EPISODE

[INTRODUCTION]

[0:00:00] Announcer: Zachtronics is a legendary independent game studio known for creating intricate engineering-focused puzzle games that merge logic, creativity, and code. The studio was founded by Zach Barth in 2011, and it has become a cult favorite among programmers and tinkerers alike with titles such as SpaceChem, Infinifactory, TIS-100, and Shenzhen I/O., Most recently, Zachtronics released Kaizen: A Factory Story, in which players take on the role of an American engineer hired by a Japanese manufacturing company in the 1980s to design assembly processes for various products. Zack Barth joins the podcast with Joe Nash to talk about the games he makes. 

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]

[0:01:18] JN: Welcome to Software Engineering Daily. I'm your host for today's episode, Joe Nash. And today I'm joined by Zack Barth, who has in the past created beloved puzzle games such as Opus Magnum, Shenzhen I/O, Exapunks under the label of Zachtronics, and the recently released Kaizen under Coincidence Games. Zach, welcome to the show. How are you doing today? 

[0:01:36] ZB: I'm great. It's kind of early to be honest, but I'm ready to go. Let's talk about software. 

[0:01:41] JN: Yes, absolutely. Sorry. The joys of cross-continental podcast scheduling, but thank you for making it. I guess we mentioned a bunch of games there in the intro, but for folks who are unfamiliar with your games, can you explain, I guess, what your whole deal is? There's a genre named after you. What is a Zachlike? Let's start there. 

[0:01:55] ZB: Oh, god. Yeah, there is a genre named after me. I did not coin the term, I need to point out. I'm not crazy. I feel like it'd be really weird to name a genre after yourself. I actually pushed back on it really hard. But I also named our studio Zachtronics, which is kind of damning. 

Zachlike is - I still hate that term. 

[0:02:14] JN: I love it. I think it's great. 

[0:02:15] ZB: Oh, god. Somebody from Rock Paper Shotgun did it. I swear, not me. It's an open-ended puzzle game where you're given a set of tools and you're given usually a criteria. So, not like a thing you have to build, but like criteria for what an acceptable thing would look like. Like in the example, I think the overt programming games like Shenzhen I/O are a lot more obvious about this, where it gives you a test spec. And so you have to write software that satisfies a test spec. 

And in fact, there's like a hundred different random variations. So, it's almost more like elite code or something. But I think to a programmer, they're very recognizable and that they're like programming puzzles basically. You can write any code you want as long as it solves the problem and meets the spec. They're not all literally about programming, although some of them are, but they all take the form of that. They're these open-ended puzzles where every puzzle has kind of like a programming puzzle or a programming problem. Every puzzle has a huge number of solutions. And you can just find a solution that works, and then you can go beyond that to find a solution that is - often we have these sort of secondary optimization metrics. It's like maybe you're not just solving the problem, but you're trying to solve it in fewer lines of code than anybody else. Which, again, to anybody who's familiar with programming problems and LeetCode kind of things, it's totally familiar and a thing that lots of programmers do all the time and don't even think of it as a game necessarily. 

[0:03:30] JN: Yeah, that's definitely the element that I think it was TIS-100 came out during my first year of university when we were learning assembly and completely one-shot me for that reason. As you say, I think most programmers find the optimization bit especially very addictive. But before we get into it, I want to come back to the metrics and stuff, but before we get into that, I guess to round that out. 

Zachlikes. And you said you had this studio called Zachtronics. That was, I guess, your previous incarnation. You've recently moved over to Coincidence Games. Can you give us kind of a pocket history of, I guess, the game development journey and where you're up to today? 

[0:04:01] ZB: Okay. Right now it's 2025. I graduated college in 2008. And so that's quite a few years ago now, almost 20 years ago. I did a lot of game stuff in college, but for some reason, at the time, I was - I worked at a game studio. I interned at a game studio my junior year of college. And it was not a bad game studio, but not an interesting game studio. They did mod. They made - I don't want to shit on them too much. They kind of divided their time between stuff for the government and DLC Packs for AAA games. 

And so the people on one part of the office were making some kind of military training tool for teaching people how to use anti-tank missile launchers. And then another group of people were making the DLC for Command and Conquer 3, Kane's Wrath, or whatever. And so this was what was going on at the studio. And I was working on this game that was funded by a shadowy group of American businessmen who wanted to - this is like 2007. This is like peak war on terror, who wanted to spread American values abroad. And so we were making a Western values-laden game for Middle Eastern audiences. Yeah, it was crazy. 

The people who ran the company went on a business trip to Saudi Arabia and came back with all these games that they were selling in Saudi Arabia. We could see kind of what's the local market like. And that we were trying to make a game that was funded by these - God knows who these American businessmen were. But it always sounded so - 

[0:05:28] JN: Radio Free America for games. Yeah, very interesting. 

[0:05:32] ZB: Well, except it was like private people doing it, not the state department or whatever. I don't know. It was weird. The game never came out. I just worked on it early on. But it was kind of a boring project to be on. And so my takeaway from all of that - I like the people I worked with a lot, but my takeaway from that was just all programming is the same, right? It doesn't matter if you're programming a game or programming Microsoft Office, which is where I ended up, actually, out of college, and why I live in Seattle. But I was like, "Oh, all programming is the same. It doesn't matter." 

And to some degree, I actually kind of think that's true. But the thing that I realized over the years is that it's like, "Oh, yeah." Because I don't really like programming all that much. I do like programming, but I don't like programming. And professional programming sucks. And the thing I really like is design. Game design and product design, and stuff like that. 

And so when I was at the game studio, I was trying to do more game design stuff because I'd been designing games of my own. I wanted to design more games. And then when I eventually got to Microsoft, I tried multiple times. I interviewed to be a PM. I tried switching two times to be a PM. And they're like, "No, you're very much a developer." And it's like, "I am. I guess I am." I never understood what it took to be a PM at a big software company. 

But I actually think that what was a weakness in that context of only being like a mere programmer with ambitions to do design I think ended up becoming the backbone of everything I've done in my career since, which is that I can walk the line and I can design games about programming. Arguably, some of the best-in-class games about programming. And so it's like a weird intersection of my weaknesses in other contexts became my strength, and Zachtronics and stuff like that. 

[0:07:07] JN: Right. Yeah, that makes total sense. 

[0:07:09] ZB: Oh, we didn't talk at all about Zachtronics. 

[0:07:11] JN: No, no, no. It's okay. I'll get there. I guess the last bit you meant, that was Microsoft. You were trying to become PM. You're a developer. How do you get from there to Zachtronics? 

[0:07:19] ZB: Yeah. I worked on Visio. So, after I left, they kind of canned Visio. I don't even know if you can still buy it anymore, but that's the diagramming program. 

[0:07:27] JN: Yeah. Yeah. Yeah. 

[0:07:28] ZB: You can make UML diagrams with it, which, back in 2008, was a thing that people talked about and took seriously. 

[0:07:34] JN: I do feel like I've seen it under a dropdown in Office 365 recently. I think it is still floating around. 

[0:07:39] ZB: Oh, that's cute. Yeah, it's actually a really cool product. It looks like a drawing product. But behind the scenes, it's a spreadsheet. And this is what powers all their like little shapes that you can resize. And it totally changes the shape and scales it and stuff. It's all driven by a spreadsheet that has formulas that describe the locations of the points. It's actually really cool. 

[0:07:58] JN: That's really interesting. It comes to like this whole - there's this programming community called Future of Code. But they're really obsessed with like how the spreadsheet as like a programming paradigm and how that's kind of a wasted opportunity. And that sounds like a good example of where it could have been. Okay, but also hearing you worked on Visio makes a lot of sense with a lot of your games. 

[0:08:15] ZB: Yeah. Yeah. Right? And the funny thing is that I didn't even choose to be put on Visio. I just told them that I didn't care. And then they're like, "Cool, you're going to go work on Office." Because I was really unambitious coming out of college. I wanted to make not games but physical simulators. I'm like, "Oh, I want to work on simulators and immersive games. I was really into immersive physical gaming and stuff, and the idea of VR before Palmer Luckey came and rebooted VR. And I was really into that kind of stuff. And so there's nobody who hires for that. 

And so when I was graduating from college, I pretty much just got an offer from Micros - I had an offer for some other software company that I didn't take, but I got an offer from Microsoft, and I was like, "Well, I guess I'm moving to Seattle because I don't have anything else going on for me." And they were like, "Oh, you can work on Visio, or you can work on the team on Office that does the licensing and like serial key validation." And it's just like, "Oh, god, I don't want to work on that. There's no -"

[0:09:08] JN: Yeah. Visio has interesting problems. Yeah. 

[0:09:09] ZB: Yeah. That's good if you really like programming, but not design user interfacing stuff. And so Visio, actually, I guess in a roundabout way, was good for the kind of stuff I wanted to do because it's like a very weird bespoke CAD program, which is what many Zachtronics games are weird bespoke CAD programs. So I never really thought about that, but there is kind of a through line there. 

Yeah. As soon as I got to Microsoft, I was like, "I hate this. I need to go work on something that's actually exciting." You couldn't switch teams for like a year. As soon as I hit the year point of being a year in, I was like, "I need to find another team." And so I spent two and a half years trying to find. I spent a while looking for another team. And I managed to find a games team that was starting up, and I managed to interview for that. And so I managed to switch to this games team. I don't know, the two and a half. Or probably two and a half years into it, I switched to a games team. 

[0:09:56] JN: Okay, that's a good tenure. 

[0:09:57] ZB: But by then I'd already been working on SpaceChem because I'd already been making Flash games, and I'd already made Infiniminer, and I'd already done all these things on my own. Just working as an indie game developer, just kind of releasing games and giving away for free on my website. 

By the time I kind of managed to switch to a games team, I'd actually gathered kind of a large group of fans back then for just like a random person making free games. And so, yeah, I did Flash games. I did a game called Ruck Engineer 2, which was my first game that kind of got an audience. It's like a reverse engineering hacking game. I submitted it to Hackaday, and that was like a big - I mean, I guess it's still kind of a big site. But it was a big deal back then, and it got a whole bunch of people drawn to my stuff. 

And yeah, then I made Infiniminer, which was the inspiration for Minecraft. I made a bunch of those games. And when I switched games to the games team at Microsoft, we were just wrapping up SpaceChem, which is a game that I'd actually started building with one of the guys I worked with on Visio. On the side, we worked on it. And what became a growing team of other people, an artist in the Philippines, and another guy I was friends with from college who came to work at Microsoft as an intern. 

And we built SpaceChem. We launched SpaceChem. It did well enough. Steam accepted it. That was back when it was really hard to get on Steam, but it was accepted onto Steam. And at that point, everything was going so well at SpaceChem, our first commercial title, that most of us at Microsoft left Microsoft and then started Zachtronics full-time. 

