The Making Of Assassin's Creed, Ubisoft's Original Open-World Epic

As part of our end-of-year celebrations, we're digging into the archives to pick out some of the best Time Extension content from the past year. You can check out our other republished content here. Enjoy!

"When we finished the project, I went on holiday. I was convinced the game was a flop. No one was going to buy this. It had so many imperfections." Mathieu Mazerolle, lead programmer on the original Assassin's Creed, was wrong. Imperfect the 2007 game may be, but a flop it most certainly was not.

By the time the sequel was released, Ubisoft announced Assassin's Creed had sold more than eight million copies – an impressive debut for any new IP. In 2022, the franchise it spawned is celebrating its 15th anniversary, with total sales standing at over 140 million.

As has been well documented, Assassin's Creed began life as a Prince of Persia project, a follow-up to the hugely popular Sands of Time trilogy. It was the first title from 'the POP team' (as they were known internally) for the next-generation consoles, PlayStation 3 and Xbox 360, and one that diverted significantly from their past projects.

The bulk of the game is spent wearing the hooded robe of Altaïr Ibn-La'Ahad, a member of the Brotherhood of Assassins who is tasked with eliminating nine individuals from the Templar Order, eventually uncovering a plot to claim a mysterious artefact: the Apple of Eden. Interspersed with this narrative was the plight of Desmond Miles, Altaïr's descendent, who is forced by the Abstergo corporation to enter a machine known as the Animus, which allows him to love through the assassin's memories.

Unlike the Prince's largely linear and more contained adventures, this was an open-world affair that allowed players to approach the central assassinations from any angle. "We wanted to push hard on the living, breathing world aspect," says Mazerolle. "That was very unique to us; the scale of the world, the crowds, the interaction between characters that wasn't really there in Prince of Persia. Those were really new; they were there from the beginning and, from a gameplay point of view, those were the biggest challenges."

You could go anywhere, make an assassination, and do it within 20 minutes. You meet the bad guy, you sneak in, you kill him, you escape and you have the revelation at the end, and that's all done in the open world

Associate producer Simon Tremblay adds: "We wanted to have that sense of adventure, where you go out and find your missions, you develop the story as you progress. You had so much freedom in the first Assassin's Creed. You could go anywhere, make an assassination, and do it within 20 minutes. You meet the bad guy, you sneak in, you kill him, you escape and you have the revelation at the end, and that's all done in the open world."

Most of the game was spent in one three cities: Jerusalem, Damascus or Acre. And Mazerolle says the decision to set Assassin's Creed during the Crusades raised more than just a few eyebrows: "It was dangerous territory for us. The legacy of the Crusades, it was all loaded with religious and cultural history. We got anger from Muslim groups for how we were portraying certain things, we got them from other religious groups, but I think it's cool for games to explore that sort of history and lore and take their own fantastical slant on it."

Perhaps one of the most unusual aspects of the game is its structure: Altaïr is never given a glowing line to follow or a map marker to track his target. Instead, he needs to investigate, completing shorter missions and activities that will give him clues as to where he'll find his mark and, crucially, what they look like. Even then, if you're told to take out the man in the marketplace with a beard and red cape, you may find multiple such men, so more information is needed to ensure you don't kill an unsuspecting innocent.

It's a level of freedom and player agency that many modern-day titles, arguably including later Assassin's Creeds, lack. Mazerolle tells us the original intention was for there to be even less guidance. "In the beginning, it was an open-world game, and even that level of structure wasn't there," he says. "It was literally 'Here's the mission' and there'd be a sort of win condition and then you went into this living, breathing world where people would have a disposition to you, your actions and the consequences of your actions, and it would be completely different every time – but we found out that was just super not fun. There wouldn't have been enough structure for the player to stay engaged."

We'd test them out, and if we could make it fun with different coloured dots in a small Flash demo, we knew we might have something

The developers experimented with how they could build missions and set objectives without making them feel too scripted, which might distract from that living, breathing world around them. One method was to prototype missions in 2D using Flash, with different coloured dots to represent the player, guards, key characters and NPCs. This would help simulate an eavesdropping mission, for example.

