Game Guides - -

An interview with Cellar Door Games, makers of Rogue Legacy 2

Cellar Door Games is the developer behind the Rogue Legacy series and Try Hard Guides was able to get an interview with the co-founders.
Featured Rogue Legacy 2 Promo Image
Image: Cellar Door Games

The first Rogue Legacy launched for PC in 2013 and became a hit due to its unique spin on the Rogue-like genre. Players take on the reigns of a legacy of potential heroes, all attempting to discover the castle’s mysteries. Each new attempt at the fortress would be a new descendant of the previous heroes, complete with their unique combination of traits and skills that can make the job easier or harder.

Rogue Legacy 2 is launching in 1.0 on April 28 after two years of early access development. As a new family, the player will take on the mysteries of the Kingdom. Whether or not the player can make it that far will depend on their descendants and the level of investment.

Fortunately, Try Hard Guides was able to get an interview with Cellar Door Games‘ co-owners, Teddy and Kenny Lee, where they discussed some of the finer details of Rogue Legacy 2.

Rogue Legacy 2 Boss Eye Image
Image: Cellar Door Games

What interests you the most about creating games?

Teddy Lee: The thing I enjoy most about making games is definitely the problem-solving. Using tools from feature X, to solve problem Y is super fun and will never get old for me. It’s like I’m playing one of those old-school adventure games, like Monkey Island or King’s Quest, except it’s real life!

How did your team change between Rogue Legacy and Rogue Legacy 2?

Kenny Lee: Honestly, the team didn’t change much.  After Rogue Legacy, most of the original team went on to do bigger and better things, so it was really awesome that we were able to bring everyone back to work on the sequel.  I think that’s a big reason why RL2 is such a step up.  After separating everyone gained a lot of experience, and upon their return they brought that knowledge back with them.

Of course, we also grew larger.  We brought in an art director, another programmer, a technical artist, and a slew of very talented independent contractors, many of whom we’ve worked with in the past.  But even with that growth, I think a lot of people would be surprised at the size of our team.  All-in-all, there were only six of us working full-time, and another handful working part-time.  I truly believe everyone punched well above their weight to get RL2 to where it is today.

What led to the decision to put Rogue Legacy 2 in Early Access as opposed to a standard 1.0 release?

Kenny: We actually jumped back and forth on that decision nearly until the day we announced the Early Access release date.  On the one hand, it seemed like Rogue Legacy 2 would be perfect for Early Access.  We thrive on iterative design, and the genre has proven it can do well in EA with games like Hades, Dead Cells, and Risk of Rain 2.  On the other hand, it was scary to jump into something we knew nothing about.  For us, Rogue Legacy is more than just a game, it changed our lives, so to completely open it up to the community was a very difficult and personal decision to make.  Ultimately, we felt Early Access was necessary to make the best game we could, and it’s a relief to know it was the right choice.

What mechanics or features from the first game were the most important to improve?

Teddy: We went into Rogue Legacy 2 with a lot of the foundational changes already in mind.

  1. Make Traits more important by giving the ones that affect gameplay a gold bonus. We had already jury-rigged this into later versions of RL1, but it only applied to Vertigo because it was too late to add it to them all. We knew that once we applied this feature it would allow us to think up even wilder traits.
  2. Add more variety to each run, both immediately and throughout the play session. Creating more distinct classes added diversity to the start, but the harder problem was getting runs to change as you played them.  It took a few iterations, but we addressed this with the Relic and Resolve system
  3. Try to add true Metroidvania elements to a randomly changing world.  We’re pretty adamant that every boss can be beaten without taking damage, and with the weakest possible character. In RL1, because abilities like double-jump and dash were optional, boss encounters were handicapped in how interesting they could be. Heirlooms in RL2 play double-duty by making the game feel more like an authentic metroidvania, and allowing us to increase the complexity of boss battle sequences.
  4. Wrap the world’s events in a proper story. I really wanted to tackle the problem of trying to make a plot-driven story in a game where the world is (literally) constantly changing. Most rogue-likes resolve this by either having no story, or by front and back loading it (story at the start and story at the end but nothing in the middle). It required a lot of up-front planning, but it was really fun to write this intricate lockbox mystery.
Rogue Legacy 2 Citadel Image
Image: Cellar Door Games

Did any other games or franchises serve as inspiration for the characters? What about the Estuaries?