[0:11:15] JN: Perfect. Awesome. And so you mentioned VR there and how hard it was to hire that. Joining Valve and working on VR stuff was after you started Zachtronics? 

[0:11:24] ZB: Yeah. I left Microsoft. I graduated from college, joined Microsoft 2008. 2011, SpaceChem comes out at the very beginning of the year. And so a couple of months into that, I leave Microsoft. We start Zachtronics full-time. I did that. We made Ironclad Tactics. We made a bunch of educational games for this company called Amplify. They used to be called Wireless Generation. And funny thing, when they were called Wireless Generation, they were actually the other company that I had a job offer from out of college. 

[0:11:51] JN: You're making games then. 

[0:11:52] ZB: I didn't want to move to New York City because I was like, "I'm not really accustomed to living in cities." It seems super intimidating. And Microsoft is in the suburbs. So that's why I went to work at Microsoft. But there's an alternate universe before I moved to Brooklyn and - 

[0:12:05] JN: And stayed in educational games. 

[0:12:05] ZB: Yeah. Well, they weren't even doing games then. Honestly, they were doing weird edtech stuff with like Palm Pilots. Yeah. It was a really interesting - well, I say really interesting. It was a product. It was all based on Palm Pilots. And you give the teacher a Palm Pilot, and then they go around the classroom while kids are working, and then the Palm Pilot has a question on it. It says like, "Go ask this child this question." And so the teacher goes up, asks them this question on the Palm Pilot. They do it on paper. They write down what the answer the student gave, and they do this. 

And so it's like a machine-targeted intervention where the machine is telling them which students to talk to and which problems to give them. And they're using some sort of giant early machine learning big data backend that has all the students and all their answers and knows who to target with what interventions. It was this Palm Pilot-driven, teacher-driven intervention where the teacher can - because this is like the problem with education is that you have one teacher teaching 30 people. And it doesn't really work all that well. You have to adopt a completely different set of things than when you're teaching somebody one-on-one. And so this was - well, we can tell the teacher who to go one-on-one help, and we can do it using big data to figure that out and be smart about it. And so that was their product. They did that. 

They're actually pretty big. There were a lot of people in that office when I interviewed there in college. And sometime between then and when we worked with them later, they'd been bought by Rupert Murdoch News Corp. Rupert Murdoch is just like, "Yeah, you guys look like you know what you're doing with education technology. I want you to make a new curriculum that blows all the other curriculums out of the water." The way it was described to me is that he realized he was old and wanted to leave a positive mark on the world, which - 

[0:13:39] JN: Cool. 

[0:13:41] ZB: Yeah. Yeah. He bought it off, and they made this huge curriculum project that required games. And that was actually how we came back to it is that they never became a games company. They've always been a curriculum company, but there was this little splinter group within there that wanted to do a bunch of games. And so while we were - I mean, this ties back into the early Zachtronics stuff. We made SpaceChem, we made Ironclad Tactics, we made these educational games for this company to put on their tablet that had a charger that caught fire once, and then they killed the entire initiative and switched to iPads. 

But, yeah. We did that. We did Infinifactory, we did TIS-100, we did all those games. And then I was just like, "Well, I'm completely burnt out." And so we shut down the studio, and I went to work at Valve. And I worked on VR at Valve. I would say it was a mix of really wanting to make a - Infinifactory, and TS-100, and SpaceChem did pretty well in retrospect. But at the time, it was like, "Yeah, they're doing okay." 

For most of my career, I've had this inescapable feeling that I need to be making games that do even better. Because it's really easy to look around and see all these people who are doing way better than we are. And it's like, "Well, if only we could make a game that was genuinely successful, we'd be set for life like all these other people." And every time I've tried to do that, it's just been ineffective, right? You can't just - I don't know. At least for me. I can't just be like, "I'm going to totally switch gears and be a different person and make something that'll make me successful." It's like, "No, I'm only successful in the sense that I did this thing that came naturally." And nobody would have ever thought it would be successful, but it was. And it was successful enough that I got to keep doing it. And so trying to make a game that was successful and it just completely broke me. 

And so this was when all the VR stuff was coming up. And like I said, I'd always been interested in physical games and VR stuff. And so for a real chance to do modern VR and not just having it be this like ghost of '90s VR that was all the hardware was inaccessible and everything was dead, it was really exciting. It's like, "Oh, VR's back. And this is my chance." And so, yeah, I got kind of FOMO-y and I went to work at Valve. And I was there for 10 months. I also couldn't work at Valve because - I don't know. But I I was at Valve for a little bit and I did work on VR. 

The early Valve VR product was called The Lab. And so, it's a bunch of mini games. And so there's a mini game in there called Xortex 26XX that I programmed and designed. And it was a fraught production. It turned into a totally different game. For most of it - Oh god. I don't even want to talk about it. It was so stupid. 

[0:15:57] JN: Well, that's very normal for Valve, though, right? They make things, and they change. As someone who's been trying to follow Deadlock, it's how they do things. 

[0:16:06] ZB: Yeah, it started off - God. I showed up there. They put me on the Counter-Strike team at first because they didn't put me - because you can go anywhere you want in the company, do whatever you want, because there's no managers. And so I let myself be put on the Counter-Strike team because they're like, "Oh, it's good to work on a team that ships stuff all the time." And I'm like, "Yeah, okay. I get it." 

But I got there, and I'm just like, "I got the Counter-Strike team." I didn't really know how to contribute. It was hard to tell what anybody was doing or how to contribute. And the people I was talking to, they assigned me like a mentor who was like a technical artist. And I'm not a technical artist. And so he was working on a new hand animation system. And then everybody was drawing stickers. Because the best way to make money at the time was selling stickers and selling battle passes for their little battle pass system they had back then. And it's just like, "Cool. I see nothing here that I can contribute to." 

And they had - I talk about this all the time. I don't know if I should. It was a long time ago, though. Meanwhile, they're making stickers and hand models, and they had a version of PUBG. I don't know what you call those kinds of games. 

[0:17:05] JN: Battle Royales. 

[0:17:06] ZB: Oh, yeah. They were working on a Battle Royale. And I was just like, "Hey guys, are we going to work on the Battle Royale? That looks awesome." And they're like, "Yeah, we're working on it a little bit." But they were not really working on it because it didn't make any money. And they were maybe a little bit more short-sighted back then with like stuff that - focusing on stuff that makes money. 

[0:17:21] JN: Yeah, that's fascinating. 

[0:17:23] ZB: 3 days in, I roll my desk downstairs to the VR team because I had a friend down there, and he's like, "You should come help us. You know Unity." I'm like, "I do know Unity." And that was the only place in the company where I could write C#, which is everything else there is like C++. And everything in Microsoft was C++ on Office. And I am really bad at C++. There's too many features in the language. I just never know what to do. I had been writing it in college, but it's just like there's just so many different ways to do stuff. There's so many different kinds of smart pointers. And it's just so complicated. 

And seemingly, every time I go to write C++, it was wrong when I tried to do it professionally. And I think other people would be like, "Oh, I want to learn more about C++." It's like, "No, I don't give a shit about C++." At that point, I'd already been using C# a bunch." I'm just like, "This is just like the best programming language ever made." It has all the stuff that's good about other languages without all the stuff that's bad about C++. 

And so I managed to work on - when I was on Visio, it was all in C++, but the new web-based stuff was in C. So I managed to kind of get to do some of the web stuff in C#. And then when I went to the games team, it was all in C#. And when I did Zachtronics, it was all in C#. It's all we've ever written is C#. And so when I went to Valve, I managed to find the one team writing C#. And it was the VR team, which I wanted to be working on anyway. So I was just like, "Fuck it. I'm not going to waste time pretending to learn how to do this." I know how Valve works. You do whatever you want. You do the right thing. And so I'm going to go where I'm useful, which is VR, writing C#. And so I did. 

[0:18:43] JN: Perfect. 

[0:18:44] ZB: And I get down there and I'm just like, "What am I doing? What are we working on?" And they're just like, "We're just building a bunch of stuff." And it's like, "Oh, I guess this is what they meant by like you shouldn't go work on a new team, because no one knows what's going on." And so I'm like, "I need something to work on. You guys have to give me something to work on to cut my teeth on." And they're like, "Oh yeah, there's this - one of the guys had been programming a remote-control helicopter simulator, because he was into like real-life remote control helicopters." And he's like, "You could work on that." I'm just like, "Okay. I don't know what that means, but sure." 

And so we started trying to build a game with this remote control helicopter thing, and it was terrible because remote control helicopters are really confusing because the controls are from the vehicle's point of view. And so it's actually really unintuitive, right? That's actually really hard. You can't just like put that in front of random people. They're going to struggle with that completely. 

And so I rewrote all the controls for the helicopter to be from your perspective as the person looking at it. And so we have the motion controls. You tilt it to your left, and it goes to your left. And so we built this. And then we built this whole game that was like a bullet hell shooter. You'd stand on the edge of this platform and like look into this huge chamber, and you'd have this like helicopter flying around in 3D, and then all this stuff would spawn and you'd shoot it. It was like a bullet hell shooter where you're in this chamber. 

And we just spent so much time on it. People still could not figure out the control scheme. For me, it was super intuitive. It felt, really, you could do these big, huge swoopy flying maneuvers. If you identified the thing and understood how to control it, it felt really free and exciting, and like remote controlling a bird or something. But for most people who played it, they just couldn't figure it out for the life of them. 

And we spent so much time trying to do it. They have a thing at Valve called Overwatch, confusingly, because it's like a Blizzard game also. But they have a thing at Valve called Overwatch where other people in the company who have been there a long time and kind of know what's up will review your product. And so you give it to them and then they try it and then they - you haul them into a conference room and they like pick your shit apart and just destroy you. And their feedback on Xortex was - they're like, "Did somebody from outside - this is like a contractor who made this. This doesn't feel like a Valve game at all." Because all the other things feel very Valvy. This feels like somebody outside the company made it. And I'm just like, "Yeah, I made it." And they're like, "Cool." 

Nobody liked it. Nobody liked the controls. And so it kind of became clear to me that it's like, "Oh, they're going to kill this unless I can make it so that they like it." Because other stuff in the package did get killed. I'm like, "Okay, I need to turn this into something that they're not going to kill." I tried a whole bunch of different alternate control schemes. I just spent two days just like angrily programming at my desk, trying out a bunch of things. And I made two alternatives, and then I just shoved them at the guy who was next to me, who was this guy, David, who was very nice and had been there for a very long time since like the Half-Life 2 days, I guess. 

