The first annual Unite Developer Conference in 2007 had 70 attendees. The company was eight-people strong, and it couldn't afford to pay the catering bill.
"It was something like $10,000. It was horrible," says co-founder and CEO David Helgason, laughing nervously. "And these caterers weren't rich people."
For Helgason and fellow co-founder Nicholas Francis, Unity's CCO, the recollection causes obvious discomfort, but that is evenly balanced by their obvious pride in how far the company has come. Today, Unity has more than 150 employees, and one can only assume that the caterers are paid in advance.
We are sitting in San Francisco's Masonic Centre, one of a grand city's grander buildings. In the rooms below us some 1200 developers are revelling in the robust features of Unity 3.5, perhaps the most significant update to the engine since 2.0. They are a diverse collection of bedroom coders and mid-size studios, educators and business people, covering every type of game on almost every available platform.
When Helgason, Francis and the absent CTO Joachim Ante founded Unity with a goal of democratising game development, this is probably what they had in mind.
"[Other engine companies] are now a lot more open," says Helgason. "We would like to say that they copied us, but we did it first. We always had the community approach, but with everyone else back then you had to sign an NDA."
"You couldn't even get a price quote, right? It was crazy," adds Francis. "There was always this market. It's just that somehow most of the games industry didn't see this as being part of the games industry. And we were just saying, 'Well, why not?'"
If you work at Unity you should get a forum account, get a Twitter account, go out and talk to these people
Lucas Meijer, Unity Technologies
With version 3.5, Unity is pushing towards AAA. The opening keynote is a breathless rundown of the update's long list of new additions, most of which stir some small corner of the auditorium into a minor frenzy. If Unity was run by a more ruthless group of people, Helgason says, they could have spooled these features out into version 4.0 or 5.0, but in most cases the changes are direct responses to the community's requests. When any given feature is mentioned, Helgason knows specific people in the crowd to watch for the most excitement, or the biggest smiles.
This close relationship with its customers is the driving force behind Unity's success, but as the company grows maintaining that bond is becoming more complicated. The goal of 3.5 is to make the engine as useful to 100 people as it is to 1 person, and the top end is only likely to increase in the future.
However, Helgason insists that the features designed for larger studios are purely optional, and can still prove beneficial to the sort of customers who helped establish the engine. Besides, the sheer breadth of 3.5 is evidence of how Unity can leverage its size to offer more to its customers, and faster than ever before. Not only does the company have a greater number of engineers working on updates, but it now has capital to invest in major improvements.
Unity recently bought Mecanim, an animation studio whose technology will be integrated in the engine. It is the company's first acquisition, a step that Helgason and Francis didn't take lightly.
"We've often joked about big companies and how stupid they are: they're never timely, and out of panic they buy something to fix a problem. Maybe this is a small example of that," says Helgason. "We thought about animation for a long time. We knew it was important. Unity has a pretty decent animation system; it's just not awesome, and it doesn't really think forward in time."
"And then we did what big and medium sized companies do. You put down money, and in a few days it's like time travel. We could go back 20 years and work in animation for that long to get to this point. These guys have thought about this and really understand it in a way that we don't, but they have now become part of the team, and so now Unity knows how to do these things. It's kind of magic."
Helgason and Francis won't be drawn on their plans for future acquisitions, but they insist that Mecanim was a special case. Yes, it solved what they saw to be a problem with their product, but if that was all they would simply have licensed the technology. The more important factor was that the team at Mecanim fit in with Unity's open, communal culture.
"You'll find our engineers talking to people and also responding to the feedback we get," says Lucas Meijer, who joined Unity in 2009 to help spearhead its Flash initiative. "When we hire new engineers one of the harder things we have to teach them is that, because in other jobs you maybe just have to do what your manager says."
"If you work at Unity you should get a forum account, get a Twitter account, go out and talk to these people. If you're making a feature you should go on the forum and ask for advice. We really encourage engineers to be part of the community instead of code-monkeys."
Unite 11 is the first time Meijer and his colleague Ralph Hauwert, a Flash veteran who joined Unity in August, have shown their work on the engine's Flash export function to the public. Unity developers will soon be able to make, say, an iOS or Xbox Live game available on perhaps the most ubiquitous and accessible format without the necessity of changing their code. This, we are told, is a big deal.
