Leading The Design of APB

Lead designer EJ Moreland comes to Scottish developer Realtime Worlds with a background in MMOs, having worked on titles such as Ultima Online and EverQuest II. But that background isn't the whole story.

The concept behind much-anticipated PC (and later, Xbox 360) online game All Points Bulletin (APB) is more accretive in nature, taking in action games, console and PC innovations, and coming up with its own solutions to gameplay questions.

Here, he talks about the process, the dangers and opportunites in creating an contemporary urban online game outside the typical MMO territory, and working with Dave Jones, the Grand Theft Auto creator who has talked passionately to Gamasutra about the opportunities online gaming offers on multiple occasions.

The interview, conducted at this month's E3 Expo in Los Angeles, is essential reading for those who would consider breaking the rather staid boundaries of the MMO genre, and also provides a window into the design process for developers who have spent their time working on very disparate genres until this point -- and now are trying to marry them.

One of the most elaborate elements of APB, highlighted extensively in the game's behind-closed-doors E3 presentation, is its customization options. Amateur designers will be able to use a complex set of tools to customize clothes, cars, tattoos, and logo designs -- and then, in view of the game designers, be able to trade or sell those designs to the wider player audience.

It's the urban virtual Etsy, in other words, and has the potential to open up the game to entirely different audiences than the shooter-loving online gamers who you'd expect to pick it up.

You talked about supporting the APB marketplace with design -- enabling people to design clothing, cars, and stuff like that -- and we've seen that kind of thing in games like Forza Motorsport. Attracting and maintaining the sort of audience who wants that experience is an entirely different proposition than it is for the people who just want to shoot each other.

EJ: It's a bit schizophrenic. We have a different, diverse skill set on our design team. We have people that come from an online background, such as myself; we have people that come from a very heavy action game background, and we actually have a component of the team that we call our "creative team", that comes from that more social background.

That comes from being able to... take their inspiration from a wide variety of applications out there. Things like IMVU, and other types of really social apps. And they're the people that are driving how that's going to work, along with our business team. So we actually work with the guys that handle the business side and look for ways to really try to pull to those goals of being able to support the creator, versus the game.

From that perspective, it's also going to require not just having robust and intelligently designed systems, but also the people who would be attracted to them may not be paying attention to APB.

EJ: For us, APB's release is the beginning, not the end.

We needed something to really crystallize the base concept, and Dave's always been a fan of contemporary action. I mean, obviously, as the creator of GTA, this is kind of a natural step for him. So it started out being, "OK, let's make this really believable city with a cool action game." As we layered the customization, as more and more people got involved, we took a look at it and realized, "Hey, wait a minute... This is a whole separate product in and of itself."

But we really don't want to dedicate the resources to completely flesh it out both ways, to start with. So what we've done is we've said, "We want to use this to support this great action game -- this really different online type of game -- and then, once the game releases, based on what the players' feedback is, we're going to go in every direction we can." We're looking at much more horizontal expansion than reproducing the same content.

So APB is the action district; the conflict of the city is one part that we'll continue to support after release. But we're looking at things like racing districts, fashion shows, private housing -- you know, everything in between. It's just which way the player base wants us to take it, and which bases we want to attract after that.

So we realize that the game has -- it's not what I would call a "narrow" focus, but it has a very narrow initial potential. But we want to make sure people understand that there's much more to it. We really expect there to be a pretty good portion of our base, even from the initial part, that are just early adopters of just that expression, and customization, and that social aspect.

How do you ensure that you survive that long? Because when you look at it that way, MMOs launch... I mean, you guys have ambition, and I think, from what I've seen, you have a really promising game. I'm not criticizing the game, but promising games have come and gone.

EJ: Promising games come and go, and I've actually been involved with the release of three initial MMOs, as well as expansions. I've worked on Ultima Online; I've worked on EverQuest II; I've worked on Star Wars Galaxies; I've worked on several MMOs that didn't make it to the light of day. So, you know, we certainly have a broad experience, and we understand that.

To us, it's all about initial execution. It's why we're not throwing a bunch of extra features into the game. We want it to come out, really fulfill its purpose -- really feel like value for its money -- and for people to be really satisfied with it.

We know if we do that, we'll have all of the opportunity in the world to expand it past that. So it's really kind of this trade-off, and it's really difficult for us, because we're so excited about all the aspects, but we have to focus in on just an initial set of aspects.

I think a lot of games are -- I don't want to say "guilty of it", but they're not so good at picking what their limitations are. I think people get inspired and just go crazy...

