XCOM: Enemy Unknown Interview -- Terror From The Cutting Room Floor

I spent a disproportionate chunk of my PAX time on XCOM: Enemy Unknown, simply because as a die-hard fan of the original X-COM: UFO Defense, this is a game I've been waiting to play for literally 18 years. First I replayed the early levels using the PC UI, then I watched the Firaxis panel in which the development team discussed the crazy stuff they'd tried over the past four years of work, and finally had an hour-long chat with Lead Designer Jake Solomon, Lead Producer Garth DeAngelis, Lead Programmer Casey O'Toole, and Art Director Greg Foertsch where I got to pick their brains further and found even more insight into where decisions came from and what ideas they had for mechanics, gear, and aliens that didn't make the cut.



GameSpy: Some of the prototypes you showed at the panel for the strategic layer seemed like a massive departure from the original X-COM. How far were you willing to go to reinvent things?

Jake Solomon: What it was was I think I just kind of lost my way, and every one was a little bit further and further away from the original game, but I ended up back at the original game. I'm going to say something that's going to be a little controversial, but the strategy layer in the original game was, I think, the weakest part. I just felt like it was the original thing every time. There weren't many choices -- there were some, but once you'd solved the strategy layer, it was solved. I always played the same way: base in Europe, base in North America, base in Asia, no matter the difficulty I was fine. And I knew when I was building each of those. I wanted the strategy layer to be more interesting. I wanted there to be more choice, more explicit choice, but I didn't know how, and I think that's where I got into trouble. I wasn't sure how to do that, so I ended up going like "Ok, now the strategy layer is turn based."


And if it's turn based, that means you've got three missions up and you can only go to one, and it only gives you one thing. It's interesting in theory, but the problem is that the rewards, like if you make something turn-based you have to have a reward basically every turn, and XCOM isn't set up that way. That would mean that you'd have to go on like 50 missions to properly space out the tech tree. So it really was this real problem.

This is a little technical here, but in Civ, it's turn based. And because the turns are so small, you do such little small things every turn, they can then say, "Ok, research takes five turns, and building a settler takes this many turns, and every turn there's like a couple of things coming up. But with XCOM, if a mission was a turn? That's like 20, 30, maybe 45 minutes of gameplay is a turn, and at the end of that you had better have a reward. The player needs a big reward. Those turns were just too big and too long, and so I couldn't give the player multiple rewards and so it was the sort of thing where I wanted it to be turn-based, but it just didn't work out. So I got more and more lost, I spun off into more and more different prototypes. There was the number one that you saw, that was more like... what was the board game we played? Epidemic?

GS: Pandemic? That's what I was thinking when I saw it.

JS: Yeah, ok, that's very insightful, because that's what it was. As things happened, as UFOs appeared, they sort of trailed this noxious cloud that would then raise the fear in the city, and the cities at some point would start feeding other cities' fear, and panic would spread. And you were trying to respond in such a way that you were getting rewards, but also you'd better go to South America because it was getting infected with panic...


GS: So it was like a territory-control game?

JS: Yes. And it was interesting. That was actually kinda fun, but again, it wasn't XCOM, and the problem was that when you were staying on the Geoscape it was fun. If you just played the Geoscape it was very fast-paced, and you were kinda like fighting, and you were like "Oh god I've gotta shoot this UFO up here because it's red," that was very fun. But then you went into the base and then you did research and then you did engineering and then you did all these things, it was like this long pause of five to 10 minutes where you're just doing stuff, and then you came back to the Geoscape and it's this fast-paced thing and you're like wait, what the f***, where was I? I don't remember. So I'd made this fun little layer that was only fun if you stayed here. And again, it's one of those things that I'm glad I went away from because it wasn't XCOM, and people would've been "What the hell is this?"


Watch the pre-PAX XCOM: EU gameplay livestream.

GS: Did you ever mock up prototypes of an air-interception game?

JS: Yeah, we had multiple prototypes of air interception. I wanted it to be more air-combaty, and it was like a 2D thing. And that was not awesome. Oh, the very first one we did was a 3D version that was realtime, and it was like a flight simulator.

Garth DeAngelis: We got a lot of feedback for that. Why not make it more like a real-time classic thing where you can can control the jet and shoot? And Jake, I think wisely, said we don't want three huge games. We already have a strategy layer, we have tactical, let's keep the spirit of the original and have it be a transition to these types of missions.

JS: Right, and the problem with interception is that it's, you know, a five to 10 second experience, or even if you made it bigger and it's a 30 second experience or a five-minute experience -- and your strategy layer is like hours of work. And combat missions can be 30 minutes to an hour of work. And you're gonna tell me you're gonna overthrow the strategy layer because the player didn't play this five-minute minigame right? Dude, that is like, catastrophic. Players would've been driven crazy.

We prototyped things -- not me, but another guy prototyped a satellite game where the satellite sweep would go bing bing, and these little dots would come on screen and you'd have to hit spacebar... we prototyped everything. A lot a lot a lot.

GD: I mean, this is what makes Firaxis games great though, not just XCOM but all Firaxis games. We try things. It's ok if it doesn't work, we don't expect it to work. We just keep trying and trying new things. Everyone's ideas are valid until it's proven that they don't work very well.