And I'm just like, "Well, are any of these any good? I hate all of them." And he tried them. And the one that he liked was the one where it's just like I just glued the ship to your hand. And I'm just like, "What if we scaled you down? So rather than the chamber you're in being huge, it's just really small? And then it's like your size. And then you are - you're holding a toy spaceship and you're going, "Pew-pew-pew-pew-pew." He's like, "No, this is really good." I'm like, "No. You're full of shit. This sucks." He's like, "No. This is really good. This is better than before." And so I'm like, "Whatever. I don't even care at this point." And so I just did it. And everybody's like, "Oh, yeah. This is really good." I'm just like, "Oh, god. Whatever." And so that was what we shipped. And was it better? I have no idea. I don't care. Right? 

After that, what became the Half-Life Alex team. After we shipped The Lab, the Half-Life Alex team came through and stole half the people. Other people who were left over had no idea what they're doing, and I'm just like, "I got to get out of here." There was no project that I wanted to work on there. It was at a weird point at Valve where people were really protective of what their projects, because they didn't want too many people to join their projects. Because then if there's too many people in the project, now there's a lot of eyeballs on it. And if you're trying to figure out what your game is, you don't want that. 

And so there was like seemingly nothing for me to work on. And so I'm just like, "You know, really just want to -" while I was working on VR, I spent a lot of time with the hardware guys because I was working on some platforming stuff, too. And I've been always drawn to hardware. When I was in college, I did a lot of stuff with electronics, but I find it really frustrating because software just gives you such a bitter feedback loop. 

[0:22:55] JN: And you can't break it. It doesn't explode. 

[0:22:57] ZB: Yeah. Well, I'm not even worried about that. There's just like I'm - well, I guess so. I'm always like afraid. It's just like, "Oh god, what if I accidentally short something, it catches fire? Or how do I know what kind of capacitor to put on this?" Electronics has a lot of these heuristics that I could never get a straight answer about. And I really can't deal with that ambiguity. And I think it's less ambiguous when you're actually in practice, but it still feels like it's more of like an art than a science, electronics, even though it's clearly a science. But I don't know. Just software so much resonated with me so much more. My desire for everything to make sense and be perfect and understandable is really like a software thing. 

And so I didn't go into electronics. I couldn't have done it. But I was still interested to it. And so when I was at Valve, I was hanging out with the hardware people. And it's like, "God. This is so cool." They're manufacturing stuff and learning about all of them. I'm like, "I want to make a game about embedded programming." I've done embedded programming for my whole hobby since college. Microcontroller stuff is pretty accessible. And with Arduino, it's become even more accessible. 

I do a lot of that in real life as a hobby. And I was like, "I need to make a game about electronics manufacturing." And so I did. And that became - when we sold. I got out of Valve by selling Zachtronics to a big company in New York that did games distribution called Alliance. Yeah. 

And so one of the big problems we had at Zachtronics when we owned it ourselves and ran it is that we were always one game away from going out of business. And so it made me do stupid things like, "Oh, we have to make a game that'll make a lot of money." Right? Because I'm like worried about it. That's a thing you think about when you try to think about money too much, which I think is not helpful. And it was too much. I felt like it was kind of constraining our creative process, having to worry about making money too much. 

And so I went to Valve. And when I was getting out of Valve, I wanted to find something that would - I wanted to find somebody who would just give us a lot of money so that we wouldn't have to worry about it. And through somebody we knew from the games team at Amplify, that educational company, they actually were working - they had gone from working at Amplify to working at Alliance as a talent scout. And so it's like, "Oh, I know a talent scout at a games company." You know? 

And so we talked to them. And we met at GDC. And we managed to sell Zachtronics to them so we could go work for them and have this huge company that did physical games distribution, kind of as a flywheel. That even if we fuck up and ship a game that doesn't make a lot of money, they just keep on spinning, and then everything keeps going. Because they're pulling in like 50 million a year, right? It's like we're much smaller than that. That will keep us going when we mess up and make a bad game. So we sold to them. 

And that was how Shenzhen I/O, Opus Magnum, Exapunks, Mobius Front, Eliza, Last Call BBS all came to be was all working for them. That was over the course of - that would have been in like 20 - I went to work at Valve and SpaceChem, 2011. That's when we started Zachtronics. We shut it down in 2015. 2016 is when I go work at Valve. 2016 is also when I leave Valve and go work for Alliance. 2017 through 2022, I would say, was when we were at Alliance. Only five years. But we managed to ship - I don't even know how many games I named in there. Six? Six games? 

[0:26:04] JN: Yeah. 

[0:26:05] ZB: In six years. 

[0:26:06] JN: Yeah. And that's that was one thing I want to ask about. I recalled you mentioning somewhere that that was very much part of the deal, that you had to ship on like a regular cadence. And you had a year to develop each game. Is that the case? Am I remembering that correctly? 

[0:26:18] ZB: We didn't have to do anything. Honestly, there was very little oversight of what we did. I think things kind of fell apart a little bit at the end. Their business was physical games distribution. And their biggest customer was Toys R Us. And so when Toys R Us went out of business, it took their physical games distribution with - that also killed the distribution business. And so what started off as this like $50 million, 50-person company, it became us and the CEO and his son. And so it became a much smaller company. 

They'd picked up another indie studio at some point in there, but then they'd shut it down. And so everything got really small, and it just got kind of awkward. And just the flywheel stopped spinning. Yeah. And so we never had to do a game every year. But that was just - I think we were always very responsible. For us, it was like, "Okay, we can make about a million bucks off a game on average." We spend about a million bucks a year on the five of us, and our office, and our health insurance, and our taxes that you have to pay. If we do about a game a year, we can keep this going indefinitely, right? 

And so obviously, indefinitely, it was about 5 years long when we all started getting burnt out. Again, feeling like things started kind of closing in on us. Can we even experiment anymore? And we had done experiments in the form of Mobius front and Eliza. Neither one of them was very commercially viable. And I think with the walls closing in, it's like, "Okay, we can't really swap up who we work with." Nobody wanted to like leave. And I wasn't about to like fire anybody just to get some new blood in there. We're just stuck, you know? 

And so we kind of realized, it's like, "Hey, I wanted -" well, it's complicated, too, because I was getting kind of burnt out again. And I was getting really interested in something I had always thought about as my backup career, which was teaching. And so for years I'd been saying, "Oh, if I ever burn out and quit the games industry, I want to be a teacher. That seems cool. Our games are kind of educational. And I think I would like that." 

And so the plan was kind of to wrap up Zachtronics, get my teaching degree, and become a teacher. But things ended up happening in a weird order, where COVID happened. And then when I started getting my teaching degree, I realized that you don't need one if you're teaching computer science, and they just let me in right away. And so I actually spent my whole year teaching while we were also finishing up Last Call, which was our last game. And so I ended up what was supposed to be leaving Zachtronics, starting teaching, turned into Zachtronics and teaching both wrapping up at the same time at the end of 2022 or whatever, because I turned out I didn't like teaching at all and was not cut out for it. Yeah. And so I just sort of found myself at the end of that year with nothing going on, right? Zachtronics was over and pretty dead. 

I mean, I still maintain all the games and stuff, the idea, the possibility of any new work. Sort of at the last minute, I'm like, "Oh, what if we kind of like stuck around and made some more games just with a smaller team?" And they're like, "You could do that." I'm like, "But what about with the other programmer?" They're like, "No. We're not going to pay for him." And I'm just like, "Oh, this is over." It's like if I can't have - I've been working with this guy, Keith, for my whole career. And I'm like, "If I can't have Keith working with me, what's the point?" And so that was when I realized that the stuff with Alliance was like, "We're just going to maintain the old games. This is kind of over." Yeah. 

And so I sort of found myself - as soon as I got out of school, I had hernia surgery. I'm recovering from that. I'm just like, "Cool. I'm just down and out and have nothing going on." And that was when we started Coincidence. And so we had planned it out. So, I guess I actually knew exactly what I was doing coming off of school and starting Coincidence, because we were trying to move into our new office for Coincidence while I was recovering. My abs didn't work. 

Yeah, as we were wrapping up Zachtronics and stuff, some of the other people on the team wanted to start working on a project of their own. Not without me, without me, but like not with me, you know? Because I was going to teach and stuff. They wanted to do a project. I'm like, "Okay, if you guys are going to do a project, who's going to own it? What's the company structure?" And they're just like, "Uh -" I'm just like, "No, we need like a company structure. You can't just do this willy-nilly." 

And so I designed this company structure that would allow them to do the thing they wanted to do. And then when I realized that I needed to start doing stuff again and start making games again to make money, that's when it really came together. I was like, "Okay, we'll build this company that's like a co-op, but nobody gets paid a salary. Everybody just gets paid with what they work on. We can run our contract work through it. We can make games through it. The company can own the games and sell them, but we all own the company. It's like ownership is distributed, but also not - we all own the stuff, but it's owned by a single entity. So it's legally allowed to be sold without conflict." Yeah. And that's Coincidence. It's it's a normal company except that no one gets paid salaries. And I keep a spreadsheet, basically a ledger of all the money coming in and all the money going out and who it goes to. 

[0:30:47] JN: This is a very recognizable org structure that sounds really cool. 

[0:30:50] ZB: Yeah. It's kind of uncommon in games, to be honest. Games require a lot of seed money. And I think a lot of people, a lot of - 

[0:30:58] JN: Long development times normally. I guess, for you guys, it's quite short. 

[0:31:02] ZB: Yeah, yeah. Long development times. And there's often a lot of unknowns. And I don't want to say that game developers are like bad software developers. But I think a lot of times, indie projects have a lot of unknowns and just kind of messy logistics and stuff. And so, yeah, I don't know. People keep telling me it's kind of uncommon, and I don't hear about a lot of other people doing the same thing. I think we're special. But it's totally just like a normal - if you were just going to like run a software consulting company, that's exactly what you would do, right? I think to programmers, it seems really normal. To games people, it seems really weird. 

[0:31:33] JN: Yeah. To stay on, I guess, Coincidence's structure in the games industry a bit. We will talk about game shortly. Talking about Zachtronics, and I guess some of the comments you made recently, you talk about the tension between paying the bills and being able to experiment, and that was like a big pit of what you just mentioned was kind of the end of Zachtronics. How does that then work in Coincidence? You sold Zachtronics. You didn't have to worry about the cash flow and the salary. You have that safety net. But now it's what you make from the games. That is it. How are you thinking about being able to experiment now? 

[0:32:04] ZB: Yeah. Well, so in some ways we're less able to experiment than we were before. So Kaizen was published. That's how. We've never had a publisher before. Even when Alliance owned us, we were still self-published. It was all funded off of - Zachtronics made money every year. So we were just kind of funded off of that. And when we started at Coincidence, we needed to start finding work. 

First thing, a couple of us went and got a job at a software startup, like an edtech startup. That lasted a couple months, until I got fired. I don't know if I should talk about all that, but it was messy, and it was unclear what I was there for. Yeah, it turns out remote companies can be super isolating, and it can be really hard to know what people are talking about or what anything is. 

