Jake Solomon was starting to panic. For the second time in his career as a game designer he'd been given a shot at making the one game he'd always wanted to make: a true successor to X-COM: UFO Defense, the 1994 turn-based classic. But he was facing a problem, a really major one, in fact. Because again for the second time in his career, he couldn't make it fun to play.
About three years earlier, he'd been given a small team and just six months to create an XCOM prototype and well, in his words:
"We lovingly crafted the biggest piece of shit that anyone has ever made. I'm not joking."
That project had been scrapped and Solomon was moved over to Civilization Revolution, where he learnt from one of the best in the business: the true king of Strategy, the man with his name on the box, Sid Meier.
The game was a success and after two years working on the project and seeing it ship, Solomon was then given a second go at making XCOM, this time equipped with a much larger team and an entire year in which to create a prototype.
After 12 months they'd created a demo that looked good and ran smoothly, but there were some significant differences here compared to the game we'd eventually play. There was no cover system whatsoever and the soldiers' movement wasn't restricted to a tile grid, instead they could move wherever the player wished as long as they had enough Time Units left.
"I wanted to remake the original," explained Solomon over Skype. "But when I did that, I wanted classes and different weapon types and I wanted to add all these things on top of it. I'd made the original and then I'd added all this shit on top of it. It was really, really complicated. It was a real mess, actually."
The team finished its 12 month run and demoed the prototype to the rest of the company.
"We had this big party," said Solomon. "And everybody was so excited. Then the problem was, once everybody played it, they got a lot less excited."
Solomon was then forced to make perhaps one of the most difficult decisions of his professional life. He asked his team, who'd just spent an entire year working on his vision for XCOM to start over, from scratch. Most of the artwork and the gameplay systems had to be pulled out entirely as the developers stripped the game back to the basics.
In response to the feedback he'd received about his game being too convoluted, too messy, Solomon designed the most straightforward version of XCOM he could manage. Entitled 'Combat 2.0', this new version switched out free movement for a tile grid, added the option for units to take cover and then gave each soldier just a single number to work with. This one number determined everything: their offence, defence and health. So as you move into cover it goes up, and as you get shot, it goes down.
"It was a very basic system," said Solomon. "But you can see in the roots of this thing, what it ended up becoming."
This approach had the desired effect - when people played this prototype, as basic as it looked, they finally seemed to click with it. Rather than overcomplicating the ruleset used by the original game, Solomon and his team were now able to build their own here, piece by piece.
From these building blocks, things started to take shape. In an effort to make taking cover more interesting the team added high and low options, they gave each soldier two actions per turn, so that players would actually want to move without feeling like they were wasting their chance to fire a weapon, and they also split the various stats out into different values.
If you've played Enemy Unknown you're probably starting to think this all sounds remarkably close to what we played in the final game, and you're right, it is. But the team were about to hit another hurdle. One that seemed small, at least to begin with, but would promptly spiral out of control and to some extent, define the game's development for the next year.
You see, having these numbers on display was actually causing confusion. Players were treating them as if they were absolutes: this is my soldier's attack value and this is their defence and that's that, but in fact, the game was starting to become much more fluid than that. As Firaxis added new mechanics such as flanking enemies, or taking shots from a height advantage, these values weren't able to reflect that, which was an issue. And in his efforts to address that issue, Solomon opened up one hell of a can of worms.
He started by allowing players to toggle a new circular HUD on and off, which would give them all the information they'd need, but that didn't really do the trick either. It was only when looking at this HUD that players could really make sense of what was going on. Without it, there was still a lot of guesswork.
Which led to this.
Right, so, in this version of the game whenever you selected a soldier on the battlefield, it would draw lines from every soldier to every single enemy you know about. In fact, as you then moved your cursor around and decided where to move, you'd see these same lines drawn for each tile that you hovered over. And these colourful barber poles were trying to tell you just so much about what's going on.
"That's how all problems in design and game development start," recalls Solomon. "You agree to something without fully understanding the ramifications of what you've agreed to."
If a line is red, it means an alien can shoot at your soldier and if it's blue, it means the soldier can shoot at them. If it's blue and red, they can both take a shot. Now this clarification was needed here because in this version, XCOM's weapons had different ranges. A sniper rifle, for example, could shoot farther than an assault rifle. Okay, so, red and blue. That's not too bad. Right, and on top of that, if the line turns yellow it means either the alien or the soldier could take a flanking shot. This was important too because, also in this version, there was angled cover.