EJ: I've certainly been guilty of doing that exact thing, in a variety of situations. It's something that, I think, as a developer, you learn more from your mistakes than you do successes.

So, as the lead designer, and as someone who's at the top of helping to shape the product, it's very important to me that we definitely explore those things, and look for the ones that are important enough to include. But at the end of the day, we need to bring that down to a very focused, very specific set of executed features.

The customization system is something we have to execute on well; the action game has to be done well. I mean, everyone thinks that you can release these types of online games, and, "Oh, it can be okay, because it's online, it's multiplayer, so it doesn't have to be as good as a single player experience." And we realize that there are trade-offs, but we want it to be the best it can be, and so it's all about execution at this point, for us.

It's why we just now started to talk about the game. We don't want to put it out there and let people see it until we feel like it's ready to be there.

There's also been a belief where people seem to think that they can release an online game and it can be shaky at first, then improve -- and that idea is over, I think.

EJ: Yes. It's absolutely over. My observation from releasing games is: what you release on day one sets your curve. Sets your curve, so that no matter what you do, no matter how much money you throw at it, you never really break out of that curve. So it's very, very important for us to get that right. We don't have the belief that we can ship something incomplete and get it out there.

Now, the reality of the situation is, you know: there's always the variables that occur as you get the game to ship -- there may be areas that we could do better at. But the fact of the matter is, we're very, very focused on the fact that the game has to be executed well. It has to be stable, it has to be enjoyable, and it has to be worth the money, day one. Period. End of story.

You're talking about "curve", and I find that an interesting concept. Because if you look at some games, they start at maybe 800,000, and then they shrink down to 300,000, right? But then you look at EVE Online, and it started really small, and grown to 300,000. And that's a big victory.

EJ: Absolutely.

How do you look at those sorts of issues?

EJ: So, EVE Online is a great game, because what they did is they came out, they didn't have a lot of press, and they had a really dedicated fan base. So they used that dedicated fan base to get away with what you can't necessarily get away with in a game that's heavily publicized. They stayed quiet; they stayed under the radar; they built the core features, they built the core player base, and then they just started attracting more and more players. It's a growth curve.

That's not the average curve. Most games -- that are fairly mainstream, that have a lot of money behind them -- tend to have to make a big splash. And then they release, and the curve starts high, but it immediately starts to decline. I've seen that multiple times -- have actually been a key factor into why it occurred, so I have sins to pay for it in the past.

It's a tough issue. One thing that's interesting is the genre, right? Obviously, the high fantasy MMO thing is kind-of... "tapped out" is not the right word, but there's super-stiff competition, let's put it like that.

EJ: Honestly, there's the 800-pound gorilla. I mean, fantasy RPGs have to contend with WoW. It's great, well executed, polished; it has great pedigree now. And I still play it. It's still something that I enjoy in my time. I'm a fantasy RPG guy from way back, and the reason I got into the industry was Ultima. That's why I ended up at Origin in the first place.

So it's definitely a market that I think still has potential, and I think that there are lots of games that can go in there and make a splash. But they can't do it by doing the same things as everyone else.

We certainly flirted with the idea of taking this type of franchise and creating a fantasy game, but if we do it, it's not going to be another MMORPG that has the same mechanics. We're going to find a way to make it distinct, just like we have here.

I think that branching out of that genre is sort of uncharted territory in terms of massive games, right? I see that you've done things with the design that are definitely completely removed. Look at City of Heroes, right? It's a great game, and it has obviously managed to find a really consistent audience, but in the end it's not that much different than skinning another MMO, right?

EJ: Right. It was very important for us to not to just skin the game. We definitely flirted with different ideas on how we wanted to approach it. But it always came back to the same thing. We looked at the tags that we place on these things -- we call them MMOs, we call them RPGs -- some games want to make their own acronyms. To us, we don't care. It's not about how we classify it, it's about what the game does.

So for us, all we always say is: it's an online action game that has persistence. That's the important part for us. So, we don't try to look and see what else is out there, and go, "Well, we should have this feature, or this type of play, because that's what everyone else is playing..."

That's why we don't have levels; that's why we don't have arbitrary statistics. We want to try to see what players [do] -- what's important to us is what what we measure from players. Their artistic ability, their action game skill, and eventually the other types of game types that we introduce with this.

How do you maintain player interest without the carrot-on-the-stick of levels and similar systems? I mean, they maintain them in many games because they are a proven treadmill.

EJ: They're absolutely great. The whole DIKU-style of RPGs is genius in what it does, but there's only a certain group of people that really enjoy it. Now, that's not to say that they're [just the] 10 million people playing WoW, but it's finding what works for your game; finding out how your game works.

