May 13, 2013

The Bureau: XCOM Declassified preview — Why this isn’t the XCOM I wanted

XCOM: Enemy Unknown

In its third and final form, the combat system of The Bureau: XCOM Declassified plays an awful lot like Mass Effect. It’s a third-person shooter where you point a pair of squadmates around levels that are dense with cover, and a handful of its character abilities are pulled straight from Commander Shepard’s playbook. There’s a Lift power that levitates enemies. There’s a radial wave attack that bucks aliens out of cover. There are powers that sap an enemy’s protective armor or shield, and a Squad Heal ability.

When I played The Bureau: XCOM Declassified a couple weeks ago, I asked 2K’s producers on the project about the thinking behind the combat. The answer I got was scripted and disappointing.

What were your inspirations for the combat system?

Nico Bihary, Senior Producer: Well, a lot of the inspiration comes from the XCOM franchise. There’s a lot of... did you play Enemy Unknown?

Yep. I reviewed it.

Bihary: So you probably saw a lot of the similarities through the combat, you know—as I’m moving my guy into cover, it shows me the half-shield or full shield, or if I’m moving a guy into bad cover it’ll tell me where he’s gonna get hit from. And there’s also weapon types: laser, plasma, and human. And there’s kind of a progressive rank on how advanced the weaponry gets and how good you are. So those ties are definitely very obvious as far as the combat system goes.

Andrew Dutra, Associate Producer: The core XCOM experience was a big influence on how things were formatted and how things were displayed.

Wouldn't you say it’s fair to say that there are a lot of parallels to Mass Effect, in terms of the combat style?

Dutra: I mean, yeah, we've seen people comment about that, and it’s great, Mass Effect is a great game. Our game’s a great game.

What would you identify as differences between Mass Effect’s combat style and your combat style?

Dutra: So with our game, you have more control over where your agents go, you have more control over where they’re placed on the battlefield, versus “you just do an ability here, you just do an ability there.” So there’s more strategy involved behind that. And also the whole consequence of, “My guy’s gonna be lost, I can’t use him anymore” certainly ratchets up the difficulty there.

Bihary: We certainly push the consequence a lot, which we don’t think anyone has really done in this third-person shooter scenario.

Dutra: Which, again, is an XCOM core principle.

It’s easy to call 2K’s XCOM shooter a copycat. But I don’t consider that rhetoric a useful exercise—innovation can happen through iteration, and you could make the argument that it’s clever of 2K to rely on players’ familiarity with a widely-played series.

What it does say to me is that there was insecurity within 2K about how to make XCOM into a shooter and have it sell. Remember, The Bureau is the third full iteration of the game—when I first saw it at E3 2010 (a year and a half before Firaxis’ XCOM: Enemy Unknown was announced), it was a first-person game dripping with atmosphere, mystery, and its own personality. I wrote then:

William Carter is responding to a distress call in a small neighborhood, a cul-de-sac smorgasbord of Leave It To Beaver homes. He exits his vehicle in the street with two fellow agents--AI-controlled comrades in ties and hats, each wielding a pump-action shotgun. It’s perfectly silent. A children’s bicycle lies toppled in the road.

He makes his way through a backyard--an idling lawnmower in half-cut grass is the first evidence of something amiss. A few feet away on the patio, there’s a dead man smothered in blue-black oil, a grease that paints a trail into the house. “I’ve seen this before,” says a nameless agent, kneeling to inspect the body. “But it doesn’t get any easier.” Carter snaps a photo of the body with his camera and moves inside.

XCOM’s E3 2010 showing is still stuck in my brain, and it’s rare for a game demo to stay with you for that long. What I remember loving about this vision of the game was its pace: my write-up above shows the trickle of observations made before the player gets into combat. Scripted as that demo was, I loved the way it slowly toured you through the aftermath of an alien attack. I loved that there was some mild detective work, that it was HUDless, and how naturally clues felt set into the environment. The Bureau, by contrast, is driven by its combat and slathered with information layers: activating Battle Focus (the time-slowing planning mode in which you give commands to squadmates) fills the screen with translucent interface. The life bars of Sectoids (and other aliens) pop into the center of the screen as you shoot them (at roughly the same size and spot as they do in Mass Effect, coincidentally).

And what was most disappointing about The Bureau was how much the things you shoot—Sectoids, Silicoids, and other aliens—felt like interchangeable, generic game enemies. Sectoids lacked the creepy, bobbling gait that they had in Enemy Unknown. A Muton boss I encountered at the end was a hulking damage sponge, but he could’ve easily just been a mech or a giant ogre--there was nothing in his animations, attacks, or behavior that conveyed he was from another planet. The Muton’s one distinguishing aspect was his modular armor, which you had to crack open to kill him. But I was annoyed when The Bureau revealed this immediately by slapping a damage display of the Muton’s armor on the screen rather than letting me earn an understanding by fighting, trial, and error.

Aliens were alien in the XCOM 2010 demo—the blobs (Silicoids) animated unpredictably, lunging and slithering along walls and ceilings like liquid cats. At first brush, you didn’t know how to deal with them, or if they could be killed at all. While I was playing The Bureau, Silicoids were described to me as “alien attack dogs” that nip at your heels.

And at the end of the 2010 demo, your encounter with an unnamed alien power weapon was mystifying and intimidating. I wrote then:

Everything turns red. Something is here. Something big. Carter runs toward the street–there’s a black wind of distortion hanging in the air. A field that warps light around it. The mass manifests a strange rectangle-obelisk in the sky: it looks like a giant, textureless Jenga piece. Without warning, it rearranges itself into the shape of a ring and spits a beam of white-hot energy into the street, disintegrating a car. Then, it does the same to the remaining agent.