"We'd test them out, and if we could make it fun with different coloured dots in a small Flash demo, we knew we might have something," Mazerolle explains. "That led to our building of the game design docs, and we had this really structured process that started to establish much later as the game evolved. We had none of that for the first couple of years. The first couple of years was the Wild West, 'build a living, breathing world and it'll come out in the wash.' When we realised that wasn't going to work, we steered towards these semi-structured missions, and I think that worked out very well. [The] only thing I think we could have used was maybe more variety in the mission types, but I think the ones we did pull off, we saw some good promise with them."

Crowd Gaming

In part, the game's direction was driven by one of its most impressive technical achievements: the simulated crowds. Open-world games in 2007 tended to be sparsely populated, at least in comparison to Assassin's Creed where dozens of bystanders milled about its narrow city streets. Tremblay recalls that lead level designer David Châteauneuf, though impressed by the simulation tech, once asked: "What do you want me to do with this crowd? I can't make any gameplay with this."

Tremblay adds: "This is where the ideas like the monks you could blend in with came from, because you needed tools that used the crowd system to make gameplay and challenges, to use that crowd to make things difficult for you to approach the target."

The team would go on to coin the term 'social stealth' and enable players to identify, approach, assassinate and evade a target without ever being noticed by the many potential witnesses around them. Again, the crowd was central to this as their reactions would affect how noticeable your character was; perform actions without attracting the attention of other people and the patrolling guards would ignore you, but cause alarm or act too suspiciously, and the guards would actively seek you out, inevitably leading to a chase and potentially combat.

"That was from Patrice Désilets, the creative director," Tremblay recalls. "From the get-go, he wanted a system where you were either high profile or low profile. As long as you weren't holding the triggers to run or jump, the crowd wouldn't react to you, but as soon as you activated what we'd call the 'high profile trigger' to climb on a wall, to jump and stuff like that, the crowd would start reacting. There was kind of a barometer of your actions, and if the crowd reached a certain threshold, they would call for guards. You'd be in the cycle of needing to find your target, and that might mean climbing because you need to sneak or find a way in, but then the crowd reacts, and you might have to hide and try again."

Again, the Ubisoft Montreal team could draw on experience from a previous title – not Prince of Persia but Tom Clancy's Splinter Cell, the studio's first AAA project. In the first title, players were compelled to take a stealthier approach because if the alarm was triggered three times, the mission was over. With Assassin's Creed, the trial-and-error structure would have been too disruptive, ripping players out of the open world – so instead, the team gave players options to lay low until the heat died down: blending in with the monks, hiding in hay, and so on.

We liked the parkour of Prince of Persia. We thought that would be a really great element… I think that's still a super innovative mechanic. Almost every game I've seen to date does it in a way that's either about as good, or not as good as Assassin's Creed

This is also why, as with Prince of Persia, the assassin cannot die. While the prince's failures were passed off by the narrator as a misremembered moment, Altaïr's apparent demise is labelled as 'desynchronisation' of the Animus simulation.

Mazerolle explains how the previous trilogy also gave Assassin's Creed its signature traversal system: "We liked the parkour of Prince of Persia. We thought that would be a really great element… I think that's still a super innovative mechanic. Almost every game I've seen to date does it in a way that's either about as good, or not as good as Assassin's Creed. The technology we put in and Richard Dumas, the guy who basically programmed the Prince, and his team… The Prince and the Assassin were both left-brain-right-brain combos between Richard and Alex [Drounin, animation director for Altair]. You couldn't separate these two; they'd finish each other's sentences, and the fact there's this certain je ne sais quoi to how the character moves and feels, it's because of these two people. When you look at how other video game characters are designed and move, you get the impression they're designed by committee."

Tremblay adds that Assassin's Creed's parkour was the natural evolution of the acrobatics from Prince of Persia: while that character could only climb where the developers allowed him to, the team wanted to enable players to take Altaïr anywhere. They even ensured Altaïr automatically leapt from object to object while players effectively steered because, as Tremblay puts it, "we were not challenging you to succeed in the acrobatics; we were challenging you to choose the right tactic to get to the right place or complete the mission."

Building The World

While the vast majority of the game is spent in the cities, the team was keen to deliver on the promise of a living, breathing world. As Altaïr travels between towns, he can climb towers to reveal more of the map, collect flags, and take down Templar knights scattered around the landscape.