GS: A lot of changes were obviously made to streamline and simplify, as the original game is unapproachable for a lot of people these days. But how did you guard against oversimplifying?

JS: I think that it's probably a case of, we put so much stuff in, again, a lot of this stuff just never made it, and what we did was we just cut it. And that put us at a point where, design-wise, the game was pretty over-designed. We had the time units, we had all the class abilities, we had the new weapons, we had the cover system. And that's when it broke. We made the original game, and then we added all this extra stuff. Our first version of this game, we had these weapon ranges, and everyone was drawing these sweep cones in front of them that were colored from green to red, and you could aim by pulling the left trigger, you could contribute more time to your aim, and so you'd pull the right trigger, and the cone would come in and turn green until it's like, that's a great shot. And it was like, there was way too much stuff in the game. It was like we took the old systems and we dumped a bunch of shit on top of it.

I was never too worried about the game being too simplified because it really was like, way too complicated. And so it was more of a question of like, look, anything that's not working we're going to cut. I actually -- If i'm being entirely honest, as a designer, maybe it's not too complicated, but the strategy layer, there's really a lot to do.

GD: We made the first prototype, which was the original XCOM. And if you were a part of Firaxis, you'd have been like "Yes! Everyone loves it!" But we had people who didn't play the original, and they had a lot of questions. A LOT of questions. And I think that 18 years after the original came out, it behooves us and new fans to try to make the design a little bit more elegant. And if that means oversimplifying it -- and I personally don't think that it does -- but you want things to be a little more intuitive so that you can get into it and enjoy the experience right away. We did go for a lot of complex things, but we filtered them down until it got to a point where you see people pick up a controller or mouse and keyboard, and they start playing on their own. And that's very powerful to me.


Part 2: The Rejects

GameSpy: So, what are some ideas that didn't quite make it?

JS: We had something called close combat, where we wanted melee in the game. We knew we were going to have melee aliens -- the zombie, Chryssalid, Berserker. And so the idea was that, like now, it's the original game, where your soldiers can shoot at close range.


GS: But there's no rifle butting or anything?

JS: Right, but we did have something called close combat, which was something that was triggered when your units -- we drew these rings around them -- when you moved within that range you triggered close combat, which was like... oh man, this was not good. It was like, you basically chose stay and fight or run, and it was a percentage chance based on what you were loaded out with. GD: It was one of those things, it never felt satisfying.

JS: It almost felt like a quicktime event. It wasn't, it was a strategy, but... and if you ran away, you could run half distance. We had so many iterations of all these ideas.

GS: So there's no hand-to-hand combat in there?

JS: Well, I won't lie: I like the idea of hand-to-hand combat. But for us, that was just a bridge too far. We had our hands full, and when we realized close combat wasn't going to work -- that was about two years ago -- we realized man, we've got a lot of work to do to get this game finished, so forget melee, they're just going to get up on each other. I'm not opposed to the idea of melee -- it was something we always wanted. We had a knife at one point. So we had it, but at some point I realized... I think I fell into a trap a lot of times, which is as a designer you should never do, which is like, "Oh, this isn't fun, you know what this needs? More f***ing design. It needs another system on top of it. Oh one number doesn't work? You need two numbers. Two numbers doesn't work? You need to see the sight range of the weapon." It's a dangerous trap to fall into as a strategy and systems designer, you're just like "Man, if I put this other thing in there, finally, it's going to really come together." So it's one of those moments where it's like alright, forget melee for now, and we'll just move on.


GS: The original had only one hand-to-hand weapon, the stun rod. Do you still have that?

JS: Yeah, we have the arc thrower. And that one, I guess that is melee. It's very short range, anyway -- the same range as that close combat. I just changed the name to the arc thrower, but it is the stun rod. So what it is is you have to get in close, and then you pull out the arc thrower and you stun him with it and it fails or succeeds, so it works a lot like the stun rod.

JS: Remember how grenades worked until a couple of months ago? Some of the team missed this You would throw them and they wouldn't explode until the following turn. So you would throw it, and you would see the radius, and you'd have to move. So grenades were a way of flushing units.

GS: It was like that the first time I saw it at E3.

JS: That's right, because we said, we're going to throw a grenade, then we're going to suppress him with another guy, and that's how we force him to eat a grenade. And then after that, I was like, it was just a question of focus testing and people who were new to the game were starting to play, and they were like "What the f***, I throw a grenade and the thing doesn't blow up?" And that had been in there... it was an original mechanic, so like four years. And I remember I changed that and said you know what, grenades are going to blow up immediately.


GD: That's a prime example of external focus testing. And Jake and I joke about that because I loved how it worked. It's an example of me being close to the game, loving to set that up and using it as a flush mechanic. And we had to change the ability tree, because we had abilities that let you cook the grenade that would let it blow up right when you throw it.

JS: And if you held onto it when it was cooked and you got wounded, you'd drop the grenade and you'd hurt yourself.

GD: Which is all awesome, but to Jake's credit, he saw we had a lot of focus tests, and across the board people were like "I just threw a grenade and nothing happened, why?" And I guess it's something we could've tried to teach better, but it was something that people were confused about.

