Rogue Legacy 2 Teddy and Kenny Lee Interview: ‘We’re Turning it Into a Metroidvania’

rogue legacy 2 interview

By August 18, 2020, we’ll all be able to play Rogue Legacy 2. But it won’t be the complete game, rather a chunk of it presented via an initial Early Access launch. As we’ve covered before, Rogue Legacy 2 will go through a couple stages before it’s a full game, although the team at Cellar Door Games already has a plan and framework in place. For Early Access, the team really wants to get organic feedback from the fans and make a better game with it. And with a sequel as big as Rogue Legacy 2, the community has quite a task ahead of it.

I got the chance to sit down and chat with Cellar Door Games co-founders Teddy and Kenny Lee about Rogue Legacy 2, and over the course of nearly an hour we covered the Early Access campaign, the fundamental changes for Rogue Legacy 2, the Trait system, and frankly much more than I expected. Buckle in for a big one, folks!

Lucas White, Siliconera: What led to getting started on Rogue Legacy 2? Was it just time for a sequel, or did you guys want to find a core idea first?

Teddy Lee (co-founder/game designer): We always wanted to make a sequel, but we wanted to do something in-between (Full Metal Furies). I’ve played a lot of sequels; I’m not a fan, but I always wanted to make a sequel, just to see. There are restrictions–when you’re thinking about it nonchalantly you don’t really take those restrictions to mind–but when we actually got down and dirty with it, there were so many things we had to take into account. At the very beginning I think I poured over 2,000 forum posts, just finding out what people did/didn’t like.

The whole point of Rogue Legacy was you didn’t have to grind, right? That was the original intent, but a lot of people felt like grinding was mandatory, a lot of people wanted to grind. So [making a sequel] was like taking in those things and finding a present-day solution, especially after making Full Metal Furies. We’re taking those ideas and trying to… not appease, but… it’s hard to explain. Take difficulty for example: Rogue Legacy is known for how hard it is, right? And people want that difficulty in Rogue Legacy 2–it’s part of the appeal. But if we just make it hard for people who were good at Rogue Legacy, then people who never played it are just gonna be blindsided.

When I was thinking about it, I was thinking about how you went from Dark Souls to Dark Souls II. Dark Souls 2 was one of the hardest games in the franchise; I had so much trouble with that one! But [FromSoftware] also made Bloodborne at the same time, and everybody was like, “That’s the Dark Souls successor.” It wasn’t as hard as Dark Souls II, but it was so similar to Dark Souls. But they made these minor changes to fundamental mechanics, which changes the cadence to things. So that was one of the things we wanted to do. How do we fundamentally change the base combat, without completely destroying its identity?

Speaking of combat, when I play Rogue Legacy I feel like I’m playing a Castlevania with a greatsword–I’m always molding my approach to a situation around that specific downward swinging arc. So when you talk about changing the fundamentals, how does that relate? I assume that’s partially what led to having different weapons?

Teddy: Yeah! So we took a lot from Full Metal Furies. That game has four characters who attack very differently. We took a lot of what we learned from that game and put it into Rogue Legacy. So with what you’re talking about with the greatsword thing… everything in Rogue Legacy 2 is a little faster. Enemies are faster, the player’s faster. It was important that even when we added new weapons like the bow and arrow, which really changes up how you play, we also had to change mobility. If you look at the clip we posted of a character using a bow, you can see how much faster you are in Rogue Legacy 2, and how much more mobility you have. You can see how it all ties together, how the enemies are more aggressive.

Kenny Lee (co-founder/programmer): For us, I think adding the weapons is a bigger challenge than people might think. It’s not as simple as, “Okay, we want this weapon; just stick it in there.” There’s lots of balancing to make sure these weapons are like, one isn’t clearly better than the other. It would be a waste of time otherwise, because people would just avoid [the lesser] weapon. In Rogue Legacy, we found that, you know, people have preferences. People have tendencies and we have to make sure that some of these tendencies aren’t so negative that they’ll never choose a class or weapon.