For us, we do have functional progression; we do have cosmetic progression; we do have things for players to achieve. It's always about putting things out there, that work for your game, that players want to continue to achieve.

Some games, like Counter-Strike, have no progression, but they still have achievement. You know, people created leagues, they created leader boards -- it really goes to what your game is. Our game is an action game. For us, most of it's going to be about players comparing each other, and seeing how we can weave achievement into that. So, as you achieve, you get things that make you more famous.

Again, the game is really about that celebrity. But there is also, like I said, functional progression. You don't get everything up front, you have to earn the types of cars, the weapons; all that stuff is progression in the game.

This game has a very sophisticated design, from everything I've seen. How do you arrive at these designs that have not been seen before in this type of combination, and how do you know it's going to work?

EJ: We don't.

Right.

EJ: That's part of it. I mean, I have to say that the person who should get credit for really pushing those boundaries is Dave Jones. I mean, he's never been someone who adheres to what's already out there. He wanted to make a game, he had vision for it, and he brought people in like me, to go, "Okay, this is the broad vision -- how do we make it practical?"

And we went through lots of trial and error. I mean, honestly, we've gone through the whole gamut of functional progression. We've looked at how the action game could work on that many multiplayer.

We've just continued to iterate and arrive at what we thought was the best way to do it. And we're continuing to learn. Honestly, when we get it out there in beta, we're going to see what works and what doesn't; what we can change, we're going to change.

In terms of hard design doc versus prototyping, what kind of balance did you have?

EJ: I arrived in 2007, and the game had been going since about 2005, so they had some definite design ideas and some documentation there. And what we did was we took a look at that, took a look at where the game was, and started trying to play the game. And as we learned what we couldn't play and what we could play, we started adjusting the design.

Since then, we've continued to create documentation, and continued to maintain it, but it really is about: we apply something in the game, we see it, we feel it, we touch it, we see how it works, and then we can go, "Okay. This works. This doesn't." Dave's heavily involved in that. The whole development team is heavily involved in that.

RTW has the tag on it of being a design-driven company; what that means, necessarily, isn't that the design team drives the company, it's that everyone is involved in the design. It's my job to maintain and shepherd that. But, you know, we take the advice and consultation of everyone else involved -- whether it be an artist or a programmer, or an audio guy, or a community person. All of those people have valid concerns. We try to address that.

And it's certainly a unique development environment. I've worked at seven other companies, and this is the most unique environment I've worked in.

Another thing that's been a concern, I think -- we were talking about how launches can't be like they used to be, and I think betas have shifted, too. They've become way more of a marketing tool than they were, and I think that the audience, back in the day, had this understanding that the game was going to be broken in beta, and now the audiences don't have that. 

EJ: When you get feedback in beta, the whole point of the beta is to take that feedback to heart, even if that means making some sacrifices, and looking at your release timings... You have to push back and say, "Hey, if these people say it's not ready, it's not ready. We have to fix it."

So we're already in what we call our F&F [friends and family] phase, where we do have some public involved, but it's very small and private -- it's mostly friends of the company, or people that we trust implicitly with the game.

And we're taking their feedback as well. We have forums set up, and we have community guys already working on that. So, to us, beta is where we discover what we did right and what we did wrong, not what we thought out. Closed beta.

Open beta, we understand is pretty much a marketing exercise. It's a stress test and a marketing exercise. But the early parts of closed beta, we're going to continue to find people we can trust to look at the game objectively, rather than just expect it to be ready to play.

That's a good point, about closed versus open, because ultimately you do have to test these ideas. There's no real way to do it without having a beta.

EJ: No. With big online games, you can't have the development team play it and expect to have everything right. What we use the development team for is a real good litmus test: a lot of the guys are very specific play types. We don't have a lot of MMO players on the development team; in fact most of the ones [who do] are people who come from MMO backgrounds, and a few other guys. So we rely on those guys to tell us what's wrong with the kind-of persistence and progression -- what feels right from that point of view.

But a lot of our people play action games; they play console games; they play very mainstream stuff. So we use them as a good test for what the usability is like; what the user experience is like. We use that as our initial test, and we make our assumptions off that, but as we get it out there, and as beta players start to play, we're going to listen to what they have to say.

I mean, Dave is committed to this really being a player-driven game through and through. Not just player-driven in the mechanics that players actually use to play the game, but player-driven in how we take the game from post-release as well as in beta.

