There are always tremendous complications when developing and launching an MMO. As APB recently handily demonstrated, it's an enormously time and money-intensive endeavor to launch a game on this scale -- and even when it does launch, there are numerous additional considerations arising from audience issues.
Of course, from that point forward, a company has to operate the game as a service, and continue to update it with live content. And then there are business model considerations...
In this in-depth interview, Ryan Seabury, creative director of LEGO Universe at developer NetDevil -- itself a division of well-funded MMO firm Gazillion -- discusses the development process of the expansive toy-based title, which was first announced in 2007.
He takes in everything from the cloud-based graphics-crunching technology required to get the game up and running, to why it's a subscription-based title, whether Luke Skywalker might make a cameo, and how the team has been focus-testing the title with the same group of kids for four years now.
There's a lot of building that happens in this game. Is it complicated to manage all this creation on servers, in instances, and what other players see?
RS: Yeah, there's a huge amount of technical magic that's happening in the background to make this feasible. We thought about rendering LEGO bricks in real time, but they're complicated, actually; people think they're just blocks, and they should be easy, but the level of detail with all the studs and the details on the other side... LEGO is uncompromising about how those need to look.
As a comparison, a two by eight LEGO plate brick, a very simple brick, is about twice the polygons of say, a World of Warcraft avatar. You can see as people build on their properties and stuff, there are hundreds if not thousands of bricks in a scene, so we built a ton of tech to optimize that on the client side and also there's a whole rendering farm technology, 3D surfaces cluster on the server side.
When you build a model, it gets uploaded to that, it starts optimizing it, doing all this material surfacing on it, removing hidden surfaces, doing all this magic on it to make it run on everybody else's computer, and that comes back down the wire to everybody else. That's happening basically in real time.
You can have your best friends in there, playing with the stuff as you're building it. As you're putting it together, the physics are dynamically generated and everything. It's really cool; it's the culmination of a lot of years of work to get to that point, but now it's all online.
Will cloud technology be used to make that work?
RS: Yeah, it's a cloud type of idea, right? We have an extensible network arrangement; we can throw more hardware at it if we need to. We offload the processing that needs to happen to make these render in real time to a server location, so you don't have to have 60 high end multi-core CPUs sitting there crunching on your models, but you get the benefit of that nonetheless.
Does it raise the minimum spec very much, versus World of Warcraft, for instance?
RS: I think we're shooting for a lower spec, to be honest. It's a challenge, because when you talk about user-generated content, we don't know what people are going to do with that. But it's a family-friendly game, we're going to have kids in there, and kids' computers are going to be another step below what gamers' PCs will be.
It's actually funny to test the min spec we are going for. We can't even find the parts we want anymore; we have to piecemeal it out on the internet, and it costs more to put together one of our low-end machines than a it does to make a super high-end Dell or something like that.
Speaking of kid-friendly, everyone's got their own methods for this, but how are you going to make sure there aren't just a lot of penises everywhere when you're making things?
RS: One thing we can say is when you build models you have your own property, and you can share that if you want to. If you share something publicly, it will be monitored by a human before it's seen by other people.
Thinking about what the practicality of that means, if we have millions of players in the game, that's a lot of content; there will be some lead time before you get your models up to share, but that's not really fun, right? Say if I'm an eight year-old, and I build what I think is a really cool looking dinosaur, I want to show my friend, I want to show him right now.
So we started thinking about the whole user generated content paradigm and said, "How can we get it so it's possible to do this in real time?" I mean, the majority of people aren't jerks that will use it to make bad stuff; most people want to play the game for legitimate fun, creative reasons. So how can we get them together so they can do that?
We came up with this idea of best friends, which I think is a really cool, innovative concept. It takes a lot of ideas from what's going on in social networking today and other forms of user generated content and combines them. The idea is that you can take some extra steps on your LEGO ID at LEGO.com to verify your identity in real life. It's kind of like stuff we've seen in the bank industry and other domains, where we can establish a trusted, verified identity with the consumer.
If we know you are who you say you are, then we can trust that you can make connections with other people who have gone through those steps. That allows parents to say, "I've got kids that play with the kids down the street all the time, and as a parent, I don't monitor their chat playing in the basement. I don't monitor the models you're building."
So we replicated that in the online environment, so if you establish who you say you are, we can become best friends through LEGO ID, and then we can say that our kids can play together, and that means they can play in each other's property where the models aren't moderated, and they can chat more openly than the default chat in the game allows. It creates a real time creative conversation with the players.
At the same time, we keep it safe, because if you think of YouTube as a user-generated content thing, you don't watch 99 percent of what's on it; you go for the 1 percent that's viral and maybe some videos your friends uploaded about their vacation or something. It's the same idea.
Most of the time you don't want to see the content that's out there, so let's put the emphasis on these social networks and get those people connected as well as we can, and still make it possible to share content we think is awesome with the rest of the world. That just takes a longer lead time because a moderator has to look at it before other people can see it.
So parents will have to do that for their kids, right?
RS: Yeah, but it's also good for adults, right? If we have a group of friends... But yeah, if you want to make use of the best friends feature. You can choose not to do it, and play the game as normal, but you'd have to wait through the moderations queues.
