The Witness Dev On Overhauling The Game For Mobile, And Playing A Ton Of PUBG

Jonathan Blow led the design of two iconic independent games that dug deep into the minds of those who played them: Braid and The Witness. Both were released on consoles and PC, but Jonathan and his team at Thekla broke into the mobile platform recently by porting The Witness to iOS. It brought up plenty of questions about how this style of game was adapted to a much different ecosystem and what it means for him and his squad of developers.

The Witness received critical acclaim when it initially released in 2016 on PC, PlayStation 4, and Xbox One. GameSpot’s own Mike Mahardy gave it a 9/10 and stated in his review, “The Witness tells a human story, about people trying to make sense of the world around them.” It even earned ninth place in our Best Games Of 2016 list. Naturally, we jumped at the opportunity to speak with Jonathan over the phone to pick his brain on the challenges of game development, mobile games, what the future holds, and games he’s been playing.

GameSpot: As a game developer, talk about some of the challenges in porting your game to a mobile platform. What difficulties did you run into with getting the game to run properly?

Jonathan Blow: There were sort of three main aspects to it. Part of it was just brute force on the graphics engine, hammering on things to make them fast in a way that’s customized for mobile platforms. Our shaders are simplified a little bit so we don't have to spend as much computation per pixel.

The second part, Apple has a graphics API (application programming interface) called Metal and anytime you take a relatively complicated engine like ours and adapt it to a different graphics programming interface, that's already a bunch of work. But it's good, because in this case it lets us use the hardware much more effectively than we would be able to if we were using something like OpenGL to draw on iOS. It just doesn't let you have as much control over what's actually happening.

And the third part was just content. We went through and did lower-poly versions of almost everything in the game. It’s not like there's a lot of copy and pasting of things all over the island, every area’s unique, and so that was a lot of work, especially for our small team. We didn't outsource it, because we thought we might get subpar results. The other thing about The Witness is the gameplay depends heavily on how things look, in a way that's not true for most games, so we felt we had to be super careful with that whole process. When I play the game now, it’s like I almost don't notice that it's not the same as the PC version graphically. Part of that is probably because I'm playing on a smaller screen, but things look good for how small the form factor is! It's funny, because when we were building the game, we got a bunch of devices to test on. And one of those test devices was the iPhone SE--it's super tiny. It's not a high-powered system, but somehow the game runs great on it, and I'm shocked by that. It’s this tiny tiny thing!

Gallery

How did you come to the conclusion with the tap-to-move control scheme?

Early on this was a question, and I'm like how are we gonna control the game. And most games that do 3D aren't free roaming. Those are not good models for comparison, but some of them do this virtual joystick, which I think for this game especially would make it really unplayable if we did that; just having to walk around the whole island with a virtual thumbstick. Though, we could've done that port much much faster.

I seem to remember way back, early in 3D days on iOS there was a demo that Epic put out, I think it was called Epic Citadel. It had this tap-to-move interface; when you tap something, you go there. It wasn't that great, because it wasn't a real game they put a ton of time into. And they also had virtual thumbsticks, but I always felt like that was a proof of concept.

So, this would decide a while back that [tap-to-move] was going to be how navigation worked. In The Witness you'd go up on the mountain and tap somewhere across the island and it'll move you there. Doing all this pathfinding across the island, accounting for the states of which doors are open or closed, all this stuff took a substantial amount of work. That wasn’t necessary for the original version of the game, because in the original game there was no pathfinding, you just walked around. That was a big part of making it playable for sure.

You previously mentioned that The Witness has no action sequences because you want players to focus on understanding puzzles and soaking in your environment, free of distractions. Mobile gaming is more prone to distractions around you. It’s almost like this metagame of being distracted in your real world. Do you see that as a barrier to get the full experience of The Witness? What are your thoughts on that?

It might be. I mean I'm not sure what to think about all this, it's a little weird. On the one hand, mobile is defined by ‘I got my phone and I'm on the bus and I can play something.’ but on the other hand, if I'm gonna watch a movie I might just take my iPad in my apartment and then I'd just watch the movie in a focused way. I think those things are possible. It comes down to culturally, how do people have their lives and their expectations set up?

I do strongly dislike what has happened, even with PC games, that started sort of on mobile where it's expected that you're supposed to be doing something. You have all these notifications pop up and bother you all the time. I don't think that's conducive to a deep experience. I turn all that stuff off on Steam, for example, or any other service like that. I have it turned off on my phone.

But I think a lot of people have that stuff turned on and... well, I don't know. I'm sure some people will have a distracted experience, and if they have a super distracted experience, they may not be able to play the game very effectively, because it’s very subtle. But then I'd just hope that through shared numbers, because it's a very popular platform, that we'll end up having a lot of people play it in a relativity focused way. We'll see if that happens.

Mobile gaming tends to not be as deliberate, do you think the deeper, more abstract themes are likely to be lost on mobile? Did you do anything differently to make some pieces more noticeable?

We really didn't do anything to change the game in that way at all. In part because it would be so hard to change. With the original game, everything has constraints on everything else. Making something larger to make it more obvious on a phone, that's one thing that you might do. Well, now that thing is going to encroach into some neighboring area that's precisely sized for a reason and it might block a sightline between one thing and another thing. In most games it doesn't matter, most of the time, but in this game it matters a great deal.

