Ubisoft Scalar aims to make games more like the web

Everyone loves a good analogy. It's a great way to get your point across, especially if you're trying to explain something that doesn't exist or cannot be shown in a way that everyone understands.

Ubisoft Stockholm's managing director Patrick Bach is certainly a fan of an analogy as he frequently compares Scalar, the publisher's new cloud-native development technology, to the world wide web.

"Games aren't using cloud like many other services are," he tells GamesIndustry.biz. "If you look at how you use the web today, that's way more advanced and more mature than what games are doing. Technology used to be the thing that made games cool, we were at the forefront, but we've been lagging behind because we've been stuck in our old ways and no one has done a big shift in the way you think about how you make games."

His comments come as he explains the origins of the project. A group of Stockholm staff began by questioning why video games are developed in a certain way, and what the medium could be like in ten years' time. They then looked at what is currently lacking before games reach thatpoint, and cloud functionality was a significant factor.

Bach emphasises that this is not about Netflix-style streaming, but more about moving compute, memory and all the other fundamentals that allow a game to even operate into the cloud. The team hopes Scalar will become a new foundational technology built around what the team believes video games should be.

"We're talking about certain types of games that don't exist today because the technology limits you," Bach continues. "Scalar allows you to have greater scalability, meaning you can grow your game into sizes that you weren't able to reach before. The memory footprint of the cloud is far greater than that of the box next to your desk. Also, being able to use resources in the cloud -- CPU, GPU, etc -- will also render you possibilities that go beyond what you can do today.

"And all of those things being on demand. Once things are in the cloud, much like web services being on demand, you can just get whatever you want, whenever you want it. You can scale your game up in real time, right in front of the eyes of the players, or you can scale it down when you don't need the resources. That's just not a possibility with local hardware; your machine is either on or off, you're either using the resources or you're not. Which is a total waste of resources, to be honest."

The ability to update and change games without needing to shut down servers or put them in maintenance mode is a particularly fascinating aspect of Scalar. The industry has shifted dramatically in the last ten years towards live service games hosting thousands if not millions of players 24 hours a day. Ubisoft is no exception, thanks to the likes of Rainbow Six: Siege and For Honor, and even its single-player games like Assassin's Creed and Far Cry 6 have live service elements. Bach is confident that enabling Ubisoft's developers to make changes on the fly will allow not only for a smoother service, but also the rise of brand new game types.

"Games aren't using cloud like many other services are. How you use the web today is way more advanced than what games are doing"

Patrick Bach, Ubisoft Stockholm

"If developers could add changes to the games faster, they would," he posits. "If it was easier, they would change more stuff and make their games better on a higher cadence than they are today -- especially if it wouldn't interrupt play. So the need is there, the urge to give players better experiences in real-time is already there.

"It's back to the web again. This is done on a daily basis today: if you go to any website and they update the site, it could completely change. Sometimes when you're scrolling through a site, it starts to change -- this is already a reality when it comes to other types of entertainment and services, so why shouldn't it be for games?"

He expands on this, pointing to the abundance of real-time information at play in many cloud- and web-based services we use on a daily basis: "If you want to use a map today, you don't go to the shelf and pick up a map or atlas. You go to the web and you have all the maps in real-time. You can zoom into any place in the world, and we take that for granted.

"Today we take it for granted that you can use the cloud to browse through petabytes of data in real-time and that's just the way you use maps today... For me, it's almost obvious that this will be the solution for games in the future. Maybe not for all games, but for some and perhaps the majority."

Per Olof Romell, technical director at Ubisoft Stockholm and product director for Scalar, adds that the ability to make real-time changes will also benefit the company's developers internally.

"When we test games, it takes a long time to make a build, you test it and something's broken," he says. "Then you have to wait until tomorrow to test it again. That's not the reality anymore. Now if we see something that's not functioning, we call a programmer and ask them to fix it, and then we see that happen onscreen as we are playing it. That's mind-blowing."

He adds that the tech can enable new ways of working -- the most obvious being additional options for remote workers in this post-pandemic world -- as well as changing the structure of how games are made.

"The concept of game engines have been the same for a long time: you have a huge blob, you remove stuff you don't need, you add stuff and modify stuff and that's your game," he says. "Whereas we feel that what it should be is you have a library of functionalities and you look at the design of your game and what you need for this, and you compose what's going to be the engine for your game just with the relevant pieces. It gives you a much more nimble way of working, and the freedom to use the tech you want."

