 

Yeah, we made this experience that you can kind of play as a kind of meta multiplayer game. But you can't really release it, of course, because it depends. Even the save states have a lot of copyrighted code in them, and so on. And the ROMs that it depends on, the ROM files are also - obviously, you can't distribute. We settled for being able to show it at parties and in other kind of reasonable fair use situations. And it's one of my favorite things that I've ever worked on, honestly. It's a very unusual experience. 

And the reason that I love it is that it gets to the heart of what I like as a player, which is the moment when you're trying to figure out, "Oh, what does this joystick do? What does this button do? What am I doing right here? How do I play this game?" It's that little moment of not being sure how to do it. That's what I love. 

And in Multiball, they're just keeping you in that going like again and again and again, like, "Hey, here's a new experience. Here's a new experience." You figure out who you are for starters. Figure out who you're controlling. Figure out how the controller works. Figure out what the rules are. I love that sense. 

[0:57:05] JN: Yeah. And all that in a classic multiplayer scenario. Head-to-head as well, right? Yeah. 

[0:57:10] BF: Yeah. Because the other person is trying to do it at the same time, and you're trying to just do it a little bit faster than them. I love that feeling of being at sea. And I think Multiball kind of produces that feeling better than anything else that I've made. 

[0:57:23] JN: Yeah, it sounds fantastic. Yeah, great idea as well. And also, yeah, absolutely. We've had quite a few emulator developers, or people have ported to emulators on here, and it's like what you were just hitting on about the detective games. Some of the things we've heard are just like, as a programmer, you're just like I don't know how anyone invented that. That's absurd. So I can imagine that being very addicting. 

[0:57:39] BF: Yeah. 

[0:57:40] JN: While we're talking about, I guess, kind of - I don't want to say low-level, but different kinds of systems. We did mention Zipper earlier. So, Zipper is a game for Playdate, which is a show favorite here. We've had them on as well. And Zipper was one of the season one games that I must I didn't get very far in because it did completely bother me. Tell us about Zipper before I ask you, I guess, the main question. 

[0:57:59] BF: That's definitely outside of my wheelhouse. It started as a Flash game back in 2008 or so. I had made a little prototype in Flash, which same art style, same kind of visual aesthetic. But making turn-based games is a totally different skill set than making physics games. And I just didn't have the skills. And so I kind of shelved that till about 2015. Cable from Panic pitched me making a game for this thing, which he said would be out in a year, which was not the case. But yeah, it seemed like a kind of a fun thing to do as a lark, and it seemed like the perfect time and place to resurrect the little turn-based samurai game that I had come up with. 

But yeah, I mean, to me that was interesting mainly as coming as a novice into what is effectively board game design and grid-based tactics game design. And to try to bring my own sensibility to that, which I did by creating kind of an input scheme that's unobvious and a kind of traversal scheme that's unobvious. 

In Zipper, you kill enemies by dashing past them or up to them in the blink of an eye. In a way, I have to confess, is inspired by the anime Bleach, which is not a - I can't stand behind Bleach. But I do love the aesthetics of the combat in that. It's like a hyper exaggeration of a Kurosawa movie. And I just was trying to kind of like, in a turn-based format, get this sense of speed, like blink of an eye movement. And I think it came out okay. I still don't consider myself to be a good designer of tactics games, though. 

[0:59:35] JN: Yeah, I mean, I loved it. I feel it was very hard. But yeah, I thought it was really good. And it actually took me many years to realize that you had made it as well, which is a problem I have a lot with a lot of Playdate games. I think you succeeded in stepping into a different genres. But I guess what I want to ask you is you did Flash, which obviously is ActionSript. And I believe Baby Steps and Getting Over It, Unity? 

[0:59:54] BF: Yep. 

[0:59:55] JN: How was the experience of developing for the Playdate? 

[0:59:58] BF: I actually loved it. I mean, the Playdate is kind of a minimal API and Lua, which makes it a lot like making Pico-8 games, which I also love. What I love about it is I love not having anything provided. What do I mean by that? Well, there's something that's kind of aesthetically gross when you are making a game that just needs very, very kind of like minor. Let's say I'm making a Pico-8 game and I decide I need a particle system. And my particle system, it can really just be a table of XY coordinates. And I can just draw a pixel at those XY coordinates and update. Maybe the table has velocities on them. Maybe the table has lifetimes on them. It's just an array. And the code I need to traverse that array and move the particles around is like three lines long. 