Like I said, the individual meshes, textures, assets, and stuff are decreased in weightiness for mobile. There are lower polygons, fewer pixels, and and we worked a lot harder on compressing them. But the actual shape of everything in the world is the same. The sizes of things are the same, the colors are as close as we can get them although the screens on mobiles tend to be a little bit different in terms of color balance than PC monitors.

Charting unexplored territory is a huge theme in The Witness. So, for you exploring this uncharted territory of porting a game for mobile, how is this different from creating a console or PC game? What did you learn through trying to adapt to a mobile platform?

I mean, it actually wasn't super different. As we said, it was a challenge, because the hardware can be very much on the low end. On the high end, an iPad Pro is actually a pretty fast machine. Trying to work the low end, that part is challenging, but we didn't approach it like we're trying to make a free-to-play game with microtransactions. When we did the design, we thought of what subtle things can we do to improve the control scheme so that this part of the game works better? How can we put invisible mark up into the world to help the pathfinding better in this particular situation?

It was very specific design decisions where the constraints came from the game that we had already made. If we were designing a new game from the ground up for iOS, it would probably make a big difference, but I cannot even tell you exactly how it would be different because we haven't designed that thing.

I would say we lucked out a bit. When you get down to it, technically they're both computers that render graphics roughly the same way. I mean the details are all different between those platforms, but the computing paradigm is the same, so there's nothing shocking there. The paradigm of the game on those platforms is very different.

Are there any thoughts on bringing The Witness, or even Braid, to the Nintendo Switch?

There certainly have been thoughts in the past. There's something I could say there, but I'm not sure if I can say it. You know, we would consider it. We're not doing that right this second. We're a very small studio and you can tell this iOS port took us a while. I think we did some other stuff before that, like we did the PS4 Pro HDR patch and upgraded stuff, and with Xbox One.

We have limited work power to deploy for various things, we just need to know if there would really be a sizeable audience there, and we're not sure about that right now. I don't know; if it seems like it would be a potentially successful thing then we would look at it, but we've also been working on this game for a while. We're kind of ready to do new stuff, so it’s not high on the priority list, but I'm not saying we won't ever do it.

Speaking of Braid and porting games to mobile, any thoughts on bringing it to iOS or other mobile platforms?

I thought about it lately. Actually, a long time ago when Braid came out, iOS was new, but there was a possibility of bringing the game to that platform back then. And I decided not to do it because I wasn't convinced that the controls could be good. Because I've played platformers of the time and they were all kind of crappy. I'm not sure if I'm convinced of that anymore, I think maybe the controls could be playable. We don't have any specific plans at this time for that, but I wouldn't rule it out.

It’s been over a year since The Witness released. Now you're able to go back to it, bring it into a much wider platform. How do you feel overall about? When you think back to The Witness, what do you look at and say ‘wow, we really nailed it right there’? Even your proudest moments or when you've heard people describe their experience with The Witness?

There's sort of two aspects. One is just that we managed to make this game, it's a really complicated game where the world is very interconnected. So many things, both influenced and uninfluenced by other things in the world, have a significance of sorts, that the player won't understand until later, and that significance makes them difficult to design and build. The fact that we managed to do that and make it all work is great. I don't honestly know if I'm ever going to do another game like that again. Maybe I will, but it won't be for a bit. It was very challenging. It was the most challenging design that I knew how to do at the time. And then from players' experience, it’s interesting for players to feel that.

You see an instance of something becoming meaningful in a way that you didn't expect, and then another instance of that, and then another instance of that. At some point, it builds up a bigger, more abstract pattern and you're in a world that’s meaningful in a way that you never expected or not used to. That’s very subtle, and whenever people say that they had that experience, I'm very happy with that. I know that we managed to do something that’s not only subtle, but something that most video games don't even try to do. It's nice!

Do you have thoughts you wanted to add about bringing The Witness to a larger audience? Or is there anything you want to say about your future projects, or anything else you're working on?

It’s a little too early to talk about future stuff, not showable yet. We do have future projects, so I don't think that's a surprise. We've been working so hard on getting the game ready that I don't know what to expect from people being able to play it on iOS at all. On the one hand I'm even hoping that anybody plays it. Obviously I think it's a good game and it's the best that I knew how to make, but we have this world where all these iOS games are free to play with microtransactions, and this isn't that. It’s old fashioned, you pay some money and get the game. And broadly speaking, that kind of game is a minority of the iOS market. People just don't pay for games on the platform as much, with some exceptions. We have to see how that goes.

I have no way of predicting how that will go. Apple is certainly giving us some great featuring. We're right there right now, on the front of the IOS 11 store. So we'll see what that turns into. In terms of just people actually playing the game, I don't know what to expect. We took this PC and console game, we changed the controls around and hopefully it’s easier for people to deal with and understand. But we don't really know, we'll find out.

2017 has been a great year for games. You want to shout out anything that you've noticed or been playing this year?

Dude, I'm playing a lot of [Playerunknown’s] Battlegrounds lately. Maybe a little bit too much. Everybody's playing so it's not a surprise. A really good game came out a couple of months ago on Steam called Fidel Dungeon Rescue that I really liked. It’s sort of like a dungeon crawler roguelike. It's a little bit inspired by The Witness in a sense, it’s hard to explain because it's very different gameplay-wise, but then you can see why. That game’s good.