Unity's Future In High-Definition

Unity has become a cornerstone of the indie and smartphone game space, but does the tool set have what it takes to stand up to "triple-A" developers' needs?

This is an initiative that was first spelled out by CEO David Helgason when he took the stage in San Francisco this past September at the company's Unite conference.

"As we grew as a company and our stature in the market grew, it turned out people wanted more," he said at the time. "It turned out that Unity has fallen short in some things," but "we realized around a year ago we actually had the energy to fix this." To that end, he said, "We recruited people behind the leading triple-A engines today."

Since last July, in fact, a small splinter team, based in Stockholm, Sweden, has been working on the initiative. Their job is to figure out what developers need, and if the engine isn't where it needs to be, they will make the effort to make sure that it gets there.

Erland Körner, a technical artist, and Renaldas Zioma, who was mobile tech lead on Unity before moving into the triple-A initiative, both worked at EA's DICE studio, which is also in Stockholm, prior to their tenures at Unity, and the two are interfacing with teams both big and small to determine what features the engine needs, and what implementations of these features will serve the broadest user base.

In this interview, the pair speak extensively about how they interface with developers to determine these needs, why they set up in Stockholm, and just what they hope to achieve with the tool now and in the future.

At Unite, David Helgason talked about the triple-A initiative, but it's been a little bit quiet from the company.

Renaldas Zioma: I don't think we like the name "triple-A," per se, but it's more "high-end features." One part of the initiative, you might say, was directed on the iPhone side, so something like Shadowgun is an example.

We considered the list of high-end features for the iPhone platform, and there are many things developed for that that are not in the public release currently. They are coming in the next release [Ed. note: Unity 3.5, which has been released since this interview was conducted], or some of them may come later, afterwards, when they are polished. So, it was a quite significant internal effort for us.

Erland Körner: We are working with external developers on other projects as well. We started with this just recently, so I think it's not gonna happen overnight that we will get these releases to the public. But I think within shorter than we believe, there will be something for the public to see of this.

RZ: For us, it was more internal directions that we take for things which are not immediately seen in our public releases. But it's something we're doing internally -- whether it's refactoring our code base or working with external teams to find the weak spots and fix them first, instead of going full-out with the public features. First we are looking for the external teams to try it out. To work with them closely, to fix that stuff, or to implement certain functionality.

If "triple-A initiative" is kind of a misnomer from your perspective, what is the goal of your work, then, with the engine? What are you trying to achieve?

RZ: It depends what you call triple-A, of course. What do you mean by saying "triple-A"? That's why we don't really like it. For us, what does triple-A mean? It's very undefined.

It's a term that gets thrown around in the industry without a lot of thought behind it.

RZ: It's hard to answer this question, specifically.

Renaldas Zioma and Erland Körner at Unity's Stockholm office.

What is the goal of your work, with the engine? What are you trying to bring to it that it doesn't have?

EK: I guess we're trying to stretch more towards high-definition projects, in any sense. We're trying to open developers' eyes by doing much more challenging projects, I think, using Unity.

RZ: That [term] was something towards external developers -- yeah, we want to push some limits and show how that can be done in Unity. On the other hand, I would say we are looking more towards how artists are working in the bigger teams. Analyzing that and making sure we not only cater for indie developers, but for bigger teams, the way artists work in a bigger team, we analyze that. Even just the features -- to make it more intuitive for them to work.

EK: Streamlined for handling bigger projects and substantially larger amounts of content and such things.

RZ: The particle system, for instance, the new one, it's an interesting example because it was born while working on Shadowgun. The reason for it to be born was because people weren't happy, there wasn't enough artistic control of the particle system previously, in Unity. So, we tried to attack that problem specifically, to be more art-friendly.

We tested it on Shadowgun for a little while, but then we had a small group, we we selected individual developers with quite a lot of experience -- not from inside Unity, but outside. Customers. We've gotten quite a bit of feedback on how to make it more friendly. Intuitive.