It is still a work in progress, but the simplicity of the Flash export function and the results it produces - detailed 3D graphics, realistic physics, complex lighting - is enough to draw spontaneous applause from the audience.
For Meijer, however, the best evidence of the demonstration's success could be found in the sheer volume of questions that followed.
"I felt great. I got the feeling that people were very happy with the answers to their questions, and there were a lot of questions. It took a very long time, but the questions were on such detailed stuff that it tells me we got the big picture very much right."
When Adobe announced it was adding native 3D support to Flash, Unity saw it as a gilt-edged opportunity to address the wildly variable conversion rates for its browser plug-in. All Adobe had done was add functionality; this was a chance for Unity to give its users access to a vastly expanded audience, while simultaneously opening up its tool to a new community of developers.
"What Unity does is being able to write three lines of script and throw a ball," says Hauwert, who compares the "vibe" of Unity to the early days of Flash development. "If you look at other ways of doing that, with 3D State and Flash, that's going to involve a lot more work and a lot more code for the developer."
"I mean, yes, there is Flash, but Flash is 2D, and being able to get the Unity iteration cycle, get the Unity way of scripting a game quickly, and not having to think too much about rendering unless you want to - I think that brings something truly new to Flash."
What our users can depend on is that when new platforms emerge that matter to them, we will be there
Lucas Meijer, Unity Technologies
"One thing I've learned is that people don't care too much about how technically cool it all is, but how it looks, how they can work with it, and that the tooling is right."
Both Hauwert and Meijer seem genuinely surprised at the results they are seeing from their work. Despite being more ambitious than almost anything attempted with Flash before, Hauwert says showcase games like Shadowgun and Angry Bots are running "an order of magnitude" better than expected.
"I mean, people are used to this performance from the Unity web player, but now you can offer it to audiences who really need a lot to reach," adds Meijer.
This seems like an opportune moment to summon the spectre of HTML-5, which prominent companies like Apple and Facebook are evangelising while simultaneously hinting at the eventual demise of Flash. Advocates don't come much bigger or more influential, but Meijer believes the rivalry between the two technologies has been blown out proportion.
"We make the product for our users, and one way users often get fried is that they put all of their eggs in one basket, for one specific platform," he says. "And then it turns out that, when their game is done, that platform is no longer the new kid on the block. The place where you can make money, the place that is popular, it changes a lot. In the App Store you could make a lot of money, but now we see that it has become much harder."
"The biggest thing that we care about is that our customers don't need to care about it. We think the first platform that will be viable for bringing 3D to a wide range is Flash. We also look at Web GL, but right now we don't see it as being as ready as Flash yet. If that changes we could look into it, and if we do depends on how everything goes with our users."
"What our users can depend on is that when new platforms emerge that matter to them, we will be there."
This dedication to giving developers access to as many platforms as possible is evident in Union - a service designed to introduce Unity games to places that its existing customers might struggle to reach. The App Store, the PlayStation Network, Steam, Xbox Live; these services are now valid and relatively accessible routes for any studio seeking an audience, but the sheer number of games being made has greatly reduced the chances of success.
Union addresses the problem by targeting nascent technologies like the Roku 2 set-top streaming box, and smaller competitors in established markets like the Blackberry Playbook, the Nokia N9, or Sony Ericsson's Xperia Play mobile phones. In a sense, Union closes the circle on Unity's mission to democratise the development process by allowing the company to act as a publisher, but vice president of business development Oren Tversky shrugs off the label.
"We don't think or ourselves as a publisher," he says. "We see ourselves as an aggregator, or a syndicator, or something like that."
"We have some advantages over other syndicators, one of which is that our games share the same code-base... If you think about a traditional publisher, they might have code-bases from ten different studios, internal, external, all sorts of things. Unity games share a code-base, so we have some really interesting efficiencies there."
Tversky acknowledges that Union is "in part" a response to the increasing difficulty in finding success on the App Store - the most popular platform among Unity's vast community of commercial users. Union structures its deals with platform holders to include "preferred marketing slots and other opportunities."