He continues: "Scalar is all about providing more options for our creators to create a broader spectrum of experiences, some of which have not been possible before... It's not all about scale and huge worlds and many, many players. It can also be about super complex simulations that you're used to seeing in movies but now you'll see in games in real-time, which will be really impressive."

Watch on YouTube

Of course, we've heard many of these promises (and those in the video above) before. Improbable has been talking about using the cloud to enable never-before-seen game types, with larger worlds, more simultaneous players, complex simulations and persistent player impacts for almost ten years now. So far, only two notable games have been released that use this technology: Bossa Studio's Worlds Adrift, which was closed after one year of Early Access, and Midwinter's Scavengers, which is still in Early Access. The company has even sold off the studios it created or acquired to demonstrate the possibilities for cloud tech in gaming.

"The previously always-known limitations that have been around for the 20 years I've been in games development will be gone, forever"

Per Olof Romell, Ubisoft Stockholm

Similarly, Hadean has been raising funds and partnering with the likes of Eve Online developer CCP for experiments that push the number of simultaneous players past the usual limits, but a final product has yet to emerge. Even the early cloud gaming services have struggled. The many woes of Google's Stadia have been well-documented, and Square Enix closed Shinra after two years.

So what makes Scalar any different? Romell admits that when the project began, he was also sceptical.

"Obviously we looked at the solutions that are out there -- we don't want to build something that we don't have to -- but we didn't really see anything that did what we wanted to do," he says. "For us, the difference is we're actually proving. This is not theory, we have this working. We'll get to the point where the previously always-known limitations that have been around for the 20 years I've been in games development will be gone, forever. That's going to be a big moment."

Bach adds: "We wish we could show it to you, but we can't right now. I don't want you to take our word for it that we have it and others don't. One thing that's hard for us to know is what these other companies are doing, what's the actual solution, what's the code and so on. The thing we do see is this is natively built to do these things, not something you plug into an existing technology straight off and then all of a sudden your game can do things it couldn't do before. We're flipping that around completely -- this is the core of what games can use and then you can plug different things into it. So it's a different way of approaching the same problem.

"One thing that's also important is we're not trying to sell you anything, we're just talking about what we want to do. Ubisoft doesn't want players or other companies to buy this technology, we're just hinting at what the future will be like."

The proof of the pudding, as they say, is in the eating, so when will we see Scalar in action? Bach can't offer an exact date but offers us the rough ballpark of "as soon as possible."

In addition to Scalar, Ubisoft Stockholm is hard at work on a brand new IP that makes use of it, since new games are "the best way to show off any technology," according to Bach. It's perhaps surprising that Ubisoft isn't leaning on an established franchise to drive this new tech, but Romell emphasises the technology is being built around the game, not the other way around.

"This is the opportunity we have: a new studio, a new IP and the ability to start thinking about these things from scratch, rather than being stuck in doing a sequel or having to carry a lot of luggage [from previous entries]," he explains.

Bach adds: "We're not the creators of [other Ubisoft] IPs. You need to know what it means to bring an IP forward, and it would be disrespectful to just take an IP and squeeze it into a different technology. You should start from the idea and the goal of the IP, and then look at what technology is needed."

Regardless of the ways Scalar may enable new ideas and make development more efficient, there's no doubt the main takeaway for some people -- likely avid gamers -- will be that it allows bigger worlds. But with Ubisoft already synonymous with sprawling virtual landscapes in games that take hundreds of hours to complete, we have to ask: do game worlds really need to get bigger? The reveal of the full Far Cry 6 or Assassin's Creed Valhalla map is enough to turn some players (including yours truly) away.

"Do we need games to be bigger? No," reasons Bach. "Are some games going to benefit from being able to be bigger? Absolutely. It depends on that game, and the goal of that game and its creators.

"No part of a game should be driven by 'more is better.' This is technology, and that does not dictate what games you build, but there are games that will definitely benefit from being bigger, more detailed, being able to scale and being greater than they are today. I don't think there's a real connection between games being bigger and them being better or worse. It depends on the creators and how they want to spend their energy achieving their vision."