Something I've been having a lot of conversations with about these new features in Unity is how adding them helps everybody. Just because things aren't necessarily targeted towards "triple-A" -- and maybe that's why you don't like that exact term. Improving the engine helps everyone, right?

RZ: Yes, we think so. That's the reason why we don't release some things immediately, even if we have implemented them internally. Anyone can use it, so it's not only for the 20 people who already spent a year with Unity, so only they could use it. We don't want that. We want indie developers to be able to benefit from that as well, as much as possible.

I do get the sense that the technology is starting to be taken up much more rapidly, and by studios as well. I don't know if there's a "triple-A" $20 million game in Unity. It seems like a matter of time, if you guys keep along this path.

RZ: There are certain MMO games you could say they are very expensive, so I don't know if you can count them or not.

The workflows of indie developers and larger studio developers are much different, right?

RZ: I don't know. From some point of view, yes.

EK: But I think the larger teams using Unity benefit from the fact that they have an engine and tools already, and that scales down the team fairly massively, I would say. So they can still manage to develop games for consoles with a fairly small team. I think in some way we're defining what triple-A teams are, as well, by providing a very smooth toolset for triple-A developers.

RZ: Some people, now, in Unity, are coming from backgrounds working with big teams and big budgets, what you call triple-A. I think what you see, is you'll realize the workflow that big teams usually have, it's not like people are in love with this workflow -- these big teams. It's just that they're just rolling this way. For them it's easy to just say, "Ah. Well, whatever. We already invested one year in this workflow and everybody already knows. We don't need to optimize it to be more intuitive."

Usually how the big teams work, the people themselves are not really happy about that, usually. It's just that, "Well, it's good enough," and they continue. When it comes to Unity, we try to do it differently. We want when you work on a bigger projects, you feel the same indie feeling to it. When we talk even with very experienced people who are using Unity, they use this term "playful" or "fun." They actually feel work is fun now, instead of just being work.

EK: Most successfully within gameplay -- I think that's something in Unity [that] games that are growing in. It's not as outlined as it usually is in a triple-A project, that you document it first and then try to achieve that result. In Unity, I think, gameplay has a tendency to evolve in a more playful manner.

That's because of the tool?

EK: Yeah, I think so. Rather than building technology many times from scratch, they can start elaborating very early on, on the mechanics.

RZ: With things like that, it's probably more interesting to talk to actual developers than with us. They would probably explain it even better. But that is something we hear, from quite some experience with the work. They enjoy working in this way.

So, we kind of want to be the fuse -- with Unity, traditionally, it's easier to enter and easy to make a small game for indies. We want to at least drag as much as possible of that feeling towards the bigger teams. And then the big thing is functionality. We are more implementing [features] for the bigger teams, and dragging that towards a little bit more user friendly, and make it more indie-friendly, or smaller team-friendly.

It's making the high-end features accessible to indie people. Trying to mix things up, so these divisions are fuzzier.

RZ: Yeah. And I think you can see that nowadays. Quite a lot of people are going from triple-A to form their own studios and make some small games. Some small companies go to web and start to multiply, becoming bigger and bigger. You see lots of people going in both directions. It makes sense for us to try to blend between those two groups, and get the experience from them.

It's not like we have this certain goal to blend between or drag. It's something that happens naturally for us -- there is a certain amount of people very active in giving the feedback to us. We are working with certain teams, like Madfinger, that give feedback to us, and influence us in a natural way. It's not like we have this goal set, ourselves. It's more what we actually see happening.

Rather than competing with technologies like Unreal, it sounds more like you're looking at the demands of developers.

RZ: Yeah. That is something we want to do. We try not to look at them. I mean, there is no way you cannot look at that. Of course you see that, but we try not to overanalyze [and] just implement something because they have.

EK: Our overall goal is to always provide the smoothest and most easy-to-use tools for developers to make games, and we don't interfere with cultures or try to dig too deep into game design, or influence how game makers are developing their games, but more provide the means for them to make the games as quick, and flexible, and joyful as possible. We try to listen in as much as we can. It's a bit of a learning process.