I’m less lamenting that The Bureau is borrowing Mass Effect’s combat than I am disappointed that The Bureau seems to have abandoned the sense of mystery it debuted with. Neither of the enemies shown in the 2010 demo were humanoid, but more importantly both of them were introduced without any explanation about how to survive or kill them. That aspect of learning in the field (through painful death) is essential to XCOM—the first time you stumble upon a Cyberdisc in Enemy Unknown, you don’t know how far it can move, whether it’s less deadly or vulnerable at close range, or what its relationship is with the Sentry drone it spawns with.

I do feel a little uncomfortable comparing two slices of gameplay to one another: both demonstrations of XCOM (and the E3 2011 demo, easily the worst of all of them) were about half an hour. I can’t know if the XCOM I saw in 2010 would’ve turned out to be the game I wanted it to be. And it’s possible that there’s a generous intro sequence that slowly introduces the alien threat in The Bureau.

I hope there is. What I saw of The Bureau characterized it as a game that’s hanging its hat on cover-based gunplay and its third-person strategic system, and I’m crossing my fingers that it’s driven by more than slightly derivative shooting and slightly cumbersome combat mechanics. 2K’s 2010 debut proved that XCOM set in the mid-century can be a captivating, original setting, and that the gulf between mankind’s technology of the time and that of the aliens creates great opportunities for inscrutable, strange enemies. I hope The Bureau makes use of that.

Apr 28, 2013

Julian Gollop interview: on X-Coms old and new, the Ghost Recon strategy game that never was, AI, auteurs and “Fork My Fruit”

XCOM: Enemy Unknown

Julian Gollop is a 27+ year veteran of the industry. He can list classics like Chaos, Laser Squad and, of course, X-Com, on his long career resume. As Firaxis successfully reboot X-Com for modern audiences with Enemy Uknown, Gollop has donned indie threads to pursue a current remake of his multiplayer wizard-duelling game, fittingly named Chaos Returns. I caught up with him at GDC for an affable chat about his work on the original X-Com, progress on the new Chaos game, and his thoughts on how the great machine of modern development compares to the tiny teams in operation during the turn-based-strategy boom.

The man himself.

The interview's a big 'un. Here's what you'll find on each page if you fancy skipping to a part that interests you.

Page 2: How auter-led development compares to Gollop's position in the original X-com team, and details of his new indie project, Chaos Reborn.
Page 3: Expanding Chess, "Fork My Fruit," the aborted "XCOM meets Ghost Recon" pitch and Gollop's thoughts on the modern version of XCOM.
Page 4: Gollop's love of boardgames, the story behind Terror from the Deep and XCOM: Apocalypse and Gollop's favourite recent games.

PC Gamer: In the original XCOM, the way the AI moved towards you, they would make use of cover and they weren't completely suicidal.
Julian: No, they weren't. I can't remember exactly how we did the AI in the original XCOM, but a lot of the time I tried to avoid moving into the direct line of fire of your guys. They tried to find cover if they could, most of the time.
PC Gamer: The new XCOM has that for the Sectoids and the Floaters and the Thin Men, which were the Men in Black which you couldn't put in because Microprose were making another Men in Black game. They had quite a complicated AI system for them, but anything which was melee orientated just ran straight towards the nearest target. You're talking about AI here (GDC) as well.
Julian: I went out of interest because back in 1995 when I was coming to GDC, guys like Neil Kirby, I was involved in their round tables. Every year we used to go to it. There weren't that many AI programmers around at that time. A lot of them were actually involved in RTS games because that was the big thing that killed off turn-based games.
PC Gamer: I remember it well. After Dune II...
Julian: After Dune II... I mean, XCOM was really just at the end of the period where you had this turn-based strategy game as being a mainstream game.
PC Gamer: They're coming back now, but they're not coming back as mainstream games.
Julian: Not as mainstream games, no, because in those days, you had XCOM, Master of Magic and Master of Orion were, for me, phenomenal games. Colonisation, of course, came a bit later.
PC Gamer: I guess there were The Heroes of Might and Magic kind of games.
Julian: Yeah, Heroes of Might and Magic developed a trend, but they weren't in quite the same tradition of this grand strategy game which had big random elements in the generation of world, lots of AI and stuff. Heroes of Might and Magic is a bit of an exception. In those days, I believed firmly that the future of computer games was all about AI. That in twenty years time we'd be interacting with NPC characters in computer games that actually had real intelligence and could respond to you in really intelligent ways. Boy, I was wrong. So wrong!
PC Gamer: Do you think it didn't happen because we never built on anything we built? As in, every time people build AI they build it anew, there aren’t AI libraries as far as I know.
Julian: I think part of the problem is a lot of effort was put more into graphics rather than anything else.

Gollop ponders the "paper thin illusion" of visually advanced modern games like Assassin's Creed 3.
PC Gamer: It can be seen to raise review scores, sadly.
Julian: Because it's the thing that immediately impresses people. As soon as you start interacting with a world of pretty graphics then you realise that actually it's not so interesting. It may be pretty but it's not really that interactive. It's always bugged me about the way computer games developed over the years. Even if you take Assassin's Creed, a phenomenally complex game with all these NPCs wandering around, it is nothing but an elaborate paper thin illusion, to be honest.
PC Gamer: It's a paper thin illusion which is very clear about saying, "This is an illusion". Inside the game, the framing device that they use to make it a series rather than a random collection of games by the same name, there's a person playing a game within a simulation.
Julian: It is, but then again - yeah, that's true. (laughs)
PC Gamer: It feels to me like a huge joke, that they've done that. "How can we get away with making a game with paper thin mechanics, which are quite obviously mechanical? Oh, we'll a simulation inside a game".
Julian: You could say that, yeah. I mean, computer games didn't develop really in that direction, and I guess what people enjoy and what they like at the psychological level is more to do with having their own ego massaged in certain ways through these very simple reward cycles.
PC Gamer: It always struck me as interesting in the Turing test stuff, that it's not that AI ever passes the Turing test but people fail the Turing test. When you have the awards in England, it's always somebody pretending to be a robot which causes an AI to pass the Turing test. Not an AI actually being convincing in any way. And there's something about it being easier to fake intelligence than it is to even get anywhere near trying to generate it really.
Julian: Yeah, obviously when I was programming XCOM stuff we were faking intelligent. We had some very simple tricks to fake it. I talked a bit about the randomness element in XCOM and how we put it in the AI. But in actual fact, being unpredictable is a way of intelligently countering someone who's predictable. If you play poker, for example.

