Which Dead Cells is the real Dead Cells? Is it the Dead Cells we here at Motion Twin released on Steam Early Access back in May 2017, a rough sketch of the game to come? Is it Dead Cells as it existed on August 6, 2018, when we celebrated our 1.0 release? Is it the Dead Cells that exists today, more than a year after that so-called “finished” version?
Of course, as artists we like to think the latest version is always the greatest. The truth is, though, that if you asked a dozen of our players, you’d get a dozen different answers — and they’re all correct. When you work on a game that evolves as Dead Cells has, whether through Early Access or post-release, it never has a definitive version. Not really.
So do we have a responsibility to preserve our past, for the benefit of us and others? We think so. It’s taken two years, but we recently released the Legacy Update on Steam, an archive of major Dead Cells iterations from 2017 to the present.
They say you can’t make everyone happy, but it turns out you can — with a little planning. Read on to find out how, and more important, why you should set up your own archive.
Nobody left behind
Early Access was important to the development of Dead Cells. We can’t stress that enough. Player feedback informed many of our decisions, both big and small, and Dead Cells would be different (and probably worse) if we hadn’t collaborated with our fans.
Feedback isn’t free, though. Developing a game in public entails giving up a bit of ownership. Not all of it, of course. After all, Dead Cells is our game, and we’re going to do what we think is best. But people are buying into not just the game that exists but the game they think will exist a year, two years, even five years down the line.
Problems arise when those visions clash, when our decisions not only don’t match player desires but also “ruin” the game for certain players. What do we say to someone who’s put hundreds of hours into Dead Cells, only to find they no longer own the game they wanted?
Traditionally, you’d say “Tough luck.” There is no Death of the Author in video games, or at least not so long as they’re still being updated. You make decisions about the game’s direction, you accept that you’re going to lose some people along the way and gain others. If the latter outweighs the former, then congratulations, you’re a successful indie developer. Pop the champagne.
An archive means nobody needs to be left behind, though. If you think we lost the thread after version 0.7? Now you can play version 0.7 to your heart’s content. We haven’t taken away the game you wanted, bought, and enjoyed for maybe hundreds of hours. It exists, your perfect Dead Cells experience.
And we’ve focused mostly on Early Access, but really, these lessons extend to the whole Games-as-a-Service era. There, too, you have a divide between “The Game People Paid For” and “The Game That Exists,” one that only grows wider the longer you support a project, the more you add or tweak or disable.
In fact, one of our most contentious changes came after we left Early Access and released the 1.1 update, now colloquially known as the Pimp My Run update because it added a Custom Game mode to Dead Cells. That was the big tentpole feature for 1.1, but there was more to it. After much internal debate we decided to do a massive balance pass, effectively dropping the default difficulty to make Dead Cells more approachable to newcomers.
People noticed — or at least, a hardcore subset of fans noticed. Was it the right decision? We think so. Others disagreed, and they did so quite loudly. With the Legacy Update, that more difficult 1.0 experience is readily available for those who want it.
A weight off our minds
And that leads into our second point: Keeping an archive is good for you, the developer, as well.
In the past, we worried about making sweeping changes to Dead Cells. That’s not to say we didn’t, as anyone upset about balance passes or legendary drop rates can tell you. Major updates were stressful though, especially ones we thought might fundamentally alter the game or prove contentious. Should we make the “right” choice for Dead Cells, even if in doing so we made a lot of our players angry? (See: The hornet’s nest we kicked over by locking achievements for certain custom mode setups.)
The Legacy Update doesn’t completely remove that stress, but it does diminish it. Players are no longer shackled to our vision, which in turn means we can experiment with elements of Dead Cells that before seemed untouchable—or at least volatile. There’s a creative freedom that comes from knowing any change we make can be easily reverted by naysayers.
Whether it will actually have an effect on day-to-day development? It’s hard to say. We like to think we’re pretty good stewards, and that we’d make the tough calls even without the Legacy Update. Psychologically, it’s an interesting change though.
We’ve also found the archive a useful resource, as we get further and further into development on Dead Cells.
There’s an old adage, Chesterton’s Fence, that essentially states that if you come across a fence you should figure out why it’s there before tearing it down. The same goes for game development. It helps to know why you made certain decisions before changing them again, and the archive’s allowed us to check our own work, trace a line between past and present — even settle disputes, on a few occasions.
OK, so you’re convinced. You need an archive. Now how do you do it? It’s easy, albeit a bit of an unofficial hack. You’re probably familiar with Steam’s beta branches, a feature theoretically meant for internal testing or maybe for players to access an upcoming update early. At Motion Twin, we’d already been using branches for our alpha and beta test pools.
For the Legacy Update we simply created 16 different branches, one for each major version of Dead Cells.
Again, easy — though not exactly elegant. We achieved our goal, allowing players to access archival builds of Dead Cells on demand. The process isn’t very intuitive though, hidden in a Steam sub-menu of a sub-menu that only dedicated players will know to seek out.
And there are limits, as we learned the hard way. Valve caps beta branches at 20 per game — a perfectly reasonable number, unless for some reason you’ve decided to (hypothetically) use branches to archive your game’s history. Oops.
We’re safe for now, having limited ourselves to 16 versions of Dead Cells plus the alpha and beta builds, but a few months from now we might have a problem. Ideally we’d like to see more official support from Valve, both a more intuitive front-end for the player and a more robust back-end for us. For instance, branches don’t allow us to see which versions of Dead Cells are most popular, which is data we’d love to access.
One eye on the future
Saying “It’s easy, just put it on Steam” is a bit of a misdirect though. The hardest part of building an archive is recognizing that you need to build one, and then finding the time to do it.
We only sort-of planned for an eventual Legacy Update. The idea was floated two years ago when we were still in Early Access, but Motion Twin’s a small studio and nobody had the bandwidth to do it properly. When it came time to actually do the work, we knew we had maybe a 50/50 shot at success.
And we got lucky. We’re the first to admit it. It took us about a week to roll back Git to the proper points, compile our builds, and fix any missing asset errors or critical bugs, then upload them all to Steam. Not too terrible, all things considered.
It could’ve been easier though. Now that we’ve released the Legacy Update and know this is something we want to keep doing in the future, we can plan for it. It’s part of our workflow. We actively archive every build, and it should be simple to choose the definitive 1.6 branch in the near future and add it to the list before moving to Dead Cells 1.7.
Ideally you’d build your entire game like this, with an eye towards future archiving. Do the work early, and you’ll avoid having to clean up a mess later. Of course, that’s easy to say in retrospect with the Legacy Update safely rolled out. It’s much harder when you’re a small team and everything’s on fire.
It’s worth it though, both for you and your players. Hopefully this article’s given you something to think about, especially those of you working on Early Access projects. It’s (probably) not too late to start archiving!