What's the thinking behind going subscription-based for this versus free-to-play?
RS: Well there are a couple thoughts on that. We didn't think of the business model up front, we thought of the vision we had, which was: what is it like to live life as a minifig in a LEGO universe? First and foremost, we wanted to make the game a pure game experience. That allowed us to craft the design that made the most sense and captured the LEGO spirit without worrying about how we were going to monetize it.
As we got into it and the game evolved, subscription just felt like the natural choice. We didn't want it to be a commercial experience where we are constantly pushing you toward microtransactions here and there.
The other thing we saw was that while micropayments were taking off in Asia and starting to get some traction in Western markets, when it comes to parents, they really don't like the nickel-and-dime effect; they'd rather pay once a month and have access to all the updates and upgrades and everything. That kind of steered us more toward subscription to begin with.
The other thing is that LEGO is an established brand that is all about quality; you might pay a slightly higher price for a LEGO toy, but that toy provides a much more high quality experience than most of the other toys you find in the market, right?
It's the same idea here; we want to have something new in LEGO Universe once a month for subscribers, and we want to make sure that the people that are playing are committed to having a fun, creative, online world.
A lot of what you see in the freemium model is you have this mass of anonymous people coming in -- and speaking of objectionable content coming into the game, those are the kinds of people that would really pour it in there. We want to make sure the game has a good, strong, core community first before we look at expanding those business models.
Maybe five years down the road we might have some trial free-to-play mode, we might add micropayment if it makes sense in some other content extension of the game, but for now, we feel that subscription works best for what parents wanted and what players are asking for.
Interesting. Because other folks who do the free-to-play model tend to cite parents and accessibility, but in the other direction, saying that parents don't want to have a large credit card charge, and kids don't have credit cards, so they can buy the prepaid cards. It's interesting to hear the other perspective.
RS: I'll be frank, there's a huge industry movement right now towards the free-to-play and micropayment model, and I think it makes sense in a lot of, or maybe the majority of, cases if the game is coming online.
But for this particular title, if this was the same exact game, but without the LEGO brand on it, and no one knew the brand, it would make a lot more sense to use a free-to-play model to start with, because that's how you attract a lot of attention to a game, because it has so much visibility. Since everybody knows and loves LEGO, we don't need to swarm it with the anonymous internet, right?
It's also nice because LEGO is a brand, not a specific property that you have to be beholden to in the ways you are with others. Of course you have to have the blocks be right and everything has to look correct, but you don't have things like, "This staff of ultimate smiting has to be stronger because in the canon…"
RS: Yeah exactly, until we get into the cross-IP stuff like LEGO Star Wars or things like that, and I'm sure we'll run into some of that. But you're right in terms of the fiction, the mechanics, and so on.
The other thing with those free models when you have a big brand, imagine if three million people tried out the game day one, and they're playing it for free, it's not free for us to operate. There are real costs involved with that, but I think it's a great way to start a new IP or a new kind of game where you're going to be smaller to start with.
It sounds like the art is a bit challenging on the technical side, but does it make it easier from an art direction standpoint when so much of what you can do has already been established for years?
RS: I actually think it was a very difficult job to get to where we are now -- happy with the visual direction. Since it's this blend of non-LEGO with LEGO, and things that imply LEGO, and how the minifigs move, everything is under a microscope when we look at it. We have to ask, "Does it meet the brand values with LEGO?"
The look of the toys is very established; they're manufactured, right? They've been doing that for 75 years, but there's really only a few other examples we can look at in the video game space, and most of those are the Traveller's Tales games, which have been very successful. But those draw from the IPs they are associated with, so we really carved a lot of new ground with LEGO Universe, and made some unique and new style choices.
It took us years to get to a point where we were all happy, where LEGO was happy, where NetDevil was happy, the kids were liking what they were seeing. But even so, the interface is still undergoing constant iteration all the time. It's kind of like the blocks; you think it might be easy, because it's such an iconic and established visual brand, but when you get down to the details of what it means in the context of an MMO, it was really quite a challenge.
How many rounds of kids did you have to go through, focus testing-wise?
RS: Oh, we've been through thousands of kids. It's interesting; we started with a group of 19 kids before we even had a line of code or a drawing on paper, brought them into our offices in Louisville, Colorado, and asked them, "Hey, what would you want from an online LEGO game? What do you think that would be? What would be cool to you?"
And the same group of kids, all of them -- they were about eight to 12 at the time, now are 12 to 16 -- are still coming back and testing it on a weekly basis.
So every week we have kids coming into the office, and of course the beta has tens of thousands of people playing, and that's been going on for over six to seven months now.
We've been gathering metrics and feedback on a weekly basis from that, so there are tens of thousands, if not hundreds of thousands, of kids who have played this game to give us feedback.
And you do not have PVP combat, right?
RS: Not day one. We have a great prototype we build almost a year ago that looks really promising. There's a lot of potential for the way you get specialty kits and all your abilities on the different factions for PVP scenarios, but what were trying to do is tie it to a piece of combat that makes the most sense.