A mote of randomisation made X-Com's enemies more fiendish.
PC Gamer: I knew you were going to make that reference. My friends hate it when I play poker because I'm random. I don't really understand what I'm doing.
Julian: The good poker players say, depending on your opponent of course, they'll say sometimes you need to mix up your game. Not necessarily that you're completely random but you're doing something which they're not predicting. You're maybe just changing the way you'revalue something and it throws them because (they) can no longer predict what you're doing. In the original XCOM, as I said in the talk, we always tried to make sure that the aliens did not do things on a purely binary yes/no thing, to always have a little bit of randomness in there. 10% of the time they'll do something really stupid perhaps, but most of the time, within some kind of reasonable constraints, what they do is reasonable even though it may have some random element to it.
PC Gamer: That randomness actually sometimes gave them a good chance of survivability as it meant you might have seen something disappear round a corner but you can never walk round the corner because you can never quite predict what will happen. There is a thing it should do rationally, but it might not be doing it.
It's interesting, the other person who was talking about the unpredictability thing was Gary Kasparov, when he writes about playing chess against a computer. Obviously, that whole peak of computing intelligence with rule sets, of chess, where the chess computers memorise the rule sets that every single Grand Master had learnt, Kasparov writes about it and says that the way he found of getting around it was having to always try and work out a way outside what somebody had done before. Going outside that rule set.

The team that built X-com was minute compared to modern blockbusters.
PC Gamer: The other thing that struck me about your Making of XCOM talk was the humility of how you describe how the game was designed. You describe it as you'd done the battle bit and then all of these other bits were suggested by Microprose. It's unusual in this industry, especially with the superstar developers that are around at the moment.
Julian: Yes, it is unusual, but then again if you work with a lot of creative people over the years like I have, you realise actually that you depend a lot on them. I've worked as a producer where I've had to try and build teams of people, get them to work together and you really have to make sure people are leaving their egos in their pockets or parking them at the door because you can get into big problems. What I did for my post mortem, actually, was I tried to contact all these people over the last few weeks to try and figure out what their recollections were of particularly the origins of the game. It was very interesting. There were some conflicts in what people remembered, for sure, and there were some things that I learned because I had no idea about the Spectrum Holobyte cancellation story.
PC Gamer: You didn’t realise it had been cancelled?
I did have some inkling from the QA team many, many years ago, someone some years ago saying that there was a threat to cancel it but I never realised that Spectrum Holobyte actually did make that decision, to cancel it and that the Microprose UK guy said, "Hmmm, nonono".
So I got this information when I spoke to people a couple of weeks ago, I guess. So I wanted to try and do an honest record of the development. Particularly guys who made a contribution which was never really recognised. Steve Hand, for example. because he wasn't in the credits or anything. Also, for the guys that did work on the project all those years ago: John Broomhall, the composer; John Reitze, the graphic designer - these guys really contributed something fairly unique and memorable to the project, without a doubt. Really, without my input to a certain extent. They were just doing this based on their own creativity.
PC Gamer: It's interesting that you had such a relaxed approach to the development. It was like, 'We have these people making music. We trust them, because Microprose UK have told us that they're going to be good at it.' You didn't select these people yourselves?
Julian: No, not at all.
PC Gamer: It was almost like it was, "We're doing our bit and they're going to do their bit and it's all going to work together in the end, so that's OK!" Nowadays you get people like David Cage or Ken Levine, the auteur theory, who have to go over every single detail in the game.
Julian: I think stuff today is so overdesigned, it's unbelievable. There are people obsessing about tiny details about stuff. Especially when you have marketing people involved about how your main character in a game's presented suddenly becomes a huge PR and marketing issue
PC Gamer: The whole thing with Booker holding a big gun on the cover of Bioshock Infinite. It's like crossword magazines in the UK, always having a very attractive blonde girl biting a pen. It sells more copies, amazingly.
Julian: What a shame.

The original Chaos was released on the Sinclair ZX Spectrum.
PC Gamer: What are you doing at the moment? I know you're working in Bulgaria.
Julian: Yeah, I'm working in Bulgaria. I am establishing my own independent games development studio. I'm working on a turn based strategy game. It's a sequel/remake of a game I made back in 1995 on the ZX Spectrum called Chaos which was originally published by Games Workshop. This was this just fantastic multiplayer turn based game where you're a wizard, you summon creatures, You're just looking at a black screen as an arena with your wizard but it gets filled up with creatures and magic fire and gooey blobs and stuff. It worked brilliantly as a multiplayer game so I want to update it with proper internet multiplayer connectivity.
PC Gamer: I recall looking at your blog with the concepts on there.
Julian: We've got concept art going on now. Although the concept art is obviously a lot more sophisticated than on a 48k Spectrum, we wanted to have some kind of feel or some kind of reminiscence of how the original game looked with it's completely monochromatic but brightly coloured, primary colour sprites and this black background. We're not going to have a black background but we're certainly going to have a dark background, for sure, and a bit more of an abstract, stylish graphics which is more illustrative than purely real rendering stuff.
We're just working on that aspect at the moment, but the actual core gameplay, I made a decision that I'm going to retain the actual core gameplay from the original game. We will elaborate a bit on the spells, for sure, there'll be more spells. I think the core gameplay was actually very simple and going back to this whole poker mechanic thing, it had this great bluffing mechanic in there where you could summon a creature as an illusion.
There's a lot of probability in the game, every spell has a certain probability to be cast, so the more powerful spells tend to be the most difficult ones to cast. You roll to make a creature like a gold dragon and it was something like 20% I think it was, for casting it. If you cast it as an illusion you would automatically get it. There was no possibility that you'd fail, which was cool because every player has a disbelief spell. If somebody summons a gold dragon, probably most players would think, "Well no. Now, that's probably an illusion. I'll try and disbelieve it". But if you disbelieve it and you fail, you've wasted your opportunity to cast a spell and you could be in trouble.
So, this little simple mechanic creates little bluffing strategies between players. Because of the high element of randomness and probability in the game it kind of makes the gameplay less predictable and controllable for each player which in some ways is more fun because there's always a possibility to win the game, however small. The gold dragon could come out to your wizard and attack you, you might survive. Not very likely. You might then attack the gold dragon and you might kill it. Not very likely, but you could, for example. The odds are in there. Trying to analyse why it works is quite interesting but I know for sure it does work well as a game and I want to bring it back.