Teddy: One of the coolest things is, originally we were going to have the sword, and maybe another weapon like a bigger sword that had longer attack range and shorter delay. But then we decided, well, let’s use the Barbarian as an example… he’s available in Early Access. He had an axe, and originally that was going to be a bigger swing. But then we thought, “That bigger swing feels awful, so what if we turned it into a 360 spin when he jumps into the air and attacks for the entire duration?” Then the spin led to low knockback, so when you jump into an enemy it creates all this chip damage. That led to this chip damage system, where we have hit triggers for when you hit a guy every 2.5 seconds, so he became this super fast character that’s one of my favorite classes.

He’s a super chain damage character, but when he does his ground attack we made a new one to keep that hefty axe attack people would want. All we do there is revert back to the older Castlevania style where you’re movement locked. And in a game like Rogue Legacy, being movement locked is an insane detriment. So we were able to shorten the swing, but with the movement lock it feels completely different from the sword. When we came up with that for our first class, it set the standard for every weapon from then on. Every single weapon has so many nuances, which I’m really happy with.

I imagine the Early Access period is contributing to that balancing process, but what about before then? How do you personally feel when you’re tweaking a weapon to the point where it feels “good?”

Teddy: Our goal is to make everything feel “good.” Using the axe as an example again, we shaved and shaved away [at the animation properties], and even though it does obscene damage, it still feels good. When I say “good,” I’m talking about how mobile you still are, because mobility in Rogue Legacy is a big deal.

Kenny: The way Teddy has it set up, he’s got a test room, where he can test all the different classes and go back and forth. He tests out how fast they move, how mobile they are, how good they feel. Of course this is one of the advantages to Early Access. People will actually get to play it for us. If we’re off in some way, if it turns out to be too high of a learning curve for certain people, Early Access will be there to let us know. So that’s one of the things we’re looking forward to.

A couple years ago, before we started work on Rogue Legacy 2, I posted on Steam asking people how they felt about Early Access. That in no way was a sign we were doing Early Access; we were just getting our feet in the water. It was a really long thread with lots of responses, but even after all that we only decided to go with Early Access relatively recently. The game had already been in development for about two years; we did think about it a lot and didn’t take the decision lightly. I think in the long run it’ll be best for the game.

I noticed you guys had a lot more context than I was expecting on the Steam page. Was that part of finding a middle ground and get people more averse to Early Access on board?

Kenny: I understand the aversion people have to Early Access. I actually don’t buy Early Access games a lot of the time. If it’s big enough I have, but usually I’d rather wait for the finished product. So we figured the best way to help alleviate those concerns is to give as much information as possible. Hopefully it’s enough. We understand that it will turn people off, and it’s unfortunate that’s the reality of it. It was a risk I think was necessary, especially considering the scope of Rogue Legacy 2. It’s much larger than the first one, there’s lots of moving pieces. There’s so many places we can add content, but it’s difficult to just dump [lists everything]. It seems to make more sense as an interactive process where we add some classes, tweak them, then fully release them instead of having like 15 classes and ten of them aren’t good enough because we didn’t have time to balance them.

Teddy: The original pitch was like all the way up here [at this point Teddy, who is only showing the top of his head in our meeting, gestures into all the empty space above him] in terms of stuff we wanted to do, but it’s not all feasible. It’s really going to help us with scope and what needs to be in the game.

Can you give some examples of stuff from the community [mods and other user-generated content referred to as a motivator for Early Access] that really surprised you? Stuff you felt you never would have thought of?

Kenny: That’s a good question [laughs]

Teddy: The biggest one is definitely the Zors Mod; a guy took the entire game and added four new biomes and expanded the difficulty greatly. That was just mind-blowing, and I really… it’s tough to make a moddable game! We’re trying; we don’t know how moddable it will be in the end. But Zors Mod really opened my eyes, because there were a lot of things he did in that mod, which were things we had cut away because we thought we couldn’t do them. For example, he made the castle massive, right? The game couldn’t support that – we had all these ideas for a Mega Castle early on, but we said, “oh forget it; it’s not gonna work.” A lot of that stuff he did was inspiring for us, showing us what we could do. We’re doing it in a different way, but that was definitely a huge inspiration and insight into what people like.