Angled cover doesn't exist in the game as we know it, where everything is positioned carefully at 90 degrees to work alongside the tile grid. Angled cover, as you might imagine, made it extremely difficult for players to know if they were in fact flanking an enemy. Terrified that players wouldn't be able to make sense of it all - the weapon ranges, the angled flanking - Solomon and his team were convinced they had to have the barber pole system in place.
But it was ruining the game. Players focused their attention on these shifting lines, checking every tile until they saw yellow. It didn't matter if they were fighting on a bridge, or a rooftop, or in the middle of a street. Nobody was paying attention to that stuff, all they really saw were these lines. And likewise, the developers were so deeply invested in this system, they couldn't see the bigger problem, either.
And that remained the case for an entire year of development. This tangled mess of striped lines that only grew more and more confusing as more and more enemies appeared on screen. This was considered an essential part of the game, despite the problems it caused.
Until one day, the XCOM team met, as they did each and every week for an event called Mutator Mondays. Anyone from the team was welcome at these meetings, as long as they had an interest in game design. You could be an artist, or a programmer, or Solomon himself, it didn't matter. Everyone would turn up and discuss the current state of the game and collectively come up with a mutator for that week, something that would change the way in which XCOM played.
This would usually be phrased as a question: what if every unit in the game had double the amount of hitpoints? What'd happen then? This idea would then be introduced as that week's mutator.
For the next seven days, whenever anyone booted up the current build of the game, it would ask if the player wished to enable the current mutator. Everyone would be given the chance to play around with this new feature and in doing so, learn more about the game.
Usually these mutators would cause a bunch of problems, or just make the game less fun, but even then the team would gain a better understanding of how everything slotted together. Upping the hit points, for example, slowed down the pace of the game and made the player's individual actions feel less impactful.
But sometimes, often by chance, the XCOM team would stumble upon an idea that immediately improved the experience.
"So the way that grenades used to work," said Solomon. "You would throw them and they wouldn't blow up on that turn. Again, this comes from the original X-COM, where you could prime them. One week, somebody was like: what if grenades just... blew up?
"The whole design crew thought that was a terrible idea. You get into this mode where you just accept the way things are. But we gave it a try and suddenly there was a way to get around the feeling you have when all your shot percentages are really low. It was this eureka moment. Grenades are amazing. When you have a bad shot... you can throw a grenade!"
This weekly brainstorm challenged so many of the game's features that had previously been considered fundamental. The biggest example of this was the barber poles. What happens if the team were to get rid of those? Well, they'd also need to drop unique weapon ranges and angled cover too. It was probably worth a go, thought Solomon.
And in that moment, XCOM was saved once again. Much to the dismay of the art team, I imagine, who would now need to rework a lot of angled cover assets.
I love that idea. Game developers change their games, sometimes drastically, all the time. That's a given. But I've never heard of a team approaching game design quite like this. It's easy when talking about strategy games, even as a fan, to be tempted into theorycrafting. But the XCOM team, aware that they themselves didn't even understand how all of their systems could interact with one another, avoided that temptation. Instead, they asked simple questions: what happens if we change this? Or if we remove that? And in doing so, they slowly figured it all out.
XCOM: Enemy Unknown is, without a doubt, my favourite video game. It's the reason I fell in love with the genre. Permadeath. Ironman mode. Missing 60 per cent shots. It's an incredible game made by an incredible team.
But the story of its development is not the one I'd expected to hear. There wasn't some perfect vision for what Enemy Unknown should be right from the very start. Rather the success of this game came as a result of a lot of mistakes.
And there's something extremely valuable to take away from that, I think. Something encapsulated by Solomon's Mutator Monday experiments. The team at Firaxis never grew too proud to question their own work and were, in fact, actively encouraged to do just that on a weekly basis. They somehow turned game design itself into a game and in doing so, ensured the XCOM reboot wasn't covered in those awful barber poles.
Will you support Eurogamer?
We want to make Eurogamer better, and that means better for our readers - not for algorithms. You can help! Become a supporter of Eurogamer and you can view the site completely ad-free, as well as gaining exclusive access to articles, podcasts and conversations that will bring you closer to the team, the stories, and the games we all love. Subscriptions start at £3.99 / $4.99 per month.