The Unity 3 Interview
October 11, 2010 Page 1 of 7
The Unity platform has quickly become a very popular engine for game development for the web and iPhone. Next up: Android. And the 3D game engine middleware and tools may be becoming relevant on higher-end platforms like PlayStation 3 and Xbox 360 as well, as the team at the company readies these platforms for inclusion in the latest version of the tech.
Two weeks ago, the company released Unity 3, which includes many new features -- notably among them, a new default project called Boot Camp. It also includes enhancements to its lighting, audio, and other systems aimed at making it more of a viable choice for high-quality 3D projects.
Now serving 200,000 users, according to the company, the five year-old Unity engine has benefited enormously in recent years from the explosive growth in browser, social, mobile and online game development.
As it unveils Unity 3, it's celebrating a new honor: the Wall Street Journal has awarded it its Technology Innovation Award in the software category.
In addition to many indies who use the low-cost engine, Unity lists Bigpoint, Cartoon Network, Coca-Cola, Disney, LEGO, Microsoft, NASA, Ubisoft and Warner Bros as major clients -- along with Electronic Arts, with which the company just signed a significant multi-year license deal.
Gamasutra sat down with CEO David Helgason and product evangelist Tom Higgins at the company's San Francisco offices to get a preview of the latest version of the engine and have a wide-ranging conversation about the company's, and the technology's, future direction.
Did you achieve what you set out to achieve when you sat down and said, "This is what is going to be in Unity 3"?
David Helgason: That's a good question; there's a couple different ways to answer that. One is: yes. (Laughs)
We wanted 3 to be such a massive release. We rewrote the entire rendering system, optimized it very heavily for forward rendering, and then added the whole deferred rendering on top of it so that if you're on modern hardware you can have innumerable light and all this cool stuff, putting Unity on par with basically other AAA engines.
We licensed this very expensive middleware in bulk. You get Beast, which is a super expensive package normally, as part of Unity pro, and the same with Umbra -- so two incredibly expensive packages just going into Unity and really drove the nail in.
Everyone recognizes Unity engine 2 as a really cool platform for smaller games; Unity 3 is a very cool platform for anything, really. So yeah, we did.
The other answer: Of course not. There were features --
Tom Higgins: There were features, there were more that we wanted to get in, and as jam-packed as this is there were a couple of other features that we really would have liked to get wrapped up and in 3.0; but we had to make the cut at some point.
DH: We failed! (Laughs)
TH: I think for us, internally, we're a bit of masochists and we're like, "Oh! Here's what we missed!" So there were a few things like that, but the bigger picture: Is 3.0 as cool and meaty and gravy as I envisioned it when I first saw that list? Yes. Did a couple of things have to get taken off the list and deferred to later in the 3.x lifecycle? Unfortunately, they did, and that's just the hard life of development. You can't always put in everything you want.
But on the whole, is there an ounce of disappointment in my body about this release? Absolutely not. It's just, like I said, in those masochist moments when we need that little extra whip to keep going -- there's more to be done. We could have crammed a few other things in, but not at the sacrifice of quality.
DH: The temptation now would be to start doing Unity 4 and have a really long release cycle again, but because we have some of those things almost ready we will be doing point releases, getting the good stuff out before we go into the long haul again. We want to get this stuff in the hands of people really fast; also, we want to maintain our generosity.
We've been really generous in the past, making many point releases. Where most companies would just put out a point release with some bug fixes, we would put out a point release with a shitload of bug fixes and a major feature.
TH: It's been three years since 2.0 came out -- it was October 2007 -- and there's been a lot of meat added in. Of course, that was stretched because of the Windows launch 2.5 in the middle.
DH: Which probably took more than a year out of our calendar just for that.
TH: Yeah. But we definitely like the idea of, again, bug fixes and features. We want to keep giving people that value so that they're committed to us as a platform, and we're committed to them as users. There's definitely going to be some more coming out, and at some point we'll start to turn our minds to 4.0. But we're not there yet.
As more and more potential uses of Unity drive social games and different phone platforms and consoles, does that really complicate it for you guys in terms of what to include? How do you determine what audiences you serve with that?
DH: It does complicate it.
TH: It makes the wish list longer. If you're serving one narrow market, it's very easy to kind of focus on them. But I think one thing that we're doing that's a bit different from other companies is that we're really staffing up with unique teams for unique platforms. There's not one team that has to spend their time with desktop and web and then turn their mind over to iPhone and then turn their mind over to Android. We've got multiple teams working in parallel so that each one can focus on that platform's needs.
Then, as we talk with new and different types of users across all of those, that's where I think the extra complication comes in; but we're here to solve the complicated problems. It just takes a little extra time and a little more forethought. We're not used to thinking about architecture visualization and what they might need. We're going to have to spend some extra time doing the research and looking into it, and the nice thing is that there's so much overlap between these markets. It's not unsolvable. It's just going to take a little extra time.
DH: One thing that simplifies it, I think, is that we kind of try to focus primarily -- almost only -- on adding enabling features and extensibility to the platform. Then, because we have this big community and so many types of people using it, they then build the extensions and the functions they need and often share them with others. That sort of simplifies the problem.
For instance, should you have a really good Facebook integration? Well, there are several people in the community, including some open-source and sort of commercial extensions, that can do that and probably do it better than we would if we just tried to make it as a feature and then forget it because Facebook integration is something that's alive; you have to keep updating it. If we tried to stretch ourselves to every single thing like that, we would probably fail; but we're lucky that we have this amazing community that has that and does that.
Page 1 of 7