Building an Infinifactory: How to turn science into a successful game

At the end of Janurary Zachtronics, headed up and partially named after Zach Barth, released Infinifactory on Steam Early Access. Bucking the trend, it was almost entirely complete: A release that had no visible gaps in content, or major bugs. It was essentially a release version, soft-launched through Early Access. 

Gamasutra talked to Barth back in October about his motivations for this approach, and the benefits of having people play a game that isn't considered finished through Early Access; it appears to have worked out exactly as he wanted. 

With its swarm of conveyor belts, pistons, and welders, Infinifactory extrapolates the ideas that Barth established with SpaceChem, translating them into 3D which understandably increases the complexity of the puzzles Zachtronics has created. 

I talked to Barth about how things are going with Early Access, as well as why the small indie studio decided to invest in a fully voiced story in an abstract puzzle game, and what it's like to have Jon Blow and Notch publicly playing your game. 

It seems your motivation for doing Early Access was slightly different to that of most developers, in that you wanted to sidestep people judging the game before you’d had a chance to adjust it to the public. 

Premature reviews, certainly. The primary reason, though, is that it’s hard to telegraph to players that, "Yes this game is out, but the scope of content and what you see in the game isn’t final, and it might change, especially early on." Letting players know that we are listening to them and want to build the game around their reaction. 

And you’ve got Steam workshop support with Infinifactory, right? How’s that been?

It’s been pretty good. So… kind of like SpaceChem, we have this quirk where people think our games are hard with the puzzles we make, but the user-created levels are always so much harder, and already that’s the case.

Ours are designed to be very open ended, and not about "tricks," necessarily, but just building different kinds of things which require thinking about how you’re going to construct things. A lot of the puzzles in the workshop are: "Here’s something we managed to do that’s really strange, see if you can figure it out too!" So it’s a totally different kind of puzzle, and I often have no idea how to complete them. 

It seems that often with SpaceChem and Infinifactory there’s a wide spectrum of success. You can create a really clumsy and convoluted solution that just barely works, or you can make a really tight and elegant solution. Is that always your intent?

Absolutely. That’s the point of the whole game. Any puzzle game can create a solution, tear it apart and then get you to put it back together, but with our puzzles they’re so unlike traditional puzzles that they’re more like a task -- a puzzling task -- than a puzzle. 

There have been a lot of games that have used a coding-like interface to communicate hacking, or interacting with computers, but the product is often not much like coding. Whereas with SpaceChem and Infinfactory it seems much more like you’re trying to convert coding into something more understandable and less conceptual.

So SpaceChem had a slightly more direct comparison to computer architecture, as it was based on a sort of very convoluted computer architecture that doesn’t exist. This is much more directly distilled from factories, and industrial engineering. It’s more engineering than programming.

There are a lot of "programming" games around there, with air quotes around that, and they sort of superficially are about programming, or their interface is about programming, but I don’t think any of them communicate what it’s really like to do programming, especially the problem-solving aspect of being a programmer. I feel like that’s the really amazing thing about SpaceChem and Infinifactory, that other games, even if they’re about programming, can’t do. 

And was that your intent when you started making these games, to communicate to people who don’t or can’t program or code, what it’s like to do these things?

Probably not, because the people who play these games are probably good at this kind of thing, or are already on that track. It really was an earnest expression of the fact that I really like this kind of problem-solving, and I like engineering systems -- I have a background in engineering -- and it just seemed like a fun thing to make. 

The reaction in the game development scene to Infinifactory has been very positive; I’ve seen Jon Blow and Notch both talking about playing it, with the former streaming some of his sessions. I guess that’s a kind of vindication, where these prominent developers are enjoying your game. 

When I go to game dev meetups, I find that a lot of people haven’t heard of SpaceChem, which isn’t surprising, and they ask “Oh, what’s SpaceChem?” and the first thing I ask them is “Are you a programmer?” and if they say "yeah," I’ll tell them to just go play the game.

[Ed. note: If you're curious about SpaceChem, you can also read Barth's postmortem of it.]

I think the thing with these games is that they really just click -- and it’s not that they’re not for other people -- but it’s that they really do click with programmers, and people who do technical problem-solving as their thing in life, or one of the things that they do. You’ve just named two people who have long programming backgrounds. They’re both programmers who like games, and this is straight up the game for them. 

Have you seen any direct positive response to Infinifactory’s sales in the wake of their attention?

I’m sure it’s not hurting. There hasn’t been a noticeable effect, but the biggest spike was releasing the game. But I’m sure there’s been some sales from it. 

In general, it hasn’t seemed like you’ve aggressively marketed the game, and I know when you talked to Mike, you said you’d barely put out even more than a few screenshots. I know you’ve put out a trailer now…

Yeah, we had to! In the past, trailers have always been a really hard thing for us to do; we don’t have anyone on the team who has a background in video, but we actually pinged out that trailer in two hours, and had a lot of people complimenting us on our trailer, which is really surprising. So we learned a lot about scope and effect of trailers there.

But in terms of advertising, we’ve been pushing it out to reviewers. We haven’t been holding off on doing any of those things because it’s Early Access. I know some people are put off by Early Access things, because they’re usually kind of crap, and people don’t want to come out and say it’s crap, but in the case of our game it’s not, so I’m not really afraid of that. 

Are you saying it’s difficult to overcome the stigma of Early Access?

No, not really. As we were getting close to the end of development, I was talking to people about it as a sort of crazy idea I was having, to do this game as EA, even though it’s very polished. And some people were saying "I don’t know, people might think it’s not good enough," but one look at it and it’s obvious that it’s good.