It sounds like while you're not trying to dictate anything -- what you want to do is let them influence you.

RZ: To a certain degree, yes. Of course we have...

The classic quote you hear all the time lately is that Henry Ford said, "If I asked my customers what they wanted, they would have said a faster horse." So, you can't exactly…

RZ: Of course, you can't really. If we just listened to what people want, we would have to do everything in various ways. [laughs] It probably happens. We listen to feedback. But then we have certain ways to pick certain directions all the time.

EK: That's why these triple-A developers become a very good source for us to listen to, I think. They tend to have a lot of experience in using a variety of toolsets, and developing their own proprietary tools, maybe, before that. That's why it's a very simple way for them to directly communicate with us, and try to get those features that they're accustomed to into Unity.

I think it was [Unity COO] Nicholas Francis who said, looking at feedback, something along the lines of "people are looking for specific things, but what we have to do is take a step back and figure out how these things correlate." They might be asking for one specific thing, but if you go back a step, you can see that it's people asking for different things that can be encompassed by the same feature.

RZ: That is a plus. It probably was in the past more that way: at some point we were kind of scared what would happen if we had a big customer, and we didn't want to have two big customers who would influence us too much.

But once we had quite a lot of small customers, and nowadays we have various sizes of customers, and since we get quite a lot of feedback, we can see these general directions in certain ways. Like, bigger picture in some ways, which helps to rethink and maybe do something not exactly how people asked us, but something hopefully better.

EK: Something a bit bold. Something that we can continue building on as well. A feature that is a new take on it, but still will still supply the demands they have.

RZ: If you would look at our process, how many alpha phases and beta phases we have usually, for a certain product -- it's probably kind of public information, but if you're on the beta list you know how many iterations we go through.

Time-wise, it's quite a long process before the feature goes from alpha, being presented to some people, then to the bigger beta group, and then ends up in the public release. There's quite a lot of iterations, which we get feedback. We get almost complete feature functionality, and then people bang on it, try it out, give feedback, and we see if it makes sense for us.

It's a lot of iterations, where we're getting feedback and iterating on this. Of course, quite often, there were situations where we did something quite differently from what people expected, or what they wanted us.

At the end, what matters for us is to have actually good features and that people are using it. We didn't want to cater them specifically to cater for them. We wanted to solve a problem for us, most importantly. We want to solve a certain problem, and then people can use it.

Solve a problem in a way that creates an effective tool.

RZ: Exactly.

Regardless of how it's been done before, or if it's something people were anticipating.

RZ: Exactly. That is our ultimate goal, to achieve that.

You're working with Massive Black on Mothhead. You said you worked closely with Madfinger on Shadowgun to see how certain systems in the engine would function in a game like that. It seems like these projects help you figure out how things are working.

EK: I would say Mothhead was a really good experience for us. Because we worked with artists that are not necessarily technical, in some ways. There are some people at Massive Black that are really technical, and some that are more or less painters, and it was really useful for us to really listen to their experience and how they are used to working with this, and also how a person with a complete artistic background is able to achieve something in Unity. I think that proved a lot of things about Unity, how smooth it is to use for someone who has doesn't have any technical interest, or technical background.

RZ: Shadowgun was a different example, because we worked very closely with them. But they are more technical, of course, than Massive Black. They already shipped games in Unity before, and they were very technical. So, the Shadowgun was more of a battleground for certain functionality, like the path navigation.

I remember the demo for Shadowgun at Unite was all about lighting.

EK: The particle system.

RZ: It was more to trial new features, but we worked closely to figure out the interface and what they actually need, how they would use it in the game. We had an idea with the light probes, but it was actually changed a couple of times in how it was presented, how it's structured, and how easy it is to use. Just put it in the game.

Madfinger, when they started, they were four people, and I think when they finished the project, they were seven. They were a small team, and it's very good to test on them, because they don't even have time for anything. All the features we give them, they have to be easy to use, and take as little time from them as possible. That was a different experience, but very interesting as well, working closely.