Kenny: One of the advantages of a mod is that people understand it’s a mod. So it can be buggy, or maybe not as polished as a final product needs to be. We don’t really have that flexibility; people look at it with a magnifying glass. But Early Access gives us that same freedom. We can start to adopt some of the things in the mod, some of that flexibility and testing. So we’re excited to see what we can get away with in Early Access! [laughs]

rogue legacy 2 interview 5

With that in mind, since it’s been so long since Rogue Legacy came out, can you speak to how external changes in the games industry may have influenced the project? How does making Rogue Legacy 2 in 2020 differ from making Rogue Legacy in 2013?

Teddy: There’s definitely a lot more roguelikes [laughs]. There were only something like three [popular indie roguelikes] when we came out. I’ve definitely played them all save for some of the new ones, and it’s very tempting to say, “oh we should turn Rogue Legacy into an arena-based game where you put in [a number/seed] and start over.” But that just isn’t authentic to Rogue Legacy. The whole point was to let anyone of different skill levels beat the game. One person might take three hours, another person might take 40. That’s why we like RPG systems!

I like hard games, but I also like Dark Souls because if somebody wanted to, they could grind to level 100 and beat the boss in a couple of hits. So we wanted to keep that, but at the same time we wanted to modernize it a little. So while in Rogue Legacy our runs are [usually] around five minutes, we wanted to have the player have something new for the next five minutes. But one of the main complaints was that Rogue Legacy really didn’t change. Even though it’s new at first, by the end those five minutes feel the same. So we really revamped the relic system, which is a way of letting your character change during the run. You can’t just copy Binding of Isaac, because the whole point of Rogue Legacy is five-minute runs. Rogue Legacy also has so much permanent progression, compared to Dead Cells’ more lateral progression. So we can’t do that either and retain Rogue Legacy’s identity.

Kenny: One big difference is the number of different influences we have. We don’t say, “oh we’re definitely taking that from this game,” as it never fits one-to-one. But there are a lot of smaller influences that occur. For example Teddy was saying Rogue Legacy 2 is more mobile. That’s because people are more used to roguelikes now. They play Dead Cells or Exit the Gungeon which are faster, so Rogue Legacy has naturally become faster and more fluid. That’s what people’s expectations are now.

Teddy’s a very high-skilled gamer, and he always loves to make things faster, more fluid, and harder. But it takes people time to catch up to his skill level so we’re always worried we’re making our games too hard, too fast. So with Rogue Legacy 1 we went slower, but now that the genre’s more established we can start speeding things up.

To that point, I noticed in some of the materials the UI is a lot busier, at least in its current form. Looks like you can just do more with the controller. How does that intersect with the difficulty balancing?

Kenny: It’s tough. We’ve gone through many iterations of the UI. We want to add more content, but the more content or features you add means you have to explain more. For us, one of the tricks we do is we like to piecemeal information to people. For example, unlocking the Blacksmith. When you unlock him, that’s when the explanation comes in. You only need to worry about that functionality once you get to it. A bit of my concern is that when we release the game people will be like, “oh this is just like Rogue Legacy 1.” But that’s because they haven’t played enough to unlock stuff we added that expands the world and what you can do. We intentionally block it off so people aren’t overwhelmed.

Teddy: There are a lot more mechanics coming in for Rogue Legacy 2, and we have different paths for introducing the information. One of the paths is definitely going to take a lot longer; you just have to dig deeper into the game to get it.

Kenny: I think we also learned some lessons from FMF. That combat system was fairly complex, and I think with our tutorial we didn’t expose people to it enough.

Teddy: That one was tough because we had to make a multiplayer tutorial, and that was not something we had experience with.