And I'm not a fan of remote work in remote companies. We have an in-person office. We're all like physically in each other's presence often. And I think that's important, right? Because otherwise, stuff gets weird when people don't actually get to be together and bond as people. I don't know. It was a weird thing. 

[0:33:06] JN: No, I totally get that. Yeah. I had a manager at a remote job. Sorry. I had a manager at a remote job who said that like although we were all remote, he said it's much easier to be remote after you've at least met them. And you can hear the messages in their voice in your head. That makes everything easier. But if you never have that experience, it's - yeah, I don't know. 

[0:33:20] ZB: Yeah. Well, and sometimes hearing their voices in your head and their voice doesn't help, right? Yeah. It's tough. I would say remote work works really great if there's no conflict. 

[0:33:32] JN: Sure. 

[0:33:32] ZB: Right? But I think remote work, maybe if there's conflicts, which comes - I mean, I don't know. Sometimes software is boring, but sometimes software is heated, and there's lots of personal stuff. 

[0:33:41] JN: Especially in a startup. 

[0:33:41] ZB: Yeah especially in a startup, and especially like doing something that's kind of unknown. Games are weird. They're like a creative product, but they're also a business product. Software is often not creative, or it is like a little bit. I think games are just weird. And so I think it's really bad for resolving conflicts. And that's what happened with the thing with me is that it's like something that in-person with normal people could have been solved really quickly and really easily. And the remote work aspect of it and the isolation of the remoteness just kind of spiraled it into a very messy situation. 

[0:34:16] JN: Fair. That was part - I guess like a tangent from Coincidence. So now you all come back to Coincidence. 

[0:34:20] ZB: Well, that was the first thing we did at Coincidence. Because Coincidence allows for people to be contracting or whatever. It's a structure for anything. And so that's what we started off at. And then when I came back, and when I kind of tried to bring back the people that I had gone there to the startup with. And the first project we did that made us money was another educational game for Amplify. As it turns out, I convinced them that they wanted a game, and we made them a math racer game. And so that is - they were nice enough to let us release it on Steam, so that way it wouldn't just - 

[0:34:51] JN: Disappear into some proprietary hardware. Yeah. 

[0:34:53] ZB: Yeah. Exactly. We made those three games for them previous at Zachtronics, and then there's nowhere to play them. They sold them to a weird Irish multinational company that put some of them on the app store, but then those got pulled also. And so they just die. And I'm just like, "We just need to be able to sell this on Steam so we can just keep like a version of it for ourselves alive forever." That became Add Astra. 

The funny thing of that is everybody read into it way too much when talking about our studio and they're like, "Oh, they make educational games now." It's like, "No, no, we don't." We did that for money. It says on the Steam store page, "We made this for a client and they let us put it here." And people are just like - 

[0:35:28] JN: I remember you announcing that. And there was a lot of description about why and how you did that. 

[0:35:33] ZB: Yeah, exactly. No. People are like, "They make educational games now." And it's like, "It's because he was a teacher." And it's just like, "No, that is not. Because we needed money, because we were desperate, and they're good people." Yeah, they're a good customer. So we did that. And then that gave us some time. And then we found this publisher called Astrological. And no connection to Add Astra. And they were very excited to work with us. And they funded Kaizen just off of me waving my hands around, giving them a pitch. And so, yeah. They funded Kaizen. And we made Kaizen. And it came out. 

[0:36:04] JN: Perfect. 

[0:36:04] ZB: Yeah. 

[0:36:05] JN: Let's talk about the video game. What is Kaizen? 

[0:36:07] ZB: Kaizen is a - we should drum up some tales for Kaizen while we're doing this podcast. 

[0:36:11] JN: Yeah, absolutely. 

[0:36:15] ZB: If I was being totally honest, right, Kaizen is a Zachlike. A Zachtronics-style puzzle game. It's a lot like games we've made at Zachtronics. In fact, I would say, mechanically, it's a lot like what if you crossed Opus Magnum with Infinifactory.

[0:36:27] JN: Glad you said that. 

[0:36:28] ZB: We were tasked with making the most - literally, when they gave us the money, they're like, "You guys have to make a game that's really approachable." Even Opus Magnum is kind of intimidating for people. 

[0:36:38] JN: This is the publisher. 

[0:36:39] ZB: The publisher. Yeah. And this was their - and they've changed kind of who runs it. And their missions changed a bunch. They used to be like a puzzle, like a thinky games publisher or puzzle publisher. Now they're strategy games, which are a lot more tolerant, I'd say, of complexity than like tweet indie puzzlers. But the objective we were given when we started was to make the most approachable Zachtronics-style puzzle game. And so we did. 

And I'll be honest. I focus on the negatives more than the positives. And so we're sitting at about a 90% on Steam. And the 10% of people who thumbs down our game are almost entirely people who liked our old work and thought that this game was too easy, and too dumbed down, and too short. 

[0:37:15] JN: I can see them. These are the people still doing Opus Magnum competitions on the Discord. I know these people. 

[0:37:19] ZB: Oh, I love those people. 

[0:37:20] JN: I know exactly who they are. 

[0:37:23] ZB: The Opus Magnum fans are - I feel the people on our Discord are some of the ones who've been most open to Kaizen, weirdly. But it's the sort of casual hardcore Zachtronics fans. There was a tweet by Notch. That one, it really cut straight to me, is that he was just talking to some other game developers about how, "Yeah, Kaizen kind of sucks. It only took 5 hours to beat. Just was getting good, it ended." And it really cut straight to my heart. Not that I am friends with Notch. But - 

[0:37:49] JN: Yeah. That's rough. 

[0:37:50] ZB: There's a lot of people like that. And I totally get it. And I honestly kind of agree, right? And we've added like a bunch of bonus content to the game, some of which is much harder. But it's still fundamentally a game that was designed to be approachable. And so I'm talking about this as a weakness, right? Because I focus 10 times more on the negatives. In my heart, we're sitting at 50% on Steam because those 10% negatives are so much more strong than the positives. 

But I think the real less toxic version of interpretation of this is that I think we succeeded at actually kind of making a game that is a tutorial for all the other Zachtronics games, which is not helpful for me because I don't own all the other Zachtronics games. But I think we made a game that's kind of an introduction to those games that we really like sanded off so many of the things that were unnecessarily complicated. And maybe we sanded off some of the stuff that was necessarily complicated, also. But we really made a game that does away with a lot of the stuff that I think made it hard for people to pick up Opus Magnum. 

[0:38:44] JN: Interesting. What kind of stuff are you thinking of there? 

[0:38:47] ZB: In Opus Magnum, one of the big parts of the game, probably one of the strengths of the game, is that you build a machine that builds these molecules, but it does it in a way that's automatically pipelined. And so your program, we call it tessellation, right? Is that you build this little program, but your program just repeats like back to back to back to back to back. 

And so we slide them as close together as we can. So it creates this tessellating pattern of instructions. And the way that we show this in-game is with little lights that cycle and tell you which instructions are running. But it can be really hard when your program starts looping. And specifically, when your first instructions start running before your last instructions have finished running, people are like, "What is going on?" And it's like, "Well, it's tessellating. You need to use the time stretch instruction if you want." 

It's actually really hard to explain. I've been unable to explain it to people. It's kind of like - you just have to like get it. If you see it and you think about it, you'll probably get it. And I think a lot of people do, right? Opus Magnum was super popular, made millions and millions of dollars, which is really good for us, right? Well, really good for Zachtronics. But it did very well. And so I think people are clearly able to understand that system. But also, that is one of the things that kept people from picking it up and figuring it out. 

And so in Kaizen, you make one product at a time, right? And we still loop it. Visually, you get the satisfaction of watching it run over and over again. But sort of what happens in one iteration of your program has nothing to do with what happens in another iteration of your program. And it simplified a lot of stuff. It made it a lot easier to do another thing, which people complained about Opus Magnum, is that in Opus Magnum, you hit start, it runs. And then when something goes wrong, all you can do is hit stop. And then you have to remember like what went wrong and what do I need to change about my program. 

In Kaizen, you can just scrub back and forth and you can edit your program at any point in time. And so you can just like kind of keep working forwards where you run your program a little bit, you build the next step, you run it, you build it all your own, and then you can slide back and forth. And so it turns out this is actually kind of confusing in a way because a lot of people, I think, often think of times going forward and not also backwards. And so you need this sort of weird - what's the - like Slaughterhouse 5 kind of, loose sense of time and when things take place. People do struggle with that a little bit. 

But I think it opens up this thing where you don't have to remember the state of your machine. You just like look at the state of your machine. And honestly, I think that's maybe part of why people think the game is too easy, is that if you're good at the timeline and you're okay with time going backwards, it makes the game a lot easier. You don't have to remember where everything was when something crashed. And it makes it a lot faster to play the game. 

And so I think we succeeded on these little things. But it is a much easier game. The puzzles are much easier. The tool set is simpler. The tool set is smaller, right? It's just very straightforward game because we wanted it to be. And it's both the strength and the weakness. And so I think if I were to recommend games that I've made to anybody, Kaizen is absolutely the starting point, right? Without a doubt, it is the best, the most polished, the most approachable. It gets at the fun thing, right? It lets you build machines and optimize them and compare against your friends and stuff. Yeah, it's also the weakness of it, I guess. 

[0:41:52] JN: Yeah, I think it's very - you talk back on saying how it's a good entry to your past games. But I guess also, it's the first game - I know you've published some physical games and card games and stuff with Coincidence. But as the first video game from Coincidence, I guess it's also a good introduction to what you're going to put out next, presuming you're intending to keep working on Zachlikes. 

[0:42:11] ZB: Well, hypothetically, Coincidence could do any kind of game. But I can tell you that, yeah, I'm not working on anything other than Zachtronics-style puzzle games at this time. Yeah, we have a bunch teed up that we're working on. 

[0:42:23] JN: Cool. Awesome. 

[0:42:23] ZB: That's one of the challenges of working with a publisher is you're always wondering like, "Okay, cool. Well, we just spent all that money while we're making the game, so we're back to having no money. What do we do next?" Right? And Kaizen has not done well enough that we're making like mad royalties off of it, you know? So, it's been kind of a - I don't know. It was like our strongest launch for the first two weeks. And now it's kind of flattened off quite a bit. It's no Opus Magnum, that's for sure. 

[0:42:51] JN: Everyone's playing Silksong. 

[0:42:52] ZB: Yeah. Exactly. Right? And so not that Opus Magnum was in the news for very long either, but yeah. I don't know. We're trying to figure out right now exactly what are we working on next? Where's the money for it coming from? I have no idea. Yeah. 