For Chaos Returns, Gollop wants to retain something of the stark visuals of the original.
PC Gamer: There's the iOS and iPad version of the Settlers of Catan. Obviously Settlers is a dice based system so it’s random. They have a system in it where you can also choose a stacking system where the 36 possible results are treated as cards so you have to get through all the results before you move on. It kind of balances against pure randomness, with that.
Julian: So you know there is going to be at least one of each result there. It makes it a bit less arbitrary. Yeah, you could be screwed in Settlers of Catan, I've played it many times. I guess they’re trying to make it a little more controlled, but still retain some of the randomness. I'm just not worried about it. Basically, if you lose, you lose. If you win, you win. If you're a good player, you will tend to win and if you're a bad player you will tend to lose but it’s not automatic.
But I'm adding a whole meta-game to the game as well, this is another aspect. A single player meta-game. But you might have some multiplayer effects as well.
PC Gamer: Is this the second level type thing the same as you had in XCOM?
Julian: It's going to be a little bit simpler than XCOM, actually. The idea is that you're asked as a player to name the world that you wish to explore. This is used as a random number seed generator for the environment. So you have a world which is full of different regions, different types of terrain, and you're exploring. Your objective is basically to kill the Chaos King in the region but your secondary objective is to find stuff because there's lots of artefacts in the game which are going to be useful to you in multiplayer battles or single player battles, so there's a slight RPG element to it as well. So you've created this world and you're exploring it. You go from region to region, you'll fight any enemies in each region who have their own sets of spells or own personality. There's different terrain types in each region. There's special places within realms, places where you can learn your spells, places where you can teleport, places where you can move things around the world. It's a place that people can explore, still bearing in mind they have this requirement, this strategy, to find and locate the boss and kill him. Very simple.
PC Gamer: But it's all procedural?
Julian: Well, it's procedurally generated in the sense that yeah, you're still within an environment that consists of distinct regions, but they're randomly put together. A procedurally generated adventure, if you want to call it that way.

Chaos Reborn - naked giants confirmed.
PC Gamer: It's nice to see you're still genre-busting.
Julian: Well, yeah. I really like games that generate stuff for you. I complained about stuff being over designed. My obsession was always with scenario generators, if you want to call them that, where things are generated for the player to explore and it may be something nobody else has ever played because it's pseudo-randomly generated.
PC Gamer: Which saves you programming time to some degree.
Julian: It saves level design, that's for sure. Yeah. It does allow you to create something vast and complex to explore with less effort, sure. Because you're not designing every single possible experience the player could have in the game at all. Yeah, it's one of my little obsessions I guess, and I've still to see it done well in games. Rogue-like games have randomly generated environments and that's part of their attraction, because apart from that they're very simple games.
PC Gamer: Well, that and permadeath.
Julian: It's true. So I still think this style of game has an attraction for a lot of people. We're going to keep it nice and accessible and simple like back in the Spectrum days, but obviously there's much nicer updated presentation of course.
PC Gamer: And the ability to patch.
Julian: Yes, and add extra content as you're going, of course, and proper multiplayer online. The thing about this generating from a name you type is that you can say to a mate of yours, "Look, try this particular word because in this particular region you will find a tower of mist where you can get the Cloak of Fortitude". You'll be able to exchange stuff with other players and discuss what you can get where in a particular realm. Of course, there's millions of possibilities of things that can be generated this way.
PC Gamer: Have you worked out how many possibilities?
Julian: More than millions. It depends entirely on the limits of the random number seed, I guess, but it would be a lot.
PC Gamer: Sounds wonderful. Have you got an idea yet when you want to release?
Julian: Next year. I can't be more specific than that, really. I'm trying to build the team and get resources for the game as well, so this is all part of the process. When you're an indie developer you don't necessarily have to start with a fixed budget and a fixed schedule and fixed resources.

Gollop wanted to make an XCom-esque Ghost Recon game at one stage.
PC Gamer: Why Bulgaria? Is that because there are established programmers out there?
Julian: No, it's where I live. I've lived there since 2005. Because my wife is Bulgarian and I've got two children as well, two years old. I worked for Ubisoft from November 2006 to March last year, just over 5 years.
PC Gamer: What were you working on?
Julian: Ok, so when I first started at Ubisoft Bulgaria - it's a small studio, 13 people - I was employed as a game designer. The first project they wanted me to work on was Chess Master which I was a bit surprised at. So that's 2006, 2007 they were working on Chess Master 11, I think, for PC. I'm not sure why they wanted me to work on this because I thought the game of Chess had already been designed. Actually, what they wanted was a DS version of Chess Master. We added some mini games based on Chess which I designed.
PC Gamer: So you redesigned Chess?
Julian: I actually designed some original games using some Chess-like rules. There was one, my particular favourite, called Fork My Fruit where the Chess board had bits of fruit on it and you had to, using the forking principle in Chess, you could fork fruit. You got the fruit from the board.
I did Chess Master, then I worked on some projects that were cancelled. Then I worked on Ghost Recon: Shadow Wars, which was a launch title for 3DS. This was a project that I actually pitched. I just wanted to do a decent turn based strategy game again. I wanted to do something similar to XCOM still and I thought, well, you know, what's Ubisoft got that could be used here? Looking at the franchise, OK - Ghost Recon, possibly. So I pitched it initially as XCOM meets Ghost Recon. One of the guys at Ubisoft central office in Paris said yeah, OK. He OKed the project and we did a working demo and design. I think we spent maybe 3 months on this. It did have aspects of XCOM. There was supposed to be a world view you know, generated battles and maps, different bosses in different parts of the world that you had to tackle. The tactical game that we had for this design was very much like the new XCOM where you'd have two actions per turn for each character.
PC Gamer: What do you think of the new XCOM?
Julian: It's great. It's very very good. It's different from mine.
PC Gamer: Jake Solomon, the lead designer, seems to have been very respectful to elements of it and has obviously just gone "but we need to make this work on consoles".
Julian: He was worried what I would think of it. He's changed so much. I think he was probably worried that I'd come up to him and say "Jake, you've been a naughty boy. What have you done to my XCOM?" but no. It wasn't like that at all.