When you guys started working on the game, were there any specific pain points or negative feedbacks you really wanted to hop on addressing?

Teddy: So a lot of people say the game is “floaty,” which drives me crazy [laughs]. They’re like, “oh it’s so floaty compared to Celeste or Dead Cells.” A huge part of that, is their jump height is so small. They use mostly realistic physics. We can’t do that, because we have a lot of projectiles coming in. So we have to give you full air control. So that was one of the things I wanted to tackle. We tackled it in terms of combat; we can’t tackle it in terms of people who want that lower jump. We’ve done a lot with the combat, and actually it’s in the iOS version of [Rogue Legacy 1]. We took what we made for Rogue Legacy 2 and were like, “hey the iOS version’s coming!” It’s called the Attack Intent System, and we just shuffled that into Rogue Legacy 1 to make it feel better.

Kenny: There were a few, minor creative things that people had beef with. Traits sort of devolved into a gimmick. We added some systems in to make Traits worth picking, making a more meaningful choice instead of, “oh you’re just stuck with this trait.” You’ve got Vertigo, you can’t play the game; it’s like suicide. We got rid of those scenarios. So there’s some very low-hanging fruit that we wanted to address, and I think we did. It’s actually funny; since we announced it there’s been a lot of people like, “I hope they fix this or I hope they fix that,” and we’re just sitting there smiling because we fixed that. [laughs] That makes us feel good; I hope we crossed all the boxes that people had issues with.

Teddy: Like I said we looked at forum posts, and were like, “what are the most common beefs, and what are the beefs I knew about the game already?” So those were the first things we tackled; we rebuilt the mana system as well. That was a pain point in how much trouble that’s been. I’m gonna spoil one of the traits, too. It’s one of my favorite traits. I can’t remember what it was called, but it makes a character who dies in one hit. You have 1 HP no matter what you do. But we added a thing called the Universal Health Care System. You get bonuses based on how detrimental your trait is, and I think that one is tuned at 300% gold gain. You just kill like five dudes and you’ve already made 600 bucks! It’s a lot of fun because even if you’re a pro you’re playing that first castle area super, super safe. We want to make every single run different, but rewarding.

Speaking of traits, we’ve been gameplay heavy so far. But there was some critical discourse when Rogue Legacy was new surrounding using things like disabilities as mechanics. Has that impacted your approach at all?

Teddy: It’s definitely on our minds, especially in today’s climate. But one thing we keep reminding ourselves that’s super important, is that when we made Rogue Legacy, the amount of fan mail we got from people with these disabilities was huge. A lot of people really appreciated it.

Kenny: We didn’t get hate mail, they were polite about it but we got people saying things like, “this is not appropriate.” So that definitely is part of it. But Rogue Legacy was supposed to be inclusive. The tagline was, “anyone can be a hero.” It wasn’t supposed to be, “oh look at this guy, what a joke!” That was never the case. But people who had these afflictions, who have to deal with them daily, were like, “thank you for representing us.” So we have to keep it. We understand it will rub people the wrong way, but it’s more important for the people who are getting that kind of enjoyment out of it.

Teddy: In total we got two emails from two people who had two afflictions who were upset about how we represented [them]. One was Ehlers-Danlos syndrome which isn’t in the game anymore, and the other one was Tourettes which was turned into Coprolalia. So when someone came to us and said they were really offended we changed it. But in regards to that, when we changed EDS to Flexible, we got an email from someone who said his girlfriend has EDS, and was wondering where it was because he’d always pick it when she watched him play. That made us feel really bad. So we don’t think we’re muting ourselves or whatever; we’re trying to do something different because we understand it can be hurtful for some people. So we want to have it available for people who want to see it, but people who don’t can change it.

Kenny: There was one problem. It wasn’t intentional, but people noticed that generally Traits were disabilities, and they were negative. That was never intentional, but it was what we came up with when we were thinking it all up in our heads unfortunