Austin GDC: Interview with BigWorld Tech’s Gavin Longhurst

Guest contributor Kumar Daryanani, who covered Austin GDC for us, met with BigWorld Technology’s VP of Business Development Gavin Longhurst for an interview. BigWorld Technology is a middleware provider currently delivering technology to a multitude of MMOG projects.

The MMO Gamer: Tell us a little about yourself and BigWorld Tech.

Gavin Longhurst: I’ve been with BigWorld Tech for two years. BigWorld itself is about seven years old at the moment, having branched out from Micro Forte. BigWorld is based in Sydney, Australia, and we have offices in the US, Europe and China.

We have licensed the BigWorld Engine to developers all over the world for several years. We’ve been working on the engine as long as several of the major MMO developers out there.

At this point, our technology is mature – we update it every year, with the latest server technologies and graphics/tools updates. We aim to provide a complete MMO solution – tools, analysis, client: a one stop solution for starting short and longer term, commercial projects.

Our viewpoint is that developing and MMO is not a trivial undertaking. There are people out there that think that if they can get a license for free they can do anything, without having a solid grasp on the realities of an MMO development cycle or the time and resources required to create a successful game or virtual world. There is a market for Indie Virtual Worlds but at the moment, we are focused on commercial companies.

The MMO Gamer: How did the company get started, and why did you decide to make this particular product? What niche did you perceive to be filled?

Gavin Longhurst: When BigWorld started up I don’t think there was a lot of MMO/VW specific middleware available. BigWorld Tech sprang from Micro Forte, a game company in Australia.

One of the big advantages of MMOs is the billing model – with an MMO, if you’re building your own content and deliver through the web, you can circumvent the publisher-developer model, which is one of the factors that pushed us to develop the engine.

The MMO Gamer: Tell us about some of the games currently under development using BigWorld.

Gavin Longhurst: Well, I can and I can’t, as with so many of these things. There’s two games coming live in China soon, ‘House of Flying Daggers’ based on the movie, and developer Tian Cheng is producing another game, Floral Fire Online. And later on in the year we have a social world going live in Europe.

And next year and beyond we have games like Cheyenne Mountain’s Stargate Worlds, John Romero’s Slipgate Ironworks title, UK-based Kwari and their namesake game, and Bitcasters’ Stormhawks mini MMO. So that’s half a dozen games right there, and then several every year after.

There are others also I can’t talk about yet as they are not public.

The MMO Gamer: What advantages would a developer have using BigWorld as opposed to developing their own tools in-house?

Gavin Longhurst: Cool tech, cool tools. But three key things – Money, time and risk. We’ve been developing and redeveloping our engine for over 7 years, and we’ve created a very solid set of tools over that time. There’s very few games that have been in development that long, so this has allowed us to create tools that can cater to the needs of many dev studios.

So using a middleware solution like BigWorld saves developers the investment of having to build their own engine in-house. The second is time – while making tools for your game, you’re not making the game. You’re spending more time on the tools and not the game which limits how deep, polished, and varied your game can be. So having more time allows you to create a better product, a better game.

The third, and by far the most important is risk. Time and money you can secure more of. Risk is the killer. Risk wakes you up in the middle of the night. ‘Do I have the right tech, the right people? Am I going down a blind alley?’. There are blind alleys in engineering, in design, in story. Why go down a blind alley with ‘can my engine run my game?’. With Big World, you know that the core engine is solid, and will run, you don’t have to worry about whether the tech will actually work. We tear out a chunk of the unknowns and give devs the freedom to move forward without doubts of whether they can create the engine to run their game. These are the core tenets of any MMO middleware pitch. What we can’t sell is good design, visual aesthetic or game polish, but with BigWorld, we can give you the time to work on these without having to worry about the tech.

The other thing about middleware, we provide the tools and API’s for development. Nobody’s middleware is perfect, nobody’s engine is perfect – we’ve been working for 7 years and we still find things we can add. There’s a million more ways you can stretch the tech. If you show the code, if you expose your code, with the code at the client’s control, they can add, change and enhance the systems they want, and that’s a crucial point. One of the weirdest situations we get talking to game engineers in a pitch is when they tell us that middleware in general will limit their capabilities. The opposite is true, middleware allows you to put crazy stuff in there. The elemental components are there, and as a developer you can go wild, instead of working on the basics every time you start a project. Once you’ve made a game, you don’t want to have to develop the chassis again, you’ve already made the chassis a million times.

So the use of middleware allows to skip the groundbreaking work, and get right down to working on your game. The idea is in a few days – weeks, you can get the core of your game on-screen and live, prototyping immediately on the same platform you are going to market with.

The MMO Gamer: Do you get much feedback from your licensees?

Gavin Longhurst: Yes. It’s an interesting situation. We have a number of devs around the world, they provide feedback and we fix bugs. We do go through feature requests and itemise and prioritise them, and work on them in each of our dev cycles. It’s always a juggling act with people who want to do things themselves or wait for us to implement it in the engine’s next development cycle. So in general they have the capability to enhance whatever they want with it. If you want to do something clever and different you may still have to do it yourself, depending on our schedule and priorities.

For the most part we have to work with what is most common usage for each feature. We also use middleware for certain things. We work with companies like Umbra, Vivox, and Speedtree, and industry partners like GNI and IBM for hosting.

The MMO Gamer: What advantages does BigWorld have compared to other middleware solutions such as those provided by, for example, HeroEngine or Multiverse?

Gavin Longhurst: Well both of these companies are doing cool things in different ways. Multiverse is doing something very brave and difficult, almost like a creative commons thing. But as they’re finding out, when you’re working on tools and systems, it’s a big investment, they’ve been working for 2 years. They’re showing small game alphas by now which is impressive. The hero guys are doing some interesting work too, they’ve done some neat stuff with simultaneous design space collaboration, something that we do in a slightly different way, and they’re making a game too, Hero’s Journey. Making a game can sometimes detract from delivering the most flexible middleware, because your customer might not necessarily be interested in the same things that your game needs to succeed – but in making a game you also stumble across a great deal of functionality and workflow related to making any game, as we found in the past. There’s a fine line between making a game-specific engine and making middleware that can be used by a broader range of game developers. With few exceptions, most engine middleware has grown out of a game specific project, which makes sense. In the end, we’ve got a lot of experience working on our tools, so in a developing sense perhaps we’re further along in the cycle. We’ve also got a number of games that are launching soon, while it’s probably going to be a while for our competitors’ engines to be putting out games. We’ll see.

The MMO Gamer: Do you think that reduced development costs will mean that developers will have more of an incentive to innovate, or will they just do the same old thing for less money?

Gavin Longhurst: The temptation is always there to do what you’re familiar with or what others see as a successful product – hence the proclivity of Tolkienesque fantasy in MMOs. CCP’s Eve is trying to do something different, but even then where are the magical realities? The Cthulhu Mythos, all the different alternate realities that exist in other forms of media? Second life is trying something like this in the way of polycreativity – the user rules scene, but the temptation is often to do the more traditional thing (or fund it at least). The industry is always in danger of doing what the FPS genre did for a while, cranking out the same game with one minor adjustment or new gameplay mode each iteration. But hopefully, with the advent of good middleware, they now have greater opportunity and time at their disposal not fall into that trap.

If you go to Asia, Korea is like a time travel device, its like 5 years ahead of what is done in the West.

A lot of people in the past were making very similar games. Eventually, developers – publishers – realized they had to make different games and use different art styles and billings models to differentiate their game from the rest. Chinese developers look to be making ‘exactly the same game’, generally focusing on their mythology and history – at least when viewed with Western eyes. On the surface they might look the same, but there is an increasing amount of variation in gameplay and execution.