We’re 100 percent -- and I don’t want to jinx it -- but on the Steam user reviews we’re 100-0 on positive reviews. And people can play it and quickly realize that "Oh yeah, this isn’t just your average EA game." I feel like it sort of worked, in that regard. It’s also something that I want to do in the futurebecause I really like the benefits of being able to involve our players in the development process, which isn’t something we’ve been able to do in the past. 

Do you think you would continue to hold off until the game is quite far in development before pushing it out as Early Access in the future?

It depends. The big thing that made us hold back Infinifactory before we released it was the story. There are a lot of story elements, and the puzzles you can play again, whatever, but the story is stuff you can only really get once, and if the first time you go through the story is all buggy and crappy, the next time you go through isn’t going to be fun. I was ready to push out the ship date if we couldn’t get the story in. But we got it in! And it’s really cool.

I’m interested about that aspect of Infinifactory, actually, as I could easily see a version of both SpaceChem and Infinifactory where there is little or no story, and they’re both just a series of satisfying puzzles, or tasks. And I’m not saying it would be fine without them, but what’s the reason for the inclusion of the story when it’s clearly more work for you to do?

This is actually something we’re asked about a lot, and I mean.. Portal, right? You could easily see a version of Portal without the story, and it’s like saying, "I could see a version of cake without anything sweet in it. It’s still cake," right? 

The reason I ask is because it can’t necessarily be the easiest thing to do as a small developer, especially as you’ve got voice acting and something like cutscenes in Infinifactory now. 

It’s not free, but I think it’s important. I think it’s really easy to say for most games, with the exception of storytelling games, whose mechanics rely on storytelling mechanics. You could sceptically see it as an unnecessary addition to the game, but everyone keeps doing it; everyone keeps adding stories to games. I think it makes it feel more impressive to play. There’s something about it that makes it better, and it definitely improves the experience. The tradeoff is that it improves the experience and it costs a lot of money. 

With Ironclad Tactics we ran into this, because there’s a really big graphic novel in that game, it’s like 80 pages, and we spent a fair amount of time and money on it, and people didn’t really respond very positively to it. It had a lot of words, and not everyone is into that, and especially if you’re playing a game and it’s throwing all this text at you and things to process, and you just want to play the game… you didn’t opt into it, and that’s not something that goes well with players.

But even though that didn’t go well, I don’t think that’s an indicator that there shouldn’t be story in games. With Infinifactory we decided to do audio-logs, which we’ve seen in other games and by comparison is much cheaper. And people love it. I’m surprised at the extent to which people enjoy them. 

They seem to be successful because you have to go a little out of your way to engage with them, and they don’t get in the way of you playing the game, as you can listen to them while you build your factory. It means that it is completely optional, which seems important. 

Yeah, I think that helps people not have a bad experience with it. And with a lot of this stuff, honestly, it’s like you don’t have to have great usability in your game either, but it helps! It’s really not essential, and that’s one of the reasons we have Early Access, it’s the principle of minimum viable product applied to games: how little can you get away with? Just because you can get away with what you have, doesn’t mean you should. Usability is very important to us, and so is story. It really makes the product come together as a cohesive whole, and takes it from being good to being able to be exceptional. 

The funny thing is that with our story is that it doesn’t force you to engage with it, but in a clever way, I think, it is woven really tightly with the gameplay, I don’t know how far in you are…

Oh, not that far at all. I’m terrible at this kind of thing. One of my coder friends saw one of my solutions for SpaceChem and he said that it made his head hurt trying to figure out what on earth I had done. I very much take the "blunt force" approach to finding solutions in both SpaceChem and Infinifactory. 

The funny thing is that that is what separates a good programmer from a bad one. It’s not that a bad programmer won’t solve the problem, it’s just that it’s messy, and hard to follow the solution. It’s another case where it mimics programming or technical problem-solving skills, in that you can usually hack your way through it, but it’s not going to be pretty. 

But with the story thing, there is some very tight thematic cohesion between the puzzles and the plot. I don’t want to spoil too much, actually, but the core premise of the story is that you’re abducted by aliens and forced to build factories for them. A lot of the arbitrary nature of that experience mimics that arbitrariness of puzzle games. When more people have played it, we can spoil it more. 

The other thing I find interesting about both SpaceChem and this are the post-level graphs. I could see those graphs as being either encouraging or discouraging, depending on the player. For instance, if someone has just completed a tough level and sees that their solution is way worse than even the average, that could rob them of their satisfaction, although of course the inverse is true. Is that something you thought about when making them visible to the player?

That was something that was in SpaceChem, so we got a lot of room to test it in that game. It’s not that I’ve never heard someone say that it turned them off, but that is by far the minority opinion. The thing we see way more often is that people see themselves at the wrong end of the graph and say to themselves, "I can do better." 

I think it’s especially because your first time through, you’re just trying to find a solution. But once you’ve found it you can go back and make it faster, so it feels like a very tractable challenge to go back and make it better. And the fun thing is that those graphs are populated with people who are doing their first passes, people who aren’t doing optimizing, and people who don’t care. So it’s not like a leaderboard, where you’re seeing who’s best and then how many thousand places behind them you are. It’s a pretty honest spattering of data. If you try hard, it’s not hard to beat the average. Even if you don’t try hard. 

How has it being doing in Early Access? Has it been doing well?

It’s going well. Our big goal was to engage with the players, and that’s been a great success. We use Reddit for a lot of our forum-y kind of stuff, and we started a subReddit for our roadmap for the game, and there were like 500 comments posted last week of people giving feedback and suggesting things. I’m struggling to get through all the engagement we’re having with players, which is kind of crazy. 

For more on what makes SpaceChem great, read Margaret Robertson's Five minutes of... SpaceChem.