[0:43:06] JN: Set Kaizen for a bit. So, speaking of difficulty, obviously, a lot of the difficulty in all your games isn't necessarily in the initial solve, but in the optimization and the graphs, which, as in all of your titles, have completely ruined my life. For folks who haven't encountered your metrics before, can you tell us a little bit about that system, and I guess how it drives the gameplay and how it's implemented? 

[0:43:25] ZB: Yeah, so with SpaceChem, SpaceChem was the first commercial game in this pattern, but I had made some Flash games and stuff before that. The codex of all chemical engineering was a Flash game that I had on Congregate. And Congregate has leaderboards or had leaderboards. I don't even know if you could still go to Congregate and play games anymore. Certainly not Flash games. So, they had leaderboards. And so I was just like, "I guess I'll hook up leaderboards to my game." And so, I ended up hooking it to - like when you solve a puzzle, how fast is it? When you solve a puzzle, how small is it? And I'll hook these up to leaderboards. 

And people on the forums on Congregate got pretty into it. And they started - one of the weird weaknesses and strengths of the codex of chemical engineering is that I knew how to save progress in Flash because the game keeps track of which puzzles you've beaten, but it doesn't save your solution. It instead gives you a chunk of base 64 text that you can copy into a text file. I have no idea why I did that. It doesn't make any sense. Clearly, I could have made it autosave. Maybe I thought people would want to like version multiple saves, but it's like I could have just added save slots. I don't know why I did it. It was a long time ago. I was just getting started programming games. I had never made web games before. Things are a little different in a web game. 

And so maybe I was afraid of people losing their solutions if they cleared their Flash cookies or something. I don't know. So I did it this other way that was weird. But the side effect of it is that people were able to share their solutions with each other on the forums and improve on each other's solutions. And so when we were working on SpaceChem, I'm like, "Okay, we need a way to allow people to compare their scores against other people's, like leaderboards. We need a way to allow people to share their solutions with other people in a way that lets them explain to each other as a technical person, "Here's what I built, here's how it works. And so we ended up adding export to YouTube functionality where it would record a video of your solution and then upload it to YouTube. You would type your YouTube credentials in like plain text into the game, and then we would like sign into YouTube in plain text, and then HTTP, and then upload a video. And so this is really a long time ago. We were really bad at programming. We did that. 

And then we knew that leaderboards were bad. And so as like people with a background in math, we're like, "Oh, we'll just do histograms instead of leaderboards." Because it's like we don't really need to like attribute the top hundred people who all copied the best solution, right? We can just show what's the distribution of scores. And so we did histograms in SpaceChem, and they were very successful. I would say it worked really, really well. 

Portal 2 briefly had histograms in it for scores because they were SpaceChem fans. Not that it made a lot of sense in that game. But ever since then, we've been doing histograms instead of global leaderboards in our games. We have this thing where you can solve a puzzle any number of ways, but we can add arbitrary metrics, like speed, or size, or instruction count or area. And if you start adding metrics, people can start optimizing degeneratively for those metrics, and then we can make that kind of become part of the gameplay. 

And so we end up shipping that on all of our games now just sort of by default. Because if we were to not do that, people will complain probably. And it's very cheap to add. I actually think it's less. It doesn't seem like it's as effective as it used to be. It used to be that people got really excited, and everybody talked about it. I sort of don't know if new players aren't responding to it in the same way, or if it's just that social media's kind of gone silent on a lot of topics. And so we just don't know how people are engaging with our games. 

[0:46:36] JN: Actually, I had this as a question. Because you spoke about the evolution of this feature on the Darknet podcast recently. And you spoke about its evolution over different social platforms. And so I was going to ask you. Social media has imploded. I don't know how it works anymore. What does it even mean now? And so I was interested in how that had affected this. And it sounds like it has. 

[0:46:54] ZB: I guess. It's sort of just like a large hole in my sense of the world, right? I don't even know if it's - I don't know. I have no idea. It's hard to tell what's going on. The other thing I was going to say, when we got on Steam, we added we actually got access to proper leaderboards for SpaceChem. And so we added friend leaderboards, which our intuition was that global leaderboards are garbage, but friend leaderboards are great because you know your friends. And if they're cheating, that's a friend problem. That's a you problem, not an us problem. And so a lot of this stuff is about decreasing, the attack surface for cheaters and stuff. So we just don't have to try to moderate that. 

All of the global leaderboards for all of our games, we eventually got them in the form of our fans keeping track of the best scores. And so they are way more rigorous about inspecting people's solutions, making sure that they're not using exploits. They can spot the exploits from like a mile away, where I would be like, "Ah, sure. It's fast. I have no idea." They maintain those leaderboards. And I don't know if maybe they would prefer that we did it, but I don't think I could do as good of a job as they do. I think it works out okay in the end. 

[0:47:58] JN: Awesome. 

[0:47:58] ZB: But yeah, jumping back for a second. Yeah, this was a thing that in Opus Magnum, everybody was on Twitter, for instance, talking about, "Oh, they love the competing and the gifts and stuff." And the gift export feature. Oh yeah, that was the other thing. With SpaceChem, we had upload to YouTube, which was, I guess, appropriate for the time. 

When we were working on Infinifactory, we were going to do export to YouTube, but I couldn't get a good angle. I was trying to do a thing where you pick an isometric angle and render it, but it didn't really look good. And so we didn't ship an export feature. And then one of our fans mocked up a looping first-person GIF of their factory running. And I took one look at it. I'm just like, "Holy shit. This is it, right? GIFs loop. Factories loop. We can isolate the portion of the factory and record those frames." And it's just I had within like a day or two, I had completely rewritten all of our YouTube code that we didn't ship to crap out GIFs. And it was just like, "This is amazing. I love this. This is the coolest feature ever." 

And so when we did Opus Magnum, we had that feature at launch, and it was in like a little branded frame that said Opus Magnum on it. And it just like blew up Twitter, right? And people, at one point, not - Opus Magnum came out 2017. I'm going to say a couple of years ago, like three, two years ago, Reddit changed something about their algorithm that caused Opus Magnum gifts to start popping up on people's feeds who had never heard of Opus Magnum. And this is from the Opus Magnum subreddit. 

So many people showed up, and they're just like, "What am I looking at? What is this game? Somebody explained this to me like I'm an idiot on Reddit." We had this huge influx of people who had just seen these GIFs. And it's like, "Yeah, they're really cool looking." Right? And that worked really well with Opus Magnum. And we did GIFs in Kaizen. And just like no one posts GIFs. No one cares. They're not as cool-looking as the Opus Magnum GIFs. The Opus Magnum GIFs have a lot of rotation. The pipeline repetition is very cool-looking. 

[0:49:51] JN: Yeah. I guess they're very abstract as well, right? You can kind of know what's going on in the Kaizen when you see a TV Pumbaa to GIF or whatever. Yeah. 

[0:49:59] ZB: Oh yeah. Yeah. It's just not the same. Yeah. I don't know. So it's always hard to tell. Was Opus Magnum just like truly the best game that I could ever possibly make? Or are things different with how stuff spreads now? I have no idea. Maybe we need upload to TikTok functionality on our games now. I have no idea. 

[0:50:14] JN: Probably. Yeah. With some automatically choosing the algorithmic song of the day. 

[0:50:19] ZB: Yeah. 

[0:50:19] JN: Speaking of the visuals, I realized that you said what the style of the game is. But one of the things I want to ask about that. It's set in 80s Japan manufacturing, which is a manufacturing. You said you've always been electronics, so it's not surprising. It's a common theme for you, Shenzhen I/O. You played with it. You played with it in a couple of things in Last Call BBS. You said this game is kind of like a mashup of Opus Magnum and Infinifactory. I was kind of interested in thematically. Did you arrive on this from trying to make a simple Opus Magnum, or was it you wanted to make another manufacturing game? Where did the theme come out? 

[0:50:48] ZB: I wanted to make a game about the theme. And then I made a game to go with the theme, which is the wrong way to do it. But like I don't know. At this point, it's very hard for me to just come up with like, "Oh, this set of mechanics would be cool in the abstract," because I've made so many games. I've used up all of my ideas for mechanics that came to me just as mechanics. And so the way that I make most of my games now is to think of like a scenario or like a story and then try to come up with, "Okay, what's like a cool mechanical or systems kind of - how do they do computing in this world?" And it's maybe not the best - the best way to make a game is to have like a really good idea and then make that game. And a lot of people do that for their first game. Is it's like something that they've been thinking about for a long time that they made? I've made so many games that I just need to be able to just like come up with game ideas constantly, and they have to be interesting. 

And so for me, the crush I lean on is trying to evoke a narrative thing. And so when we were at - Last Call BBS was a collection of little games. But before it was that we were trying to make a game about manufacturing in Japan in the 80s. And so it just seemed like it wasn't working, and it was too small and too boring. And so we ended up pivoting that into 20th Century Food Court, which was originally a game about manufacturing stuff in Japan in the 80s. 

[0:51:59] JN: This is actually because you said previously that Last Call BBS was a bunch of ideas that hadn't worked. And I was like, "No, there's loads of Kaizen in here between the robot making and the food court." Okay, that makes total sense. That's awesome. 

[0:52:10] ZB: Yeah. We ended up pivoting to this weird thing where it's like, "Oh, it doesn't take place in the 20th century. It takes place 800 years in the future, and they're trying to recreate - it's like a parody of the 20th century." And that was what we ended up settling on. And then we bundled it up with all these other games. But then after we shipped Last Call and a couple of years had passed, I kept finding myself thinking, "I wish we just made the game about Japan in the 80s. Because I'm really interested in the time period. I think we can really evoke it in an interesting way. 

In the office, we're all kind of intrigued by this. As somebody who like grew up in the 80s and 90s. I don't know. Just the idea of Japan was very intriguing, right? And obviously, like in real-life, it's, "No. It's just like a bunch of people who live in a place." You're being weird by like obsessing over it. But there's still this like vision of it and like watching movies from the 80s and stuff, and it just really evokes a thing. 

And so I don't know. Just wanting to make that game and wanting to like explore that story, and just focus on this time period. And people make historical TV shows all the time and historical movies. And so it's the same kind of thing, right? It's like they're not always accurate, but just kind of evoking the idea of a time period in a place is like a thing that is sometimes really interesting to do. You get to explore aesthetics of things. 

I mean, in some ways, you're kind of like reducing real life to this weird Disney version of it whenever you do historical - I don't know. Historical romances are about this all the time. It's like what if we turned the 1700s into a time of beautiful, sexy people and not just everything smelling like shit like it probably was in real life, you know? I don't know. That's what we do with history is we turn it into different things. 

And so just really coming back to this and wanting to make this game and kind of geek out on it a little bit. And so when it came time to pitch some games to Astra, this was one of them, and they liked the idea. And it kind of fit with what they were thinking. They're backed by like a math nonprofit. And so they need stuff that's kind of like family-friendly, right? And so this kind of - I hesitate to use the word cozy, but this kind of cozy story about a guy going to work at a job. And maybe there's a little bit of light romance, maybe not. It's just very cozy and safe, and nobody's going to feel weird about putting this in a school. 