Tremblay is also proud of how the open world was structured to mask loading times. One of the team's proudest achievements of the Sands of Time trilogy was the absence of loading screens, at least on the PlayStation 2 and GameCube versions. Even though they shifted towards an open-world structure, the developers wanted to remove as much loading as possible.

"I don't think we fully succeeded, but once you were outside or in the city, there was no loading gate – you were just playing smoothly," says Tremblay. "That's more like a technical feature, but it's something we were proud of and something we wanted to carry forward from our previous games."

Assassin's Creed was by far the most ambitious game Ubisoft Montreal had ever attempted, and that came with inherent technical challenges – not the least of which was building a brand new engine. At the time, Unreal Engine was the pinnacle of development technology, but Ubisoft decided to invest in its own engines rather than paying for Unreal, which was much more experience at the time. "But when we planned to build the technology, we underestimated the time it takes to build a game editor for people to actually make stuff," Tremblay says.

He uses the parkour as a prime example; one member of the team spent three years working out how to generate the physical world and enable that automatic traversal. "Back in Prince of Persia, we had to flag each individual edge and surface that we wanted you to be able to climb," he explains. "With Assassin's Creed, we decided to generate that automatically, so we had to build a system to identify that invisible world the assassin understands in order to grab every ledge and stuff like that."

Mazerolle agrees that everything was a challenge – not just in terms of making the team's bold ideas work, but also making them enjoyable. "Honestly, Assassin's Creed was one of those games where if I told you the whole game came together in the last two months before we shipped it, you would probably freak out. But it literally did because we spent so much time throughout the development cycle trying to think about how you make a living, breathing world fun."

Assassin's Creed was one of those games where if I told you the whole game came together in the last two months before we shipped it, you would probably freak out. But it literally did because we spent so much time throughout the development cycle trying to think about how you make a living, breathing world fun

He notes that there were a few key additions to the studio that helped accelerate the project. At first, the team didn't have all the components needed for full development – that is until Ubisoft installed Jade Raymond as the game's producer. "Jade assessed the team, and figured out that the Prince of Persia team was pretty great at building an engine and core gameplay systems, but couldn't split the team between the engineering and the game engine side. So she suggested hiring a lead engineer for the engine, and at the same time, there was a lead engineer for the game. That's around the time I met Jade, saw some footage of the game, and got on board as the engineer for the game side. Of course, because we were building both at the same time, the lines were always blurred."

Meanwhile, Mazerolle credits the arrival of game designer Maxime Béland as the moment the project's core systems started to be locked down, drawing the line between what used to Prince of Persia and what would become Assassin's Creed. Another challenge was keeping Désilets' vision in check, limiting the scope of the project in order to ensure it could be finished.

"Patrice was really focused on the living, breathing world aspect," Mazerolle explains. "He wanted children in the game; he just wanted to populate the world. Max came in, and he really wanted to make the world mechanical, where you had missions and things like that. We had to really bring those two things together and make a lot of sacrifices and compromises. How do you support the gameplay without sacrificing the believability of the world? I think all the resolutions to those kinds of things only came together in the final phase of the project. We didn't even have the elements needed to make those trade-offs until very late in the project, just because of the monumental effort in building everything."

The Not-So-Hidden Blade

Mazerolle is not exaggerating when he says a lot of sacrifices and compromises were needed to complete the game. Originally, the team had planned more mission types, more activities for players outside the cities, and more to do with the Animus, but much had to be cut in order to deliver a finished game.

Tremblay offers more examples: "At some point, we had a whole economic system to be part of the game so you'd be able to hunt and sell pelts, we had a whole crafting system, so we had big ambitions for the open world bit. But we had to just let those things go; otherwise, we would just never have released the first game. I think a lot of the stuff you find in the second game was actually planned for the first one."

Tremblay reiterates what is well known by now: Assassin's Creed was originally planned as a trilogy, working linearly through history until it culminated in a present-day adventure outside the Animus, set in New York story. The meta-story would have tied the three games together, but the first game (and particularly its sequel) were so successful that Ubisoft opted instead to string it out into a then-annual franchise. As a result, Tremblay feels the Animus element suffered. "I always felt we weren't using it enough or that it felt too much like two separate games," he says. "I would love to see the franchise wrap up the metagame, which just never gets resolved. There's probably a spin-off to be made on just the present-day stuff."