We don't want to just slap it in there and say, "Hey! Here's PVP, but it doesn't make any sense!" We actually have some plans for some future play themes that are going to be integrated in that will make perfect sense, and that will be the introduction for that.
Do you envision your expansions being brand-oriented?
RS: A little bit of both. We wanted to establish a unique identity that was LEGO Universe by itself, because we could easily lose ourselves in things like LEGO Castle or any of their bigger, what they call "evergreen" themes that are always around.
Rather than do that, we came up with our own thing, the Nexus Force, the Maelstrom. We reference a lot of the stuff that's there, especially the legacy stuff, like you'll see in the Nexus Force world, you'll see classic pirate stuff going on in there, ninjas, and things like that. But none of them are 100 percent there.
As we go forward, we'll definitely be integrating very much one-to-one -- there will be entire worlds dedicated to certain play themes, but we'll also continue the Maelstrom story, the Nexus Force.
What I'm excited about is when we develop and get more of those play themes in, we'll be able to start doing mash-ups of them. For example, what happens if you take the Castle guys and put them with the Space guys? There's actually story and directed content built around that. It'll be very interesting, and I don't think any MMO has that opportunity, where you can mash things up and they will still make sense.
Are you going to be able to integrate those external licenses, like from the Traveller's Tales games?
RS: It's certainly on the table; we haven't really entered discussions with all the IP holders. That was another thing -- that would have further complicated what is already probably the most ambitious MMO I can conceive of.
We did specifically make a point to architect our fiction and our world structure so it could work. For example, you could have a LEGO Star Wars world or a LEGO Harry Potter world in LEGO Universe. We'll just have to take those discussions carefully one at a time with each of the IP holders and respect what they want to see as well.
It seems like specific instances, or dungeons, or scenarios would be the best plan for that.
RS: Yeah, and it could be small integration points; maybe you see some of the characters from the Rebel Alliance walking around in the Universe occasionally, doing little cameo appearances. There's actually a whole rock stage, we could have guest stars show up and do a little rock music thing one day or something like that.
Ah, LEGO Rock Band, the most pointless of all the LEGO titles!
RS: My kid loves it.
Difficulty-wise, all the console games are super easy. How are you going to deal with that here? Is keeping in the same difficulty curve a concern here or not?
RS: I think we want to keep it familiar to the players of those games. A lot of people played those games and enjoyed them; there's definitely an emphasis on accessibility over complexity and challenge. Not to say there isn't challenge; it will be there, but we want to make sure it's accessible to a wider audience.
A super hardcore MMO raider that used to do 40-man guild raids might not find anything quite that deep and complex that requires so much coordination here, but there are elements of that here.
I've played MMOs for a long time, and I know people that consider their MMO of the day as a second job. I find it ridiculous that someone could mentally conceive of playing a game as work; it should always be fun.
You should be able to hop in, have fun, play on your property for a while, build something, oh maybe you don't have enough bricks for the idea you have, so you'll play the game some more, get some leaderboard stuff going with some of the minigames with your friends, a little bit of fun, indirect competition going on, do a little co-op, go back to your property, build some stuff, go to your friends property, give them some models to help them build their property up -- that's more of the loop we're going for. Your own creativity is driving the depth you will get out of the game.
And you're not really going for the hardcore guildies market anyway.
RS: Yeah, we're trying to take a more Pixar approach right now. Pixar makes family movies, but you can enjoy it, right? They have good technical execution, they take their stories seriously, there's entertainment value there for all ages, or there might be stuff that goes over the kids' heads that adults get. We're trying to do the same thing.
You'll probably see some pop culture references in the game that kids will be oblivious to, but an adult might enjoy it. Some of the depth in the abilities might be too deep for a younger kid, but they'll just enjoy seeing the cool effects and stuff while the adults can use it more strategically and tactically in the gameplay. We're trying to take that approach, but if you're a super hardcore raider… will you get hooked on the property stuff? I hope so, but it's not there for everybody.
Well, not everything has to target the hardcore.
RS: We just want it to be a fun game so a lot of people can play it, and if you're a fan of LEGO, it's just super cool to be your own minifig. You can take that character through a bunch of adventures and get cooler and cooler looking, and hopefully someday -- we're talking with the manufacturing guys -- you'll be able to order your minifig in real life, and even the models you build in real life, as well.
Most of the other popular LEGO games have come to console. What do you think about LEGO Universe doing that?
RS: I would love for our game to be on consoles. The problem with MMOs on consoles is just a business thing with platform holders. They're still trying to figure out how to make third party work in an MMO context. There's no technical reason why we can't be on there. I would love to see it happen.
Square Enix seems to have figured it out. They're doing it for the second time, with Final Fantasy XIV.
RS: Yeah, and there's a lot of discussion that goes on, and they're starting to figure out what they can do, but a lot of it comes down to who controls the service and the servers. LEGO treats their consumers very close to them; they don't want to give up control to anyone else because it's a trusted relationship they have. They've built a lot of trust with their fans over the years, and the platform holders feel the same way about their consumers, so a lot of it has to do with figuring out what to do on the business side of that.