Teddy: All of the names and places in Rogue Legacy 2 are allusions to other things, but aside from maybe sub-conscious influences, I didn’t use any pre-existing characters as the basis for the Estuaries. It’s probably a big reason why the writing took so long.

Are there any limitations on what combinations of traits a descendant can have?

Kenny: Absolutely, there are.  For example, you can’t have both Gigantism and Dwarfism, because then you’d just be a normal-sized dude.  This is what makes traits so tricky to implement.  With each new one you design, you need to consider how it would interact with all the others.  For the most part, traits work well enough, but there are a handful that really gave us a tough time (I’m looking at you, Pacifist).

Do the traits you pick for each character have an effect on the traits that will show up in your descendants?

Kenny: No, there is no weighted disposition for any of your future heirs, and we decided to do that for several reasons.  The first one being it wouldn’t be much fun.  The primary purpose to traits is to make your next run different from the last, so giving you the same traits goes opposite to that.  Secondly, we really wanted to dispel the notion that these are strictly genetic traits.  There are personality traits, like being a Diva, and life choices, such as being Vegan, that have nothing to do with genetics, or disabilities, or any physical characteristics.  It’s just meant to be people from all walks of life.

What are your favorite class and trait combinations?

Teddy: I like all the classes, but to avoid giving a cop-out answer, I’d probably choose the Ranger. The Ivy Canopy talent is absolutely wild, and once you master it you can do really creative things. But I also really like the Boxer and the Duelist.

My favourite trait is easily Contrarian, which randomizes your character’s weapon and talent. But if you want an OP class + trait combo, it’s hard to get better than a Bard with Clownanthropy.

Rogue Legacy 2 Clouds Image
Image: Cellar Door Games

Which monster posed the biggest design challenge outside of bosses?

Teddy: It’s hard to say. Almost all the enemies in the game were penned out during pre-production, and then we just chipped away at them throughout development.

For technical implementation, I’d probably say the Zombies, Wall Knuts, and Mimics gave us the most trouble. Each of them have special mechanics that introduced unique challenges.

How did CDG decide on the different layouts in each area of the Kingdom?

Teddy: Just through a lot of iteration. We paper designed the rule sets for all of the biomes and then built them up from there. We try not to treat our design docs like gospel, so if something isn’t working, or if something unplanned IS working, then we’ll pivot very quickly.

Out of all the biomes, Axis Mundi had the most changes by far. The original design was supposed to be a bridge with an insanely dense town built on top of it (inspired by Kowloon Walled City). Art-wise it was proving to be too expensive, so then we turned it into just a bridge. That was really boring, so we prototyped a sewer system below the bridge, so you had this dual-route thing going on. That turned out even more expensive (and bug-ridden) than our original design, so we came back to the bridge with buildings on top of it, just not as dense.

Are you anticipating more speed runs from players once you release into 1.0?

Teddy: Speed runs would definitely be fun to watch. 😊

Will Rogue Legacy 2 be getting more attention outside of bug fixes in the future or is CDG taking some time before the next project?

Kenny: We’ll be taking some time off, but not necessarily to recover before jumping onto a new game.  Surprisingly, despite working on RL2 for nearly four years, there are still some ideas that we’ve been talking about.  There’s no promises we’ll be adding any extra content to the game after version 1.0, but the desire is there.


More Content


Leave a Comment

All comments go through a moderation process, and should be approved in a timely manner. To see why your comment might not have been approved, check out our Comment Rules page!

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.