The conspiracy aspect was a really important aspect for Patrice, and not just for the first game. He saw that for all the sequels as well

Mazerolle gives "huge credit" to his friend and colleague Mathieu Leduc, who designed all the futuristic aspects of the Animus side. As for the overarching meta-story, he says the full vision lay primarily with Désilets. "What he saw in where the game could go, and the conspiracy theory and all that type of thing was, it was wacky, it was crazy, it was really built out, and in his mind, it was super vivid," he says.

"The conspiracy aspect was a really important aspect for Patrice, and not just for the first game. He saw that for all the sequels as well. The games sort of diverged from Patrice's roots, the idea that you were reliving memories. The idea was that you were going to be moving linearly through history, and the series didn't really stick to that; they just picked periods of history that were fun – which is fine, but Patrice wouldn't have done it that way. He always had a different vision for what the Animus side meant."

Assassin's Legacy

Fifteen years after the debut of Assassin's Creed, the influence of that game can be seen far beyond its own franchise. It was instrumental in popularising the use of climbing mechanics in open-world titles, and it would be remiss not to mention the abundance of map-revealing towers seen during the last decade and a half of gaming. Even that game engine the team worked so hard to build, now known as Anvil, powers many of Ubisoft's biggest releases.

"To have been part of starting a billion-dollar franchise and a major game engine, that's something I'm super proud of," says Tremblay. "Prince of Persia was probably the best work experience of my life, but in terms of being in the right place at the right time with the right team to make something special, Assassin's Creed was the next level of this in terms of achievements."

Mazerolle adds: "I'm super proud that it's laid the foundation for a whole series. You see the influence almost everywhere, but the reality is with every game you can trace the roots back. You can trace the roots of Assassin's Creed all the way back to Tomb Raider and even the gameplay mechanics of things like Hitman. Nothing really new comes out in games, but there's the right place, the right time, the right aesthetic. What Assassin's Creed really did was it raised the bar of what an epic, open-world game could be. Grand Theft Auto and a lot of those games were pushing at that open-world type of game, and that's the legacy is Assassin's Creed raised the bar. Everyone's still chasing that – even Ubisoft."

Mazerolle left at the end of the project, choosing not to work on a sequel – especially as Ubisoft wanted to develop two Assassin's Creed games simultaneously, a model it still follows to this day. Over the years, the original Assassin's Creed team would be broken up, with key members departing under a variety of circumstances.

Désilets left Ubisoft in 2010, seeking more creative independence. He eventually ended up working at THQ Montreal, but when that publisher went bankrupt a few years later, the studio was ironically purchased by Ubisoft. Désilets was let go a few months later after he and the Assassin's Creed publisher failed to agree on terms, prompting a three-year legal battle for the rights to 1666 Amsterdam, the project Désilets had been working on for THQ.

Raymond moved to Electronic Arts in 2015, forming Motive Studios, but left three years later to work on a "top secret project" that turned out to be Google's cloud gaming service Stadia. When Google closed its internal studio in February 2021, Raymond formed her own company Haven Studios, which has since been fully acquired by PlayStation.

Béland left Ubisoft in 2019 for a brief stint at Fortnite and Unreal firm Epic Games, before returning to the Assassin's Creed publisher less than a year later. However, in 2020, he faced allegations of misconduct and abuse; Ubisoft placed him on administrative leave while it investigated these claims, but Béland resigned before this process concluded.

Finally, Tremblay left less than a year after Assassin's Creed debuted, spending two and half years at EA before working at a variety of smaller games companies and ultimately moving into other tech sectors. Reflecting on his time at Ubisoft, he most fondly remembers the free rein he and his colleagues were given while working on Assassin's Creed – something he believes is key to why the game is so special.

"In those days, we were just trying to make the game we wanted to make," he says. "After that, there were too many studies of what other games were doing, what GTA or God of War was going. Editorial at Ubisoft was sending us too much [feedback] on how we should do this or that, but prior to that, we were free to do what we thought was right and improvise. I think that's why it feels so genuine. It's not perfect by any means, but it was what it was and it started something new. I remember Ubisoft wanted so much to have a GTA-like game with an open world and cars and stuff, and I think we had a team working on that for five years – but then we realised we could do an open world with a horse. We didn't need cars. We were building a GTA killer from Ubisoft, and I think that led the way to let us make it big and open world. I don't think it was super planned."