[0:54:21] JN: Yeah, for sure. 

[0:54:22] ZB: They liked that. We liked that. And so it just kind of all aligned. 

[0:54:24] JN: That's awesome. 

[0:54:26] ZB: Yeah. But then it was originally going to be a 3D game. I had no idea what the gameplay was. I'm like, "You're going to build products, and I don't know anything else." I knew it was going to be about building products probably out of cubes, because we wanted to make it physical. The original idea for - 20th Century Food Court is about assembling stuff, but you don't assemble it spatially. It's almost like when we were developing the game, we talked about it being like a sort of like the algebra of products, which is cup plus purple soda equals cup with purple soda. And then you can put a lid on it and stuff. And it's like this representation of like adding these things together and coming up with - you add like two symbols, you get a new symbol. But it's like every puzzle has its own little set of symbols and how they transform when you manipulate them with these tools. It's actually not spatial at all. It's just like symbolic manipulation. It just looks like food. 

And that was originally what we were going to do for building products. A keyboard, musical keyboard, is like a plastic body, plus eight keys, plus four black keys. And it's like an algebraic thing. But it was hard - 

[0:55:29] JN: It's hard to follow that logic through a lot of things, right? 

[0:55:31] ZB: It is. And a lot of stuff hides what's in it. That was actually why we settled on food is because food very - with the exception of burritos, food almost always shows you what it is. A burger, you can look at the side. I mean, I love burgers and puzzle games. Because you look at the side, you can see all the stuff stacked up on it. And so food exposes itself in a way that most things don't. 

When we came back to this topic, it's like, "Okay, this needs to be - somehow we need to be able to show you what you're building." The idea of we're going to make stuff out of blocks. And originally, I was thinking it'd be 3D. But again, 3D has that problem where you don't know what's inside. And there's like too many dimensions and too much - too much freedom, honestly. 

And so that was how we settled it, like, "Okay, you're going to make pictures of things." We'll kind of like take all the products and put them at an angle where you can kind of see all their components that go into them. And it feels like assembling electronics. But obviously, things are - they don't have any inside. You build a VCR, it's just flat, right? Obviously, most of the magic of a VCR is on the inside, not the outside. But we're going to kind of reduce it to the outside and just assemble a picture of it. And so that was the compromise we landed on. But originally, we were thinking it was going to be like a 3D block building game or something with machines. And it didn't end up like that at all. 

[0:56:41] JN: On that Zachstyle stuff and talking about it being 3D, one of the things that - I mean, I may be misremembering some past electronic games, but it seems to me there's a lot more animated art and voice acting than previous games. Is that all a consequence of the publisher, or is that just a stylistic change? I'm used to the story being delivered via emails, right? This is a very different delivery mechanism. 

[0:56:59] ZB: Just in Shenzhen. Really, that's the only one with emails. 

[0:57:02] JN: Really? Exapunk? 

[0:57:03] ZB: Well, that one has people come to your door, and you talk to them. I don't even think there are actually emails in Exapunks. There's a chat room. There's an AI that talks to you, but that's like in a little window that pops up. And then there's cutscenes of people coming. They're simple, right? They're just a picture of a person, but they come and talk to you. And those are fully voiced in Exapunks. There's a lot of voice acting in Exapunks. 

[0:57:25] JN: I need to do some replays here is what we're learning. 

[0:57:27] ZB: With Shenzhen I/O, we made Shenzhen I/O in 4 months. I don't even know how. We can't do anything in less than a year now. Kaizen took us 18 months when it was supposed to take 12, right? We've gotten very slow and kind of overdoing it on the art, I think, a little bit. But yes, Shenzhen was like four months start to finish. I mean, we spent time adding extra content, but it was very quick. 

But the extent of story artwork is that it's like we have little postage-stamp-sized portraits of the characters, right? And it's all in emails and stuff. And so it was very cheap. No voice, obviously. It's such a minimal game. With Exapunks, we started with - there's that. So, Opus Magnum, we had like cutscenes where there's talking heads that like they're on little portraits and they like pop in and talk to each other. That was a little bit - but still no voice. But that was still like a little bit of an elevation. There's backgrounds of rooms that they're in and characters talking. 

And then in Exapunks, we did a little bit of voice acting, and we found that we actually got a really good return on voice acting because it's like - I don't know. The voice acting in Exapunks was probably like $20,000, right? And so for a game whose budget is closer to a million, $20,000 is nothing. And I think, really, players responded pretty well to it. We were able to get good performances out of these real actors who can do real voice acting. And it just kind of elevated the game a little bit. And so everything after that, we did voice acting. 

Mobius Front, which no one played, has fully voiced cutscenes in that. Eliza's a visual novel that's fully voiced. Tons of text. It's like a fully voiced visual novel. Last Fall didn't have any because it didn't really make sense to put. It's like an old computer. It not going to have a lot of voice acting. And the games are weird. 

[0:59:02] JN: Lots of spooky noises in X'BPGH. Or how you pronounce it? 

[0:59:06] ZB: Yeah, exactly. Yeah. The forbidden path is what we call it. 

[0:59:09] JN: Oh, there we go. That's a better way of doing it. Yeah. 

[0:59:09] ZB: X'BPGH. Yeah, we had all these tools in our toolbox. When we went back to do Kaizen, we're like, "Oh, yeah, definitely. We'll do voice acting for this. Do voice acting cutscenes." I think we went overboard a little bit with some of the art and the cutscenes. They all take place in places. But that was also the point of the game, right? is that we wanted to capture the sense of place. Our artist who did all the story art was in Japan drawing and painting old factories and stuff like that. There's buildings still from the era that still stand. 

And so I think for a game that one of its kind of big strengths is that it's almost a little bit like a fake travel log. I don't go on a lot of trips internationally because it's expensive, and they're hard to put together. And it's hard to see stuff even when you do go places. I think that these games are kind of like a little bit - it's like going on a vacation a little bit. And so, yeah, we can take you back in time, too, which is like really hard to go on vacation back in time. I always think that would be the coolest thing. I'm not a guy with a lot of fantasies, but I fantasize about being able to go on vacation back in time. I think aside from the part where everything like gets you killed and smells like shit. 

[1:00:10] JN: Right. If you're like ephemeral. 

[1:00:11] ZB: Yeah, that would be like the coolest thing, you know. Blend in. Don't let them know. Here's money for the time period. It would just be like the coolest thing ever to go. I'd probably be too afraid to go on a trip back in time, though. There's so many places that I'd be afraid to go in the world now. 

[1:00:26] JN: I'm trying to imagine what the Zachlike mechanics are set in this. 

[1:00:31] ZB: Yeah. Yeah. But in some ways, that's what Kaizen is, is like take a trip back to the 80s. And you get to see some stuff that looks like the place and - I don't know. See the food. Yeah. 

[1:00:40] JN: Nice. 

[1:00:41] ZB: That's why so many cut scenes in Kaizen take place in like bars and stuff. 

[1:00:44] JN: Yeah. Yeah. I mean, that also fits in perfectly. You say you were worried about romanticizing and whatever, but I think it does a very good job. I mean, the characters also just outright talk about being normal people. But I think it does a very good job of just portraying working life and what schedules are like and stuff. Yeah, it was very evocative. 

On the technical side. I believe traditionally, you mentioned C#. And that I think traditional games have been your own framework with SDL2. Is that still the case in the Coincidence era? 

[1:01:11] ZB: Yeah. With SpaceChem, we started using XNA, because it was 2009, and that was the thing to do. So we started using XNA. But really quickly, we realized that that's not going to work on anything other than Windows back then. And we have Linux fans, we have Mac fans. And so we switched to writing our own. Turns out XNA wasn't doing that much anyway. And so we made our own version of XNA, basically, that used SDL. And so we did that for SpaceChem. We used that again on Ironclad Tactics. 

XNA is all just like immediate draw calls. Sprite batch is the thing they call it. I think that's what they call it there, too, because we still call ours sprite batch. But it's basically like you turn a bunch of 2D drawing calls into a dynamic mesh and then you draw it using the GPU. And so we did that as like our XNA replacement. That's how XNA works under the hood, I'm pretty sure. 

But one of our programmer, this guy Keith that I've worked with for a very long time, he had this idea, where it's like, "Well, what if we built like a scene graph dynamically every -" I don't even know where he got the idea. He was doing a lot of Haskell and functional programming. And so it's like what if you did graphics as functional programming? Where instead of it having the side effect of drawing, you just like return like an object that gets plugged into a tree that gets returned, turned into an object and gets plugged into a tree further up? 

You make all these function calls, and you end up with a tree that was constructed of things that all get drawn after you built the tree. And so this was called like scenes. It's like a scene graph. But rather than like a scene graph - a lot of scene graphs. Unity has a scene graph, too. A lot of them do. But you create an object and it stays there. So you create an object once, and then it stays there until you remove it or change it, and it just gets drawn every frame. And so we were generating a scene graph every frame, which is nice because you don't have to keep track of the state. It's just like, "Oh god. What's in the scene graph? When did I last update it? Whose job is it to remove this or update it?" These are all things that I don't want to - these are all things that create bugs in software, right? If you create an object and you don't know who is in charge of it, that's a bug just waiting to happen. 

By returning a scene graph, building a new scene graph every frame, it's kind of like drawing your game every frame. Stuff's only going down if you want it right now. It's kind of like an immediate mode thing, except you're generating a scene graph to describe it all. And so we did this. Space started off being like XNA and then turned into this scene thing. And we had this like horrible hybrid where like some stuff is drawn with these scene graphs and some of it is drawn immediately. 

Ironclad Tactics was drawn entirely with scenes. And so entirely with this like scene graph thing. And two of the educational games we made were same thing entirely done with scenes. The performance on it was terrible. For making little 3D games, it was actually - we spent a lot of time drawing text because we had to assemble these huge scene graphs for like text. And we try to build like, "Oh, this is an object that describes a block of text." So we don't have to have a bunch of little characters in the scene graph, which is one text block thing. 

And it's just - I don't know. it just kind of got out of hand. And then around that time was when Unity was picking up. This is 2013. And so for one of our educational games, we're like, "Should we -" I feel like it must have been our other programmer who was way more into using tools that other people were using. We had two programmers, Keith and Colin. And so Keith was really into making his own weird shit. That's like the guy doing Haskell. 

And Colin, he worked with me on Visio. And he liked C++. And he was very into using tools that other people were using that are things that are contemporary. And so they used to argue about programming and stuff because they had these two totally different philosophies. And so it had to have been Collin. Said, "We should use Unity for a game." So we did. And it was messy because we didn't know how to use Unity. And you just get all these objects created all over the place. 