PC Gamer: You've played it then, I take it?
Julian: I have played it. I've actually restarted it twice. Maybe I should try it on an easier difficulty level because I haven't managed to get to the end yet! If there's anything that's a problem with the game, it's that you can be playing it for quite a while without knowing that you are actually completely screwed and you should have stopped and started again.
I think my second playthrough I did a lot better but it got to a point where I could see I was in a bit of a downward spiral, and I just couldn't see a way out of it. I thought, well ok. I've got to restart again. I was losing too much funding. It's quite unforgiving, actually, in that sense.
PC Gamer: I was lucky that I never had a satellite shot down but I forgot to put any more up. I was just running with that minimal level.
Julian: That's the mistake I made on my first run through. I wasn't paying enough attention to the satellites. I wasn't getting the funding.
PC Gamer: Yeah, you need to circle the world. It's something that you learn as you play. Which is an interesting game design element.
Julian: It is, and pretty much every decision you make has to be fairly carefully considered, because there's always a very distinctive trade-off in decisions. I think Firaxis did a really, really good job. If you ask me, would I have designed the game in the same way? I would have to say no.
PC Gamer: How would you have designed it?
Julian: (laughs) I certainly would have gone back to my idea of generators again. I would not have accepted anything less than pseudo-randomly generated maps. I probably would have had less contrived elements to it. I felt that the... was it the Terror missions? Where you had to pick one out of three spots. Aliens are terrorizing three places. You've got to pick one of them and you have to -

Failing to maintain good satellite coverage proved punishing in 2012's Enemy Unknown.
PC Gamer: Ugh, God, yes. You know that the other two continents are going to be on minimal support and if something goes wrong, you're going to lose that funding on those two countries.
Julian: You're going to definitely lose out somewhere. You have to choose which one you're going to lose. I would have designed it differently, for sure. Would it have been as successful as the new XCOM? Probably not. No, I'm afraid.
PC Gamer: They probably wouldn't have given it the marketing money, to be honest. An awful lot of it was that they actually backed it, which was amazing. They backed a turn based strategy game on console.
Julian: That is absolutely incredible. I mean, it's unheard of really, unless it's Civilisation. Civilisation was the only game that was really surviving as a turn based franchise.
PC Gamer: And thriving, with Civilisation Revolution as well which was wonderful.
Julian: Exactly. It's actually made Take 2 Interactive the new Microprose because they're the only company that's got these really popular well known, established turn based franchises. Civilisation and now XCOM.
PC Gamer: Was there anything you would have added to the XCOM as it stands?
Julian: Well, yeah, the Geoscape is kind of missing. In the original game, the position of your bases - what you put in those bases - was important because aliens were active in particular areas, but the position of stuff in the new geoscape from the new game is actually, irrelevant, really. It doesn't really play any part in the game, so you don't have that. The Interceptors are based in each region. I guess my original game was a bit more simulation-ny and the new game is a bit more board game-y.
PC Gamer: Which is a way the industry's going. There's a whole video games made by board game designers section in the West Hall at GDC, so everyone plays board games now. I went to Jagex and Jagex have a whole room dedicated to their employees playing board games.
Julian: Yeah. This is very good and the new XCOM shows a lot of board game-y influences, without a doubt.