"They're incentivised to make our content fly," he says. "Videogames have always been a hits driven business. The rise of the direct-to-consumer platforms like iOS has sort of increased that, because all of a sudden you have much more access and the barriers to entry have gotten lower."
"We can kind of democratise that, at least for our developers on these newer platforms, because we're in a privileged position to bring over content."
However, the partnerships that Union has signed so far face an uncertain future. Research In Motion is going through a very well publicised rough patch, including service failures, huge layoffs, and the poor performance of the Playbook, one of Union's destination platforms. The Xperia Play has also failed to shine at market, and while the Nokia N9 was well received by tech reviewers, Nokia announced in excess of 7000 redundancies this year and posted an operating loss of €500m for the six months ending July 2011.
"We call them new and emerging platforms for a reason," Tversky argues. "They're not all necessarily going to become the next iOS."
"We deal with it in a couple of ways. One, we diversify; you've seen three or four announced deals, and we've got another four or five in the pipeline. Maybe not all of the platforms are going to be huge, but if we've got nine of them and we get on a [successful] platform early it's a huge opportunity for our developers."
"The second thing is we work closely with the platform owners, and structure the deals in such a way that we're covering some of our risks."
Some other engines...you can build an absolutely amazing game with them if you have an amazing and a large art team. You don't necessarily need that with us
John Goodale, Unity Technologies
Tversky estimates that, right now, "about half" of the deals for Union are for mobile, and half are for connected and Smart TVs. When asked about which holds the most promise for Unity's development community, Tversky doesn't hesitate.
"On mobile it's got really good potential. On TV I think it's got huge potential. It's a more open market and we have a lot of interesting advantages. From the TV side I can see Union becoming absolutely huge... But that market doesn't exist today. It's still 2 to 4 years off."
A striking thing about the way Unity deals with the press is how readily its emissaries are willing to discuss the future, and there's a very good reason: the company is growing so quickly on so many fronts that virtually everything outside of its core product is in its nascent stages.
The company's expansion into Asia is another example. John Goodale, who took over operations in the region earlier this year, claims that the company's Asian business is growing at a rapid pace: a revenue increase of 258 per cent by the end of August 2011 over all of its Asian revenue in 2010. Currently, four of the top five cities in the world for sessions logged on Unity are in Asia, and Goodale believes this is just the beginning.
"Before I joined, the company was largely reactive in Asia: it got enquiries and dealt with them as they came in," he says. "Since I got on board we've become very pro-active in that space. That has made a big difference. In July, we opened Unity Korea, and we staffed it with three individuals. In September, we officially launched Unity Japan. We currently have two employees and are aggressively hiring additional support staff."
Though all of Asia's regions are different, in China, Japan and Korea direct relationships are an absolute necessity of doing business. Goodale describes the presence of official Unity representatives on the ground in the region as "huge" for its future growth. The company can now deal with the local community in a common language, and get a clearer idea of what it needs to gain a better foothold in the region.
"We're keen on identifying trends and then marketing and selling specifically to those trends. Those are different country by country," says Goodale.
In China, there is a "huge flight" from PC client-based MMORPGs to browser-based MMORPGs, which Unity's forthcoming Flash integration is well placed to serve. In Korea, the flight is from PC-based MMORPGs to mobile phones, the platform where Unity is arguably strongest. In Japan, Goodale believes the games industry is being driven by social games, and Unity is already working closely with both DeNA and Gree to reach out to the local independent development community.
But as much as each country needs special attention - and Unity's Asian roadshow has already added Thailand, Malaysia and the Philippines to its promotional tour - Goodale believes that Unity has one quality that will ensure its success, and make it a more attractive proposition than any other engine company trying to break into the region.
"The number one reason why we're successful in Asia is also applicable elsewhere, and I would call that scalability," he says. "You can use and be successful with Unity if you're a single person working in your garage, or you can be successful with the technology if you're a huge developer, and we have that kind of customer in Asia."
"So the technology scales, but the business model also scales, whether you pay us nothing or whether you pay us a million dollars, and that works worldwide."
"Some other engines, without mentioning any names, you can build an absolutely amazing game with them if you have an amazing and a large art team. You don't necessarily need that with us. They're all trying, and they're all successful in their own rights there, but I would be surprised if they were having the same kind of growth we are."