But we made a game, we shipped it. And best of all, we were able to put it - when Amplify switched from Android to iOS, we were able to just make an iOS build, and it was easy. Whereas trying to port all of our C# games into iOS was impossible, right? We ended up actually having to - in order to port our C# educational games to iOS, we made it so that we could host our engine in Unity and then kick off a build. And so we made an object that gets dumped into the scene that dynamically renders itself using all of our old scene graph stuff. And the performance on it was awful. But we saw with Unity, we're like, "Okay, this is really cool." You can make a build and run it anywhere, which for us has always been really hard. 

And so we did Infinifactory, a 3D game, in Unity. I think that maybe part of the reason why it was 3D is that we felt really confident making a 3D game because we had a 3D engine that we were already using for 2D games. And so we did. And I made TIS-100 in Unity using their at the time new UI system. 

And the thing we realized is that the tooling on Unity is definitely - you just end up with stuff in the scene and you don't know how it got there or what it's doing there. If you change a file, you might - I was always really afraid of accidentally changing a file. And even though we had everything in text mode and could diff it and check it in, you still don't really know what all the changes are. You make a change, it changes like a million lines in this huge text file that has all your shit in it. You don't really know what you're changing. 

And so there was the uncertainty of not knowing exactly what was happening with the game. We had some build reproducibility issues where sometimes textures would get detached. Every time we'd make a new build of the game, I'd have to make sure that one of the characters didn't turn purple. Because for some reason, the texture reference kept getting messed up. I still have no idea why, right? I still have to be on the lookout for that whenever I make a build of Infinifactory. 

Yeah. So there's that. It was expensive to use Unity. The licensing fees on it were quite high. And they were switching to stuff that was even higher. And then the big thing that made Keith refuse to ship another game in Unity is that when we got to the end of Infinifactory development, it was slow. And so he was trying to figure out how to profile it and making it faster. And it was like almost impossible, right? 

And we're not the first people to run into this with Unity. The thing that people say about Unity, Unity makes it really easy to start a project and really hard to finish a project. And it was true for us. TIS-100 had an abysmal performance. The performance was so bad that a developer from the Unity team contacted me, because game programmers love my games, and was like, "Hey, can I take a look at your source code? I can help you make your game run better since it's a Unity game and I work at Unity, etc." I sent it to him, and he's just like, "Uh, yeah, you could try capping the frame rate to 30 frames a second." That was the only thing they could think of, right? Because I wasn't doing anything wrong. It's just that their UI system was really, really inefficient for some reason. I have no idea why. And the performance was abysmal. And it was a game that was trying to look like DOS. It was crazy, right? 

Between those two experiences, Keith refused to ever use Unity again. And I was like, at that point - I don't know. I was like, "How could we possibly make a game without Unity? The tooling is so good. When somebody takes you hostage. 

[1:07:49] JN: Stockholm syndrome. 

[1:07:50] ZB: Yeah. I was Stockholm syndrome in using Unity because we had built up some tooling in it and it was just I was so used to it. But he's like, "No, we're not going to use Unity anymore." And that was when I went to work at Valve. So I actually ended up using Unity at Valve before they killed Unity at Valve because they have their own engine that they want people using instead of Unity. 

But when I came back from Valve to work with Keith again, he had spent the 10 months that I was at Valve writing his own game engine. And it was based on Casey Muratori's stuff with like the Handmade Hero. Writing a game engine from scratch in C. That like you don't need anybody else's shit. You can just write your own game engine from scratch in C, presumably using SDL. But he had implemented like live reloading in C and all these things. 

And so Keith had watched all this. And then Keith was working in C, but I told him there's no way in hell I'm using C or C++. And so we built up our own version of that in C. And so we have live reloading in C. We had switched entirely to immediate mode stuff. So no more scene grass. You just call draw and it just draws something, right? And we switch to all immediate mode GUIs. 

The way you do a button is you draw a button and then the function that draws a button returns true if the user clicked on that button right then and there. And so it's all like an immediate - it's hard to describe immediate mode GUIs because I can't show code while we're talking about it. But if you look it up, you'll see that like the flow of the program is also the flow of drawing - your UI is also the flow of logic for it. Right? The idea that like a button - the function that draws a button returns true or false based on whether they're clicking the button right now I think kind of makes it obvious, right? 

Our games have so much UI in them. TIS-100, all UI, right? That this paradigm works really well for the kind of stuff we do. And I was worried about like, "Oh, is it hard to dynamically lay out stuff?" It's like, "No." You just do a little bit of math and then you put it where it needs to be. When you draw text, you just measure the text and you can lay stuff out wherever you want. And so that's how we wrote games. That's how we did Opus Magnum, everything beyond it. 

When we went to Coincidence, we managed to get a license to keep using our old engine code, basically. We don't really have an engine. We just copy the last project and delete all the game code out of it and then use that as the starting point. And we've done that probably 10 times now to get to Kaizen. But that's how we write all of our games now. And it is like superior to anything we've ever done before, right? I guess we haven't shipped a 3D game since we started doing this. So maybe that's a little telling, right? Is that we kind of have 3D support, but we've never actually used it. 

I don't know. Most of the game ideas I make don't make sense in 3D. A lot of them are like kind of CAD experiences. We have 3D in the Last Call BBS. I take that back. In ChipWizard, we have a little 3D visualization of all the pads and the silicon of the integrated circuits you're building. It's this little 3D view. And we wrote a software renderer for that. We're manually sorting all the polygons. I mean, we wanted it to feel like an old shitty 3D thing from the early 90s. And so yeah, we wrote a soft like a - I mean, we're not like rasterizing it ourselves. We're drawing the quads with our stuff, but it's like kind of like a software renderer kind of thing. And that's the only time we've shipped 3D in recent years, but that's what we do now. 

[1:10:53] JN: Okay, cool. Yeah. I mean, I guess I was going to ask if the nature of the engine constrains what you're making, but I guess it's more the other way around that the engine is what it is because that's the games you like making, right? 

[1:11:03] ZB: Yeah, exactly. And like if we wanted to make a 3D game, hypothetically, we can, right? We have depth buffer support and stuff in our engine. Keith added it hypothetically for the future, but we've never shipped anything with it. Again, it's none of the stuff I have teed up is 3D just because we don't have any 3D office artists in the office. And everything gets really expensive when you start doing it in 3D, right? It's pretty easy to hack stuff together in 2D if you're making games about programming. 

[1:11:34] JN: Yeah. Yeah. Absolutely. So, I guess to switch back to like the business of making games for a minute. I think a lot of game developers would find the idea of rebranding, especially from like a niche where you're so well known into a new one, quite a daunting task. How did that go? Did you manage to bring all your audience with you? Has everyone found you again? 

[1:11:51] ZB: Oh, I have no idea. Probably not. I don't know. I was always kind of skeptical that most people don't know who Zachtronics is, right? People have heard of Opus Magnum, maybe, but they have no idea who made it. And I think that's true for most games, that I think as a person in the games industry, of course we know everybody who makes everything. But I think gamers in general are a little bit more selective about which game creators they know about. 

I think a lot of people know like the name Blizzard. Right? There's some game studios that everybody knows. But we're not Blizzard, you know? We never were. I think that most people probably don't pay attention that way. Honestly, I think most people, they would just see Opus Magnum pop up on their Twitter timeline or whatever, and that's how they learn about it and play it. And they never know what a Zachtronics is at any point. And so I don't think it hurts us there. 

And I think all the people who were into Zachtronics enough to know us are probably subscribed to our mailing list. And I have ways to tell them that we have a new studio. And obviously, that's not true, because I still get emails all the time at Zachtronics from people being like, "Oh, I hope you make another game one day." And it's like, "Yeah, we did. It's out now. We have a new studio." And they're like, "Oh, wow. Cool." And it's like, "Okay, I failed in that department." 

I think, obviously, the switchover has not been perfect. But honestly, we didn't really have a choice. if I could have seamlessly kept using the name, I probably would have even though I'm not - I think it's embarrassing, right? literally came up with it when I was a child. 

[1:13:12] JN: You have also just launched a physical game called Zach Attack, though. This does have to be said. 

[1:13:17] ZB: Oh, yeah. Yeah. I mean, I still put my name on stuff, you know? And we've been very forward with like my name and the marketing stuff because it's like - I don't know. It's like that is my name. But we were given the opportunity to continue using the Zachtronics name, but it did not make sense. 

[1:13:33] JN: Expensive opportunity, I imagine. 

[1:13:34] ZB: Yeah. Yeah. Yeah. An insulting expensive opportunity. I don't know. It's my name, right? And so I feel like if you don't own something, you don't really control it. It's like, "Okay, unless we own the name Zachtronics, I don't want to borrow it." Because then it's going to cost more in the future if things go well, you know? And it's just like, "Fuck it. We're a new studio. We have a new ethos. We run things differently." Zachtronics really was me top down dictating everything. 

That's not what Coincidence is, right? Coincidence is everybody does whatever they want. Sometimes I dictate stuff, sometimes I don't, right? In 10 years from now, maybe people will know Coincidence like they knew Zachtronics. I have no idea. Whatever. You just got to make the games, right? The games are what draw people in. If Kaizen had a weak response, it's not because people didn't know where Zachtronics went or that it's a Zactronics game, right? It's because Kaizen didn't like shock people into buying it from either changes in - the game wasn't good enough, or tastes have changed, or who knows, right? Silksong. It's all Silksong's fault. 

[1:14:31] JN: Yeah. So, I don't want to do the horrible thing of asking you what's coming next, but we've established that Kaizen got some ideas from Last Call BBS, which is very important to me. Because in terms of your noted previous ideas that didn't see the light of day, I am dying to know, are we getting Submarines? Are Submarines happening under Coincidence? 

[1:14:49] ZB: Oh, no. That game, I don't know. We built it. It didn't work. I actually really have zero desire to make anything that's like a survival crafting game, especially now. It's been done to death. When we started that game in 2016, that was like kind of a fresh genre. Everybody does that now. Every game is like that. That's not true. There's lots of Balatro-like games, too. 

[1:15:10] JN: Sometimes the survival games have Balatro-like games in them. 

[1:15:12] ZB: Yeah, sure. I don't know. This is a thing I wrestle with a lot still. If I was making just buckets of money just by doing the thing that came naturally to me, this would be really easy, right? Because I just be like, "I'm just going to keep doing the thing that comes naturally to me." The stuff that comes naturally to me has always been good but not good enough. 