PC Gamer: You are a board gamer yourself, aren't you?
Julian: Yeah, I play board games. Absolutely. Far more than computer games.
PC Gamer: Would you design board games? Is that something you wanted to do or have done?
Julian: I do. Well, I have done, yes. Interestingly, Chaos, the game that I'm now remaking, was originally a board game.
PC Gamer: Was it a board game or card based?
Julian: Card based. Basically you had grid of squares, your board or arena. You had a wizard character, you put it on your wizard card and you had a hand of cards which was your spells. So to cast a spell you put your card down, roll the dice to cast it. If it's a creature it goes on the board, you can start moving it around and attacking enemies. If it's a spell, you have to resolve the effect of the spell. So yeah, it was originally a board game. On my blog I've got some pictures of the cards. I put them up a couple of months ago. So, I still have the original cards from this board game that I made. I often had lots of ideas for board games. I made one - a couple, actually - while I was at Ubisoft which we played with the level designers there. I've never tried getting any of them published.
I've got a question about Terror from the Deep, were you involved with that?
Julian: I had absolutely nothing to do with it.
PC Gamer: When I was a kid, I knew that it came out (and) I was extremely excited and then I played it and it and went, "This feels like an asset swap, except I can't use some of my guns on land".
Julian: I think pretty much the entire code base was identical to the first game. I don't think they really changed very much.
PC Gamer: Last year at GDC I spoke to Frederick Raynal who made Alone in the Dark. He had this thing where he made Alone in the Dark, he didn't sleep more than about 3 hours a night for a year. It got to the end of the year and the publisher said, "It's doing really well! We're going to put another one out. We're just going to do exactly the same thing. We're just going to make a clone and change a few bits” and he quit immediately. To me it seemed that Terror from the Deep had that air about it. Had you left the company?
Julian: No, no, yeah. What happened was, we started working on XCOM: Apocalypse pretty much the same time as they started work on Terror from the Deep. What really happened was that myself and Nick wanted to do a different game to XCOM, or at least do something a little bit different than just remake the original, so that's how XCOM: Apocalypse came about. There were some significant differences in the way that the game worked.
PC Gamer: Apocalypse had a bit of Sim City about it, I remember.
Julian: You were in this city and it had different organisations in this city with diplomatic relationships with each other and stuff. But they wanted a sequel within 6 months basically, this is what they wanted and we had to say "Well, it's not possible to do anything except re-skin the game with some (new) graphics".
Actually, they changed the story of course, I guess the clever bit, it was all about under the sea rather than Mars. Actually, it took a year to do the game. I had a huge team on it. Well, when I say huge I mean like, 15 people. Compared to just me and Nick and Helen, John and Martin on the graphics side of the original, this was much bigger.
PC Gamer: It must be strange to see studios with 400 staff, like Destiny, which is the Bungie game that's been announced.
Julian: Well I know from working at Ubisoft they have hundreds upon hundreds working on Assassin's Creed - more than 400. Assassin's Creed 3 is absolute bare minimum 600 people, probably, were working on it for most of the time worldwide across many studios.
PC Gamer: Their studio in Montreal, is it 2100 people?
Julian: It's huge. Ubisoft and probably other big publishers actually, they're making games by pure brute force.
PC Gamer: Having the Shanghai studio which is cheap to do lots of asset generation.
Julian: Yes. Obviously, these games require a huge amount of asset generation. It's like a factory. They're an immensely difficult undertaking, to be sure.

Gollop calls XCOM: Apocalypse "a disastrous project, even from the beginning."
PC Gamer: You had your huge team of 15 people on Terror from the Deep, is that right? Or XCOM: Apocalypse?
Julian: On XCOM: Apocalypse the team size for that actually was 5 of us at Mythos Games working on it and there was a team of artists at Microprose working on it as well. Again, it's a similar arrangement to the first game where we were doing the programming and Microprose were doing the artwork. But it was a disastrous project, even from the beginning, because one thing that happened is that the Microprose art team were trying to change the design of the game. Then they were failing to actually deliver anything that they promised. They just couldn't get the isometric graphic system sorted out in their heads. They did things which just didn't work, like they hired a guy whose name I forget to design the aliens, and this is a well known Science Fiction artist and he built these big models of the aliens and the idea was that they were going to scan them and put them into a 3D modelling software. It just didn't work. He had all this fine detail in these models and this scanning system just wasn't good enough.
PC Gamer: I do remember the aliens in it looking a bit blobby.
Julian: Then they had to recreate them basically in a 3D software they were using at the time. Yeah, they were awful, blobby things. They were nasty. Terrible graphics. It was very difficult.
PC Gamer: I still enjoyed playing it in the end, mainly because of jet bikes equipped with plasma cannons and missiles.
Julian: We had a real time system as well which was interesting, actually. It had some interesting aspects to it, but I don't think you can beat turn based games for simple straightforward playability.
PC Gamer: And planning tactically, as well. Responding on the fly was just tough, especially when you could just pause. Let's just quickly deal with Interceptor and Enforcer.
Julian: XCOM: Interceptor, yeah. That was the X Wing thing. XCOM: Alliance was an FPS one, yes. It wasn't a straightforward first person shooter, it was like a team based shooter, allegedly something similar to Rainbow 6. But with aliens.
PC Gamer: At what point did you stop being involved with making these games?
Julian: After Apocalypse. So, I had absolutely nothing to do with XCOM: Alliance or XCOM: Interceptor or any XCOM anything else. XCOM: Enforcer? Well, what happened there was that Microprose or Hasbro as it was by then, they had three Unreal licenses, I think, that that had to somehow use. XCOM: Alliance was using Unreal but because that project was going nowhere, they decided to "Well, let's just put out a straightforward Unreal-style shooter using the assets from XCOM: Alliance. We'll at least have something there to show for all the effort".
XCOM: Alliance was in development for a long time. How the development got screwed up, I don't know. As you're probably well aware, quite often games companies start and you're going for a long time and it just doesn't happen.
PC Gamer: This Milo and Kate seems to have broken Peter Molyneux's heart. They just gradually realised they couldn't make something believable. Yeah, it happens a lot.
Julian: It does. It's quite frequent.

X-Com inspired many official and unofficial successors.
PC Gamer: After Microprose and Hasbro stopped making them, suddenly in the late 90s / early 2000s people started making XCOM-inspired games with names like UFO: Afterlight, Aftermath. Some of them were really good, some of them were dreadful.
Julian: UFO: Aftermath arose out of my Dreamland Chronicles project. We did one game for Virgin Interactive called Magic and Mayhem, then I proposed to Virgin, "Why don't we try and do a remaining or remake of the original XCOM with, obviously, a different story? Make it PC and Playstation II". It was still a turn based game, still had all the elements of XCOM there. The tactical part was a little bit different because you controlled characters using a traditional third person controls for a console game.
If you've played Valkyria Chronicles on the PS3 then you've got an idea of how Dreamland Chronicles worked, because it's very similar. We had a little action point bar that would go down as you moved your character just like in Valkyria Chronicles, and when you wanted to shoot somebody you'd get the over the shoulder view, just like in Valkyria Chronicles. When you select characters it was on an overhead map, just like in Valkyria Chronicles.
So, it was looking promising, but Virgin Interactive had problems. They were sold to Interplay and then to Titus Interactive. Titus Interactive took one look at our game and said 'This is rubbish. This is so bad. Sorry, we're not interested in this.' Well, Titus were more interested in the IP that they got from buying Interplay. Whether they managed to do anything productive with it is another question.
So, we had to close the studio. We had a four-game contract with Virgin and now Titus but they were not going to fund this or any other games and we couldn't go to another publisher, so we had to shut the studio. What they did was they took all of the assets that we'd done and they ultimately ended up in the hands of ALTAR Interactive who made UFO: Aftermath. Unfortunately they stripped out our fantastic Valkyria Chronicles style turn based stuff and they put what I thought was a rather weak real-time thing in there.
PC Gamer: The last game they made, Afterlight, was actually good; good characters, a fun plot, interesting Geoscape mechanics.
Julian: I played it very briefly, I seem to remember. Certainly not very much, no. Unfortunately I very rarely finish games these days. Well, from my point of view I don't have the time. A lot of my game playing is more about research than entertainment because with limited time to play games, my interest is finding out what people are doing. At the moment, my main obsession is trying to find turn based games for iPad, for example, to figure out what is there out there that's interesting.
PC Gamer: I get an awful lot from BoardGameGeek.
Julian: There's a lot of board games coming out which is really cool. Very nice. But I'm talking about original turn based, to be tactical turn based games. There's one I like called Battlefield Academy which is also on PC, of course. That's quite nice.