So if I compare that to working in Unity, where the particle system is probably a million lines of code, and it does like a zillion things, which I might want to use. But most of the time, I don't. There's a sense of just like inelegance to that or this kind of gross bloat that I can't escape because it wanted to be all things to all people. 

Now I'm not a good enough programmer to make a Unity-style particle system myself. But if it's 2D games, if it's something like the Playdate, or Pico-8, or whatever, even GameMaker, of course, it's not very complicated to do something like that. And so I love the sense of only having what I need. 

And real programmers hate Lua because it's slow. And to be honest, getting Zipper working at a reasonable speed on the Playdate wasn't easy because there's path-finding, and that it doesn't have any CPU. And Lua has a lot of weird gotchas in it. Depending if you traverse an array this way but not this way, suddenly, it's slow. And now you use too many globals, or not enough globals, or you got to try to move these globals inside this table, in your meta table, blah-blah-blah-blah-blah. That's as boring and nerdy and stupid as writing in C. And maybe I should have just written the whole thing in C. 

But I love Lua because it's so easy to iterate. And so much of what I consider to be good game design is about having a tight loop on having an idea and having it running. And when I was making my cricket game for the iPhone in 2008, it's a very, very early iPhone API, and it didn't really have anything. It's just OpenGL and Objective-C. But in Objective-C in particular, if I want to add anything, I now have to make a separate class and an interface to the class. And I have to do all of this. 

And so just the smallest idea, taking an hour to be able to get it to happen. Whereas in Lua, "Oh, my particle class doesn't have a field that I want. I'm going to access it as though it exists. And now it exists." And this is such messy, bad tech debt-generating stuff. But for small games, tech debt is not really a problem. 

In fact, it's so quick to work on the Playdate, so quick to work in Pico-8, that if you start to feel you've got tech debt, you just start again and reimplement from scratch, and the tech debt is all gone away. So yeah, I love Lua. I love JavaScript. I love these high-level things. I don't love having to optimize them when it turns out that they're too slow. That sucks. But if you're working in the kind of turn-based land or in kind of very simple stuff, then generally computers are fast, and don't do anything too silly, then can really work in a very free-flowing way in those systems. 

And so Playdate, yeah, knowing exactly this is the hardware target, this is how fast it's going to be, this is how much memory you have, all of that stuff. There's something kind of very freeing about it, right? Just really takes a lot of the kind of shit work out of game development. And I did enjoy working on that device a lot. 

[1:03:49] JN: Yeah, for sure. I really like that. Yeah, Pico-8 is another great callout for that. Yeah, just constraints. Again, constraints are good. We like constraints. All right, I've kept you far too long. So, my last question, and you are free to turn down this question, but you have said fairly recently in response to the character design in Baby Steps, that working on character design animation brings you over into liking big butts, and that you, and I quote, "Had a lot of evidence for this." Please present your case before the jury. 

[1:04:15] BF: Well, it's hard to do that without visual evidence, but I would invite people to look at, for example, old Bugs Bunny cartoons. Bugs Bunny does a lot of flashing his bottom to the camera. There's a lot of looking over his shoulder when he doesn't need to be looking over his shoulder. There's a lot of turning around. And then as he turns around, his bottom is like pointing at the camera for much longer than it ought to be. 

And once you notice it in Bugs Bunny, you'll start to notice it in Mickey Mouse. You'll start to notice it in anime. You'll notice it in basically everything that you look at. 2D animation, 3D animation, you start to notice that the gaze of the animator is much more on the glutes of the characters than you think it might need to be, or that it might be if you were using a camera and kind of live action. 

I don't know why this is true, but it is kind of true. And I think it's particularly bad for game animators and character designers because we spend so long - we have the character in T pose or an A pose, and we sort of rotating around them and adjusting them. And then I just think naturally there's a kind of a natural focus towards their butt. I guess part of it is that it's an affordance of the technology, is that the computer's 3D graphics cards are very good at representing bottoms in a way that they're not necessarily for other targets of the sexual gaze. 

And so I would just invite anybody who's skeptical about that claim, start looking at other video games. Start looking where the camera points in the 3D, over-the-shoulder games that you're playing. Start looking at what is more lovingly animated and where the triangles have been put. All of that sort of stuff. It's definitely not just me, although I didn't restrain myself particularly much. 

[1:06:12] JN: I'm thoroughly convinced. All right, Bennett, thank you so much. This has been delightful. Yeah, thank you for joining us. 

[1:06:18] BF: Thanks for having me.

[END]