And so I'm always wondering. It's like, "Oh, should I be doing something different? Should I be mixing in different stuff?" And even when we talk about in the office, a lot of times it's like, "Kaizen did okay. How do we make a better game?" And then the answer is always like, "Oh, we could add more mechanics from other popular games." And it's like at one point, I got kind of fed up with it. It's like, "No. Why are we not focused on trying to just make the best games in our style that we can?" Why is the answer always that, to make a better game, it has to be less like what we do and have been successful doing? 

And the reality is that I'm constantly torn over this and I'm never like - I'm not so full of myself that I think that like, "Oh, I have this vision that needs to be realized." It's like, "No. I'm just kind of making games over and over again." We're so far past the point where I'm making like my dream game or whatever. But like at the same time, to a fault, I have to do my own thing. I'm just not interested in stuff that doesn't - by my ego in that way or whatever. 

And so this is why I'm not cut out for software. I think to make like real good software, you actually have to be like really low ego. And you have to not care about what it ends up being creatively because you're just trying to do a job. And you just need to do the best version of the job. And like I wish I was that person, but I'm not, right? 

A big part of what I do is being creative, right? I spent a lot of time thinking it's too bad I couldn't fit in at Valve. Because I'd be making a lot more money if I was at Valve. But the reality is that the reason I couldn't be at Valve is because the opportunities for creative are like pretty thin, right? They make a game and it has - their games are bigger than my games, right? But way more people work on those games. And so the amount of creative input you get to have on a project like that is just a lot lower. 

And if I think back to all the people I worked with at Valve, they were all people - their strength was not unbridled creativity, right? Their strength was that they're all extremely competent game developers, right? But me, I'm not extremely competent. I'm extremely creative, right? And so that's what the game is for me, right? That's what this is. I think it's so obvious at this point that it's like I just have to like accept it, right? That like my strength is creativity and putting that into practice in this weird kind of field. And so those are the kinds of things I want to do, right? 

Yeah, you're asking about what's coming next. Right now, I am working on a bunch of puzzle games because that's a thing I can do. And I know that at least to some degree it'll be received well by people. It just works. You just have to do the stuff that works. And it's an opportunity to be really creative while I'm doing it. I love the - systems, I think, are very interesting. The games we have coming up are experiments. 

If you liked that kind of experimental nature of Last Call, our next two games are going to be smaller, rougher, and experimental systems. Exploring what are new - I hesitate to say models of computation. But what are new kinds of systems that we haven't really done before that explore this kind of computational puzzle space, right? Yeah, because that's what I like doing. 

[1:18:16] JN: Definitely my favorite bit about Last Call. The Forbidden Path was awesome. That was very cool. 

[1:18:21] ZB: How would you feel about a bigger version of that with a better system that's less like random and more deep and - 

[1:18:28] JN: Absolute dream. Absolute dream. 

[1:18:30] ZB: Yeah, that's not what we're working on, but I - 

[1:18:32] JN: But you should. 

[1:18:34] ZB: I'm working with a writer on one of our upcoming projects who I was a fan of many years ago. And I don't want to announce to - I've never talked about this with anybody. I'm not going to announce too many specifics. But I'm working with a writer. And so he's very creative also. We started working together. I'm like, "Pitch me a bunch of ideas for video games." And he's like, "I've never made a video game before." And I'm like, "This is the kind of video games we make. These are how our puzzles work. Pitch me a bunch of games that work with this." He pitched me 12 ideas for games. They're really good. Right? 

And so one of them, I could imagine being a kind of Forbidden Path. What if that was expanded into like a real game and stuff? And so I'm tempted to make that one. So you saying that you like that game a lot makes me think like, "Oh, maybe there's something to that. Maybe I should do that." 

[1:19:20] JN: Yeah. I don't know if cellular automaton hits what was going on there. But ultimate model computation, when you said that, I was just - that straight away where my head went. I just thought that was a really good example of that. Well that's very exciting. 

[1:19:31] ZB: Yeah. It's not that often you get to interact with systems about stuff that grows. Right? It is cellular automaton. But I'm not - I think some people are kind of math nerds that really like cellular automaton. And for me, the thing that's really interesting is like what kind of information do you need for something that grows? That was where Forbidden - Forbidden Path, I think, is actually one of my favorite little designs. Because it came from years of trying to think of how to make a game about cellular differentiation. Because I've always been really fascinated with the idea of like how is it that a person grows into the shape of a person, right? It's unbelievable. It doesn't make sense, right? 

I think some people might think, "Oh, well that means that it can't be understood." And it's like, "No, it can be understood." I want to understand it. I want to I want to make a game that distills that so people can understand that concept of how things can grow into these complicated shapes. And the more I learned about it, cellular differentiation is just so weird. And it still doesn't make sense how - just how chemical gradients - how does a chemical gradient turn into this beautiful specimen of a man that you're talking to right now, you know? It doesn't make sense. I don't know. I think that's a good prompt, right? It's like magic, right? 

That's the thing. I feel I always fall short of it. Kaizen certainly falls short of it. Maybe that's the problem. But I want to make a game that's about an engineering system that's entirely fictional. So it's like magic or something, right? And a lot of people want to make a game. What if magic was real? But they always fall back into game design stuff where it's just making spells in Skyrim. It's like, "Oh, it's an area spell, and it does fire." But it's like, "That's not really systematic." Right? 

And so that's the thing that I always want to do. And that's been the inspiration of a lot of my games. One of my very first Flash games, The Bureau of Steam Engineering. But that was like a pretty steampunk played straight kind of thing. But before that, I was like, "What if there was like magic? What if you had magical steam mix?" I'd seen somebody made like a role playing game setting that was about that. And so I wanted to like - how does magic work if it's real? And that's always been like a thing that I kind of keep coming back to. These systems that are - they're not real systems, but they have the believe, the verisimitude of like a real system an engineer would apprecaite. 

[1:21:39] JN: Yeah. Tere's a system of logic and rules, it's just beyond - yeah. 

[1:21:42] JN: Yeah. And so anything like that is like the ultimate thing that I'm like chasing after with a lot of these games. And I almost always fall short of it because it's really hard to make stuff that's completely made up yet somehow systematically functions as if it was real and coherent. I don't know. It's just hard to carve out those spaces, right? But that's always the dream, right? 

[1:22:01] JN: Yeah. Well, I think that is a fantastic place to end. Zach, thank you so much. If folks want to - and they're not already on your channels and they want to hear about these upcoming games, where's the best place for them to go? 

[1:22:13] ZB: Oh god, I don't know anymore, right? Social media is dead. You can follow us on - I think really the best way is we have an email newsletter. If you go to coincidence.games, you can sign up for our newsletter. And that's the best one because I only send out stuff to - I've actually gotten it down to a dollar whenever I want to send out an email. It used to cost a lot more. But we self-post and do a bunch of stuff with that. 

But I only send out emails when there's earnestly something exciting going on, right? Something to buy, something cool coming out soon. I don't spam it. And so that's the best way to find out about stuff because you'll be guaranteed to get the email, right? Whereas if if you follow us on Twitter or Bluesky, you might see something, but you might not. And you might see a lot of other nasty shit if you're on Twitter, too. 

Whenever I log on to Twitter to check our Twitter, it's always just like, "What is any of this stuff?" Right? I think our email is the best way to go at this point, especially for your audience, right? I love email. I don't know. Do you love email? I love email. 

[1:23:12] JN: I am a big fan of a newsletter. I do like a newsletter. 

[1:23:14] ZB: Yeah, exactly. I think a lot of people in the kind of programming tech space have switched back to loving email newsletters because - 

[1:23:21] JN: Give me a newsletter with an RSS feed, you know? 

[1:23:23] ZB: Yeah. Exactly. I love email because it's kind of like an RSS reader without having to check an RSS reader. It's stuff comes to you and it's like the stuff you want, usually. Somehow I got subscribed to like - oh god. It's like pipes and valves. The pipes and valves industry newsletter. And so I get all this stuff about - 

[1:23:42] JN: This thing is useful for you, to be honest. 

[1:23:44] ZB: No, because it's like really hyper-specific about pipes, you know. And it's very weird. And so I guess aside from stuff - when you get signed up for stuff like that by accident. A big fan of email though. Do you zero inbox? 

[1:23:57] JN: No. Never. Absolutely not. No. I'm a demon. My inbox would - I don't have a single email account that's got like less than 100,000 emails in it. 

[1:24:04] ZB: Oh no. Well, I mean, you can archive them. It's just how many are in your inbox. 

[1:24:08] JN: I don't like to. It has to happen as process or it's not getting done just naturally. Yeah. 

[1:24:14] ZB: Don't listen to him, everybody listening to this podcast. Zero inboxing. 

[1:24:16] JN: That's it. Everyone's unsubscribed. It's done. I've killed it. 

[1:24:19] ZB: Yeah. Zero inboxing is one of the best things I've ever done in my life. Right? It's like version control. I had 10 episodes of my podcast. And you'd be shocked at how many indies came on and were just like, "What's version control?" Or like, "Oh, we should use version control, but we don't. We just live edit files on our web server." And it's just like - 

[1:24:37] JN: I mean, this could be a whole other podcast. Because version control in games is horrible, though. It is cursed. The options are all bad. You get LFS? Are you locking file? 

[1:24:51] ZB: We just use git. Well, I guess if you have a lot of people. We have really small teams, right? 

[1:24:55] JN: Yeah. How are you handling big binaries? 

[1:24:58] ZB: Not that big. 

[1:25:00] JN: Okay fair enough. 

[1:25:00] ZB: PNGs. I don't know. They're like a couple megabytes. That's not that big anymore. We actually did - well, we used to use Bitbucket. We were on their free plan. They dropped their limit to one gigabyte for all of your repos or something. To add it together can take up one gig of space. It really seems like they're trying to like kill our free plan off. That was a problem. But as long as you don't have like big caps on it - our games are actually kind of small. Source code's tiny. We only upload compressed audio. We only upload PNGs and stuff. We have other backup solutions for all the source files. 

[1:25:34] JN: Yeah. But imagine like a 3D Unity game. They're punished by Git, right? 

[1:25:39] ZB: I guess. I don't know. Infinifactory was all in Git, right? That was actually like quite a few people. Yeah, because the assets are usually not that - as long as you're not - I don't know. PNGs, it's a good compression scheme, right? And like you can do other stuff if you want. I don't know. 

I mean, that was not like a super high res. If you're making a game that installs and is like 150 gigabytes, yeah, you have problems, right? We use like Perforce or whatever. Everybody manages. It's fine, right? That's what I used when I was at Valve. They had Perforce. When I was at Microsoft, they had Perforce. It's fine. You don't need Git. I don't know. Yeah, branching sucks. How much branching do people really do? I don't know. I mean, I guess less when it sucks, but people manage. 

[1:26:16] JN: I already know game devs. Yeah. Not as much as they should. All righty. Well, thank you so much, Zach. Thanks for joining us today. 

[1:26:23] ZB: Yeah. Thanks.

[END]