FTL is the next game Gollop's lined up to play.
PC Gamer: What are you playing at the moment?
Julian: What am I playing? I know what I'm about to play because I just downloaded it before I came to GDC, which is FTL. I purposefully did not start doing it 'cause I had to finish my presentation so I guess as soon as I get back that's at the top of my list. Before that, I was playing XCOM, of course. I do play games on the iPad as well. The latest one is Battlefield Academy. Outwitters, I quite like. Outwitters is nice. Online turn based game, cutesy graphics, brutal gameplay. Chess-like.
PC Gamer: I haven't heard Chess mentioned once, apart from you, during all the time at GDC. It's not something people learn from any more. They don't reference it any more. That's really odd, considering it was, for 6000 years or however long it's been around.
Julian: I don't know. Maybe people think it's boring and that's all there is to it. If you like Chess, you'll like Outwitters. Outwitters has got a brilliant mechanic in it which is very simple. Each piece has a certain move, a certain strength - attack strength and defence strength - but you can only see the board as far as your pieces can move. So, there's a hidden area of the board, you have to be careful. You're not entirely sure what your opponent's doing. Very simply done. That gives the game a little bit of uncertainly and a bit of edge. It's quite nice.
PC Gamer: Can you see what your opponent can see?
Julian: Not exactly. You're not entirely sure what he can see. Most of the time, actually, you're not sure. Some of the time you're sure because the long range scout units, if you've got those up front on your front lines you know that you can see as much as he can see, because his scout units can't see further than yours sees. It's an intriguing game.
PC Gamer: Oh, that reminds me. The other XCOM game that was in development which has gone very quiet. Did you ever see that?
Julian: Oh yeah, the 2K Marin game. The only thing that I read is that they sort of rebooted it. Obviously, gone back to the drawing board a little bit trying to figure out what the identity of this game should really be. I think they got some bad reactions on several levels. One was the fact it was an FPS. Secondly, the presentation was a bit - this 1950s style alternate reality thing probably didn't go down too well with a lot of people, either, so it may be they're rethinking that. I'm not sure. Graphically, it was amazing.
PC Gamer: Thank you!

Apr 28, 2013

Julian Gollop interview: on X-Coms old and new, the Ghost Recon strategy game that never was, AI, auteurs and “Fork My Fruit”

XCOM: Enemy Unknown

Mar 26, 2013

GDC 2013: The AI tricks behind XCOM, Assassins Creed 3 and Warframe

XCOM: Enemy Unknown

In preparation for our not-too-distant subjugation by skull-faced machine-men, I thought I'd bone up on the latest advances in electro-brain design and stop by this year's GDC AI summit. Kicking off the summit was a trimuvirate of talks about the AI behind PCG-fave XCOM, stabby sequel Assassin's Creed 3 and the super-shiny “space ninjas with machine guns” shooter Warframe.

The talks showed a fascinating variety of uses for AI: XCOM's combat AI was the most immediately familiar, but supremely clever in insinuating the personality of enemy types - a far cry from the use of AI to determine Connor's foot placement in AC3. Warframe, meanwhile, deploys AI as a dungeonmaster, cobbling together levels from pre-built components to fit the needs of its players. It's smart stuff. Perhaps... too smart? Read on to unpick alien plans, parkour and player-centric dungeon design.

Alien nation - making XCOM's enemies distinctive

Firaxis had a problem in updating the classic X-COM (UFO: Enemy Unknown to Brits): how do you balance the game's appeal to modern and nostalgic audiences? Luckily, it seems it was a problem that they managed to solve, in part by way of a complex and hybrid approach to the AI, as described by AI/Gameplay Programmer Alex Chang (pictured right). The challenge was to revive X-COM's classic enemies, whilst keeping their behaviour distinctive and entertaining in the limited action system of the new game.

Chang's team did this by means of a utility-based system – a system that gave a measure of 'usefulness' to every possible action. This means that, at any time in the game, the AI rated each ability for each alien on the basis of its defensive, offensive and 'intangible' benefits. Each race also had its own inherent biases and special abilities which also affected different behaviour; so the Muton's 'Blood Call' ability, which buffs nearby allies, would be heavily weighted to be used, if there were other Mutons nearby and if they weren't already buffed.

Given the limited movement system of XCOM, one of the choices the AI made at a given time was to move or use an ability. Similarly, the units would generate a movement map of the area around them, to see what area gave the maximum utility. In this case, utility was generated by taking into account distance to the location, whether the location flanked an enemy or got the alien closer to flanking an enemy, the cover bonus the location gave, proximity to other aliens (to avoid grenading or rockets), the number of visible enemies (with just one being optimal) and an alien behaviour specific value. If the optimal location was where the alien was already... it stayed where it was and chose to do something else.