We’re all in danger of not innovating enough, not doing something new. People are lauding the Wii for doing something different – it might not have horsepower of the other next-gen consoles, but it has this intriguing new mode of gameplay. Ideally middleware should give developers the a buffer, a ‘period of grace’ in terms of time, so that they can focus on the creativity and do something new. I’m not saying it’s impossible to make a great game by building it yourself – there are some examples of that approach that are incredibly successful out there. Middleware should give you additional room and flexibility to make something truly unique.

The MMO Gamer: Is BigWorld or Microforté working on any games internally which make use of the BigWorld engine?

Gavin Longhurst: No comment. Oh ok, BigWorld itself no, but Micro Forte is. That’s all I can say on that subject for now.

The MMO Gamer: What is your opinion on single-shard/server MMOGs, and what do you think the future will be like for them? Does BigWorld have any specific functionality designed especially for this type of game?

Gavin Longhurst: Good question. First of all “Horses for Courses”. Sometimes a game design will require a smaller population per shard. Who wants 25,000 people in IF? It’s crazy. In some ways, BigWorld creates a model that a lot of games haven’t used so far because of previous server tech limitations – that of one giant world. There are fundamental limitations on bandwidth and connection speed. With people all around the world, it’s difficult to ensure a regulated experience for everyone. Having said that, there’s other markets like virtual worlds, where latency is often less critical because it’s an environment – a world – rather than a time-dependent game. So latency can be less important, and virtual worlds attract people from different parts of the world, large communities, so it can make sense to have everyone on the same server. BigWorld can do either – Big shards, tiny shards, no shards, instances – we can work with any of these models. In terms of where it’s going, I hope we’re going to see more singular large worlds, more large contiguous space rather than shards, because A) it can be better to only have one community to manage and it can drive different kinds of gameplay and world design, B) initial populations may require it once it is done properly by someone, C) if you have the right tech and design there’s no reason why there should be such a limitation, and D) a larger proportion of the online community is on a better link than they used to be.

If you can get past the design problems of large populations, BigWorld has the solution to be able to implement this. Think of it…

Single large worlds can offer mass migrations, mass battles, mass economies… But game design is key to doing something like that right.

The MMO Gamer: With all the possible uses for interactive worlds, do you have any plans to license your engine to research facilities and other non-game related projects?

Gavin Longhurst: That is a possibility, but currently the limiting factor is providing support. While there are some interesting projects out there involving technology that is usually associated with online games, every license of our engine requires a support from our tech.

Support costs money, so we have no plans at the moment. While there are lots of interesting research areas, we’d have to consider each on a case by case basis.

The MMO Gamer: The capabilities of the BigWorld scripting and questing language, more specifically how robust it was when compared to Warcraft’s (hence my asking if you had played Warcraft).

Gavin Longhurst: There are numerous uses for scripting – how much of the architecture the developer wants to expose is up to a particular developer. BigWorld uses Python – a robust and proven language used in many areas of software development. Google and YouTube use Python – other scripting systems out there have their various fan groups – Lua and Python are probably the most commonly used in gaming – and probably in MMO development in general. You can take either of these further, as Unreal has with Unrealscript and Hero has with Heroscript, but there is always the trade-off between maintaining, specializing and extending your own scripting development vs. what the greater scripting community is doing. Python is a community based development, so is able to draw on a lot of functionality specific to many different areas of software development – something that is increasingly important in the Internet enabled Virtual World age as some games slide towards Web 2.0 and some Web 2.0 environments slide towards games. I think Eve Online – along with a number of other success stories in the MMO space, use Python.

The thing about either Lua or Python is they are very accessible to ‘non-programmers’ – when you are dealing with large amounts of gameplay, character interaction, large worlds and so on, You can bring non-programmers – like designers, world builders or artists into the development pipeline – both simple and complex functions can be created without the need for (expensive) programmers, who should for the most part be working on more hard-core, deep engine stuff.

Which is not to say the languages aren’t powerful when used in the correct way – you can build entire game systems with them. For the end player – how much access to the scripting engine they get is up to the requirements and design of the developer.