The crime theme is a really great target. It's really compelling in a lot of media -- there have been some really successful games that, obviously, Dave's been involved with. However, this isn't a traditional MMO. That audience and the people who are familiar and plugged-in to PC gaming may not be as familiar nor as plugged in to this genre. What do you think?

EJ: I think what we're looking at is, we're looking at establishing this genre for ourselves, and it is this online genre that shares similarities with other ones, but it is different. We are looking for an audience that enjoys the game.

That may be console guys who do play GTA; that may be guys who play WoW; that may be guys who play My Little Pony Online. The honest truth is: we have some pretty clear business targets about what we want to reach, but the audience is still kind-of an unknown country. And we think we have a compelling game -- the reaction we get is compelling. So we believe that we're going to be able to attract a real core audience that's going to enjoy it, but we also, certainly, recognize that this is different.

I mean, one of the things that concerns us the most is, between now and release, the message we give about what the game is has to really educate people on what the game is, because it's different. I mean, it is truly different.

And I'm not just saying that to put the cliché out there; as someone who has worked on a variety of these types of games, it was definitely hard wrapping my head around it. It's been one of the best challenges I've ever had. So I hope it works out.

One thing that could be a concern is system requirements, on a game like this. Because I think it could draw a totally different audience, and those are not necessarily the people who are buying giant six thousand dollar PCs. But it looks really, really good...!

EJ: Right. So the good news is, it is -- at our core, we are an Unreal 3 game. So we have the same system specs as most Unreal games. But we spent a lot of time reengineering it ourselves, not because we were dissatisfied, but because we were doing some crazy stuff with the streaming and the customization.

The truth of the matter is, we've spent a lot of time personalizing. That's what it's there for. I mean, you take an engine off the shelf, you don't expect it to work right off the bat. And the games that do do that are pretty mediocre, to be honest. So we spent a lot of time making our own, and making sure it works for us. And part of that is optimization.

I mean, it's definitely going to have bigger system requirements than you typical turn-based fantasy MMO; we understand that -- and [it's] because of the customization. But they're not unreasonable system requirements, in my opinion. You don't have to have a brand new machine that has an I7 and dual-core video card. I mean, I've seen it run on a GT 7800 Nvidia card; I've seen it run on really low spec dual-cores, and and high-spec old P4s.

I guess that's an engineering question, too, in terms of how the game deals with those system requirements. Some games just choke on low systems; some games scale more intelligently.

EJ: So, we certainly are still working on it. It does choke at times, but that's just part of the ongoing development. What we've actually done is, we've reduced the fidelity of the look as we good.

Actually, I would say we don't do standard LOD stuff, we do some pretty clever stuff. We care about what's relevant to you rather than what's nearby you. So people in your group and people you're opposed to, we try to keep that detail much higher, even if there are many more people in there.

The matchmaking tech in the game is really interesting to me. You talked about the hierarchy of the way you find groups to fight with each other and put them together, and that it's asymmetric, in terms of one really skilled person could be taking out a group of four. How do you make that work?

EJ: Well, so, I mean, that's part of the magic, to be honest. We've spent a lot of time on the logic behind that. We've got a dedicated gameplay tech team that pretty much focuses on our mission system and our matchmaking, and we have a gameplay design team that basically is joined at the hip with them. Their whole major charge is making this work. And a lot of it is just really simple math.

One of the things that we looked at was: you look at how you categorize -- how you get a metric off someone -- and the first thing you think of is, "Oh, well, it's the equipment that they have, and it's this, and that..."

But you know, honestly? It's the last 30 or so missions, and success. That window, and that metric has been the most accurate thing, and the most predictable thing we've had. So, it's based all around that..

How about ping and latency issues with such complex multiplayer action?

The one little secret we have is, we have a pixel-precision shooter, and we're latency-tolerant. Now I'm not going to talk about how we got that, but it's pretty unique, and we're pretty proud of it. So, even ping isn't that big a deal -- for the combat.

For driving, it tends to be a little tougher, since the driving is truly digital, rather than analogue. Playing driving games on a console, you've got the thumb sticks, but with the WASD, you hit A and you turn left. So the driving is not quite so latency tolerant, but the combat is extremely latency-tolerant. We've had people who've played in Korea and been competitive in combat.

So it is absolutely tolerant... up to a degree. Now, obviously, after a certain point -- you know, 340 milliseconds, it starts to degrade pretty drastically, but under that, it's pretty competitive and equal for everyone, whether you have an 80 millisecond ping, a 50 millisecond ping, or a 140 millisecond ping. Same thing with framerate. We tend to framerate lock; we tend to be able to make sure that your technical requirements don't invade in that too much either.