Of course, this only really applied to the normal units – the sectoids, thin men, mutons and so on. Fliers had an entire extra range of behaviour choices, and melee enemies were configured to charge pretty much directly at the nearest troops. Fascinatingly, this latter includes mind-controlled troops.

Meanwhile the elite units – the Sectopod and Ethereal – also didn't care about cover, but also didn't care about getting near to the enemy. The Ethereal was programmed to hang back and stay close to its bodyguards; the Sectopod was programmed to get as many enemies into range as possible, given its ability to attack multiple times in a single turn. With these special utility rules, the weighting towards choosing individual abilities, and the differentiated behaviours based on custom weights, the team ended up with about 17 different AI behaviour sets.

Happy feet: starting from scratch with Assassin's Creed's movement

This might not sound like an AI problem to us lay people, but the decisions on where Connor put his feet are hugely more complex than Ezio's lumpen feet – mainly, as far as we could tell, because Ubisoft Montreal is ramping up for the next generation consoles, which will no longer limit the complexity of their simulation. “The challenge was to change everything but change nothing,” said Aleissa Laidacker, Team Lead for AI and Gameplay, Ubisoft Montreal (pictured left). “The fans would have killed us.”

The four movements of the new Assassin's Creed engine were ground navigation, climbing, free-running and tree-running. While they were revamping these, they also revamped the animation system completely, making it totally procedural, as Laidacker demonstrated with sample videos showing Connor's reactions to varying conditions.

The basis for the movement style was the movie Apocalypto, with its wild forest-running. As much of Assassin's Creed 3 takes place in the heavily-wooded frontier, this was an important parallel, but it meant that Connor had to react correctly to the environment, whatever he was doing. This meant running, fighting and assassination animations all had to take place on uneven and even moving surfaces – considering the ships and sails, but also the rocky, bumpy surface of the frontier provinces.

Once the animation and AI team had made it so that Connor's feet could stand properly on uneven ground, Ubisoft Montreal's procedural animation guru, Simon Clavet was called in. His task was to ensure that the animation system could take advantage of this, by predicting where Connor's feet would end up as he was running and ensuring that his legs moved in the correct way. He did this by raycasting possible paths and making sure Connor's feet were ready to step over the highest point, and his pelvis was properly tilted. (This is scarily similar to the procedure the human body does automatically when we're walking.) Added to this, Front Strafing meant that Connor would step from side-to-side as ran; when players started turning, Connor would strafe first, meaning the animation wasn't disturbed if the player then turned him back.

The free-running model was also changed substantially for AC3. Every movement was given new animations, with short jumps chained together and long jumps separated off by 'settle' animations, so the animators could create new variations without having to think about how to integrate them.

Richard Dumas, the Technical lead (pictured right), explained that climbing was revamped by basing Connor's movement on that of professional speed climbers like Dan Osman, who can climb a 400m cliff in 4 minutes. “If he can, an assassin can too,” said Dumas. Connor doesn't settle after every move, like Ezio, but can flow from move to move, along the more organic surfaces of the frontier. Similarly, he doesn't just move up, down and sideways, but can move 360 degrees on the rockface, climbs vertical cracks in a totally different way, and has a dynamic system of how he positions his body, depending on how close to an edge his hands are or whether his feet are resting or dangling. Frankly, this was a crazy amount of work compared to its nearest climbing competitor... QWOP.

Finally, the newest form of movement was tree navigation. The trees came in three types. The unclimbable tree was smooth and branchless. The normal tree had anchors and horizontal branches, and could be climbed slowly. Finally, the V-shaped trees acted as fast elevators, and allowed players to hop up, from V to V, extremely quickly, so they could get back into position for assassinations.

To make it so that Connor (and the other characters and animals) had proper foot placement took the best part of three years work by AC3's AI department, and even now they're limited by the hardware power of the consoles. The movement behaviour was similarly involved. However, we're betting this totally dynamic system, which was under-used by AC3 because of console memory limitations, will make a more impressive reappearance in future editions of AC.

Warframe: AI-designed levels

Digital Extremes' new shooter Warframe may suffer from the GameFace / WarFace / FaceGame associations, but it's out today on Steam and worth checking out. In it, players battle in thirdperson co-opagainst a variety of AI factions, which level and scale in difficulty with the players. Daniel Brewer, the Lead AI Programmer (who we forgot to photograph, left), took us through the development of its procedural levels, which are AI designed from pre-built components each time a level is started, and auto-balanced to ensure that they're always challenging.

When the level is first generated, the game takes pre-built elements and connects a start block to various objective blocks and intermediate blocks, where the majority of the combat takes place, and eventually generates an end block, producing something that can be entirely linear or sprawling. Once the blocks are stitched together, the game works out a navigation mesh through them all and then a combat mesh. Yet, because the team don't know the orientation of the blocks to the player's route through the level and because they don't want players to have the same experience every time, they had to be very careful in the design of the AI that manages the levels and enemies.

The combat mesh – called the tactical area map – shows the areas of potential conflict. It also allows the AI to draw a distance map between the start point and the objectives, so it knows if the players are heading the right way, whether there are AI agents in the way and, if there are, where there are obstacles they can defend or chokepoints to fall back to. As the players move, the game spawns more enemies, with a higher density in in the direction of the objective and in the direction of player movement, acting as a subtle hint to players. Areas the players have left are deactivated gradually, reducing processing power and allowing agents in those areas to be temporarily removed from the unit cap (they reactivate if players head back their way).

Similarly the game paces these spawns by judging how the players are coping with the enemies they're fighting. The game will keep ramping up agent spawns until there's a lot of dead agent and players have taken damage. Once it recognises that the players have been properly tried, it'll slow the spawns down again, giving the players a chance to mop up and then heal up. An area that's peaked like this is exhausted and players can pass safely through it – until it's reset by the game rules (such as the players reaching an objective.)