EK: It differs a lot from working with Massive Black, where we didn't develop that many new features for them. We were just bystanding, and answered questions when they had any issues, and gave them some hints and advices here and there. I think they managed by themselves pretty far, in fact. Unity didn't fill the same role as with Madfinger at all.

RZ: Well, Madfinger helped with optimizing as well. The project itself, it was a good battleground for optimizations for us. It was completely different how we work with one team, and how we work with another.

You just select different opportunities that you see. I imagine, to an extent you say, "Well, you're working on something cool, will we get something out of it if we pay attention to what you're doing?" How do you select who you're going to interface with on a deeper level?

EK: I think every project has some sort of interest. We come across them in very different ways. We try to pick out what the features are…

RZ: It's not like we have an algorithm. [laughs]

It's more about what's going to teach you something.

RZ: Yeah, exactly. It's not like it started to happen… We had it before -- to a lesser extent -- but for the iPhone when we started. I remember when we started, we had early beta releases. We were helping people quite a lot, just because people were coming from different backgrounds. No one had mobile 3D experience -- almost no one, you could say that. We had to help people quite a lot back then. I think we established certain links to certain developers.

I think Unity, even before that happened as well, was working with very closely with certain developers, just because Unity was very small. It was easy for a developer to talk to someone at Unity. I think it was happening all the time to some extent, and now we are figuring out a way to be more productive with that, and get more information out of that, and be more direct. We can actually benefit quite a lot from that. It's easy to get the feedback in a big chunk.

We are looking for someone -- for instance, with Madfinger, they wanted to release that game. It's not like we're giving them a feature they don't really want. That is the feature they want, and it's the feature we wanted to test, to get it better. It's a mutual interest from both parties. Finding people like that is always, of course, hard. We want them to be in the same time zone, for instance, so we can get feedback really fast and be able to sometimes meet in person.

EK: Good chemistry is really useful. It's not a criteria. Definitely not. We enjoy collaborating with these developers, so I think the most fruitful projects come out of a good collaboration, and the better connection we get, the more useful it will be.

I think that was a good case, from Mothhead and Massive Black, where we had really natural communication with them. It was really fluent, and smooth, and joyful, I think, enjoyable to work with those guys a lot. The same goes with the guys from Shadowgun.

I think selecting these projects comes very naturally. We have a good connection with our community, looking at all these community sites where we're always participating, and in general how close we are working with and helping out the users of Unity. It's a very direct communication we have with them.

RZ: Just coming to Unite helps, at the same time, because we meet a lot of developers, and we can see how that works, and we'll just talk.

EK: I guess the hard part about finding the developers is that there's so much going on right now. It's a lot of noise, when it concerns the developers. Almost everyone I speak with is having some sort of interest in Unity at this moment. Until you see something substantial, it's hard to determine whether it's worth putting in the time at this point. Everyone is speculating.

Why open an office in Stockholm rather than go join the team in Copenhagen? Is it a strategic thing, or just because it's where you are?

RZ: There are different reasons for that. If you look at Unity, that's how it happens. We have quite a lot of offices around the place. Opening just our office is a bit of a headache for our financial guy, but it's kind of natural for us.

RZ: There's [also] a strategic reason for that. [laughs]

EK: There are many successful studios here, naturally, so it's a good surrounding to be in, I think. There's a lot of things happening here.

RZ: It helps to get focus, as well. If everyone sits in one place, I think you can get... Because one office is working on mobile, and the headquarters is more core, and here we can concentrate on certain things as well. Maybe sometimes it's even easier to concentrate on certain things when you have are smaller group while being in the same company, but you have this advantage of being a small group. And traveling, it's like a one hour flight.

EK: I think having these separate offices -- I found it a bit different when I began working here, but it's actually beneficial in many, many ways. I think once you reach the other studios, you're much more respectful with other people's time than patting people on the back. You set up your communication, and you prepare it a bit more than you would do naturally. I think it's, in some ways, much more effective to have this spread out among the studios.