JS: And it just wasn't as fun to me. And these guys may disagree, but I love throwing grenades. They punch holes, which is nice, but they also destroy artifacts, and so you don't use grenades and rocket launchers unless you have to, like you don't want to do that unless you must. Or you do it to destroy cover or something.

Casey O'Toole: I told him, I said I'm never gonna use them. I'm just never going to do it. Yeah, I use them a lot. It's true. It's just so hard.

JS: In some cases, on classic, it's like where's my f***in' heavy, I don't care if I don't get anything out of it.


GS: What kinds of equipment did you try and discard because it didn't work out?

JS: The motion detector did not make the cut. It was something that, gameplay-wise, for me, it wasn't that interesting. We have a sound system now where you can actually hear aliens in the fog, so that kind of fulfills that role of the motion detector. But I actually didn't want the player to have a motion detector and use that to sort of "Oh, the aliens are over here now." And there's big blobs and small blobs, like in the original. Some of these things come out of my personal playstyle in the original. I didn't use the motion detector in the original, so for me it didn't have much value.

One of the things that we had... remember the Tesla Hammer?

Other guys enthusiastically agree

GS: Sounds like a melee weapon.

JS: The Tesla Hammer was like, you could open a door anywhere in a wall. You could walk up to it like you would to a door, press the A button, he'd turn his fist on, or whatever it was gonna be, and just punch through a wall and open a door. It's not that valuable now, now that grenades blow up immediately. It's not that useful.


Proximity mines! They were kinda lame. We had working proximity mines in the game. But it was the sort of thing where, the only thing I see them being valuable for -- so who knows, in the future, if we do extra content -- multiplayer, they'd be f***n' awesome in. But in single-player, you don't know where the aliens are coming from, so you'd throw them at an alien who's already there, and then he moves and he blows up, so why the f*** didn't you just throw a grenade? So they just didn't work out very well.

The Flare Shot -- that was where the sniper could load ammo and shoot, and that'd create a flare. But again, in certain situations it was like the worst thing ever. Like, does it shoot through walls? No it doesn't. It was useless indoors. It was replaced by the Battle Scanner.

GS: Did you do anything with incendiary weapons, or flamethrowers?

JS: We did have flame, but we probably don't want to talk about that too much, because they were interesting. But again, it was something that was problematic. A lot of technical issues.


CO: That's another great small example of something we quickly prototyped, I think system-wise, of all of the other weapons that we had, it was so different.

JS: Yeah, it's like "Great, I've got a flamethrower!" but you've got these long f***ing sight lines. So your flamer guy is going "Woooosh" and the aliens are like "Alright, don't walk there, shoot that guy." So it was like, well...

CO: Not to mention you set the whole entire level on fire and your framerate goes down to like two. So there are technical issues.

JS: At one point, I don't think we ever implemented it, but there was talk of having a flamethrower SHIV. And it was like, eh, that would be the worst unit ever, just this tank that goes "blaaaah!" I think it was the sort of thing that was like a snap judgement, like "You know what would be great? A flamethrower!" And then we put it in the game and it was like "You know what's not great? The flamethrower." Which isn't to say that with the right design it couldn't work, but there's just so much in the game that it got to the point where we needed to start cutting stuff that's not fun.

It also broke the class system too. It didn't fit any of the classes, so we weren't sure how to work it in. It was a shitty weapon, so we were like, we can give it to the Heavy I guess, but it doesn't suppress, all their abilities don't work with it, so...

GS: Have you messed around with deployables or personal shields?

JS: Yeah, we did have something like that, but that became the Alloy SHIV that you can take cover behind. So that filled that role, and smoke grenades kind of filled that role too.

Then there was the Tech soldier. Never even reached implementation. The SHIVs had to be controlled by him, so to bring a SHIV you had to bring a Tech soldier. Again, it's one of those things that just sounds nead from a mechanics point of view, but from the players' point of view, that is shitty. The aliens kill this one guy, and you're like, oh, there's my SHIV which I can't use. Just not a great idea.


GS: What about alien types that were left on the cutting room floor?

JS:We had the Silacoid, the rock-spitting, alien that didn't make it. When we designed the aliens, we got a pretty good list, gave them all abilities, kind of slotted them in tactically. Then some aliens we went, this one's kind of a retread, or it's lame. Like the three-story toad [which would eat soldiers who could be rescued if you killed it in time]. Mechanically it was kind of interesting, but that never reached implementation. We just cut that.

We had a lot of aliens, and it was like ok, we have a budget for aliens, some of them made it in, some of them were different. And when we started cutting back, we asked which of them were going to be interesting tactically, and the Thin Men already had a poison spit, and now we had two poison aliens and that's not very interesting. And we didn't want a lot of beast aliens -- we didn't want a lot of those, because it's more of a tactical game. So we had the Berserker, we had the Chryssalid, even things like the Cyberdisk is kind of a beast alien, a non-cover-taking alien. But we didn't want too many of those, so we sorta cut those guys. The toad was never going to work.


More to come a bit later -- there's a lot to go through here.