RZ: In certain cases, it can be more helpful to structure your time. Of course, one big thing is it's easier to hire people this way because -- it's a bit probably different in Europe than in the States, probably, that people actually don't want to travel.

They don't want to move.

RZ: Yeah, because they have family here, they have a different language here, and it's a bit of problem to move to Denmark maybe, just because there's a different language, for instance. People do understand each other, but...

But we have people from Germany, from Holland, and many places, so we don't really want to move them all the time, if we can keep them where they mostly have family, and are enjoying, because they're being more productive as well, instead of shuffling people from one place to another place.

And it helps to be able to build better ties with the local game development community. Another day you visited Might & Delight, and some small companies here in Stockholm using Unity. So we can go and talk to them and sometimes even help them even.

EK: Schools, it's the same thing. Game development universities in this area are growing. There's a lot of respectable game developers in the Stockholm area, and from that, it's become a very likely career choice for people in the Stockholm region, and a lot of talent is growing here.

Personally, I think it's a surprising amount of developers comparing with other Scandinavian countries, just in Stockholm. It's sort of a hub that has been created here, for game development in this region. Not only finding talent for Unity, it's perfect being around here.

Massive Black's Mothhead

Unity is doing something that I don't think many companies are doing: building an engine that's like this, and building an engine this way, both. Neither one of these is really being attempted, I don't think.

EK: There's many core differences in Unity, I think, from the culture of the company to the actual talent that's working there, that just differs from any other company that I've been working at, at least. I think that philosophy that Unity has is something you can actually feel.

The Unity philosophy is the democratization of game development. That's the succinct way of putting it, right?

EK: It's very exciting being part of it, I must say -- how Unity is progressing the mantra of democratizing of game development. I think not only can we provide good tools for developers, but I feel that I can be part of evolving game design, which has almost staggered the last couple of years. You can almost see it on the console market, these repetitive games sort of that are being developed in genres. I think looking at the games that has been done using Unity, it sort of differs. It's fresh. It's more playful.

You see all these people getting their hands dirty and you see all the weird shit they're doing, and you realize there's more out there than the tunnel vision you might have, if you're in a major publisher or a major studio and you're just trying to ship.

RZ: There's much more tunneling in the bigger industry, especially from a technical point of view. For three years, you know the direction of that tunnel and you dig in it for three years, maybe turning a little along the way. You get feedback once in three or five years, so that changes quite a lot of the process. Much less playfulness when you are digging that tunnel for three years, or five years. You can't really step aside. You can't really experiment that much. There's always a very important feature.

EK: The goals are very clear from start, when you're working on those bigger projects. With us, I think even for these bigger developers, Unity has played some part there. Some of them, I heard, are using Unity as a prototyping tool for working with the games.

RZ: That is something we hear as well, that even though the studios aren't using Unity for the main production, because they have their own tools already and they're invested in that. They grab Unity just for prototyping. [laughs]

EK: I think that's very flattering.

I guess you guys are hoping at some point they'll stop just prototyping in Unity, and move into production in Unity.

RZ: To achieve that? Yes. That's why we want to have this "triple-A" whatever -- I call it "internal" focus. So we can actually find out, if people prototype, the reason why they would not continue working. And either implement the missing features they need, or fix whatever is broken.

But we don't really want to do just that. We want to be more creative on the way. Not only just taking an EA triple-A studio team and working for them -- no. We want something a bit more different. Taking varied input and putting that into the initiative.

EK: I think it's fantastic also to see these teams, how they've exploited the tools to such a degree that it surprises us, almost, the way that they're using Unity. You'll see when you visit Might & Delight as well, the way they're constructing the game [Pid] and how they're using Unity to do it.

This is exactly how I see that people are using Unity. It's so flexible, and they've taken quite a leap to use Unity in a different way, to make their game come true. But it also shows how flexible it is, and how you, if you really tame it, you can use it any way you like it. There's no real root or any specific flow that you need to use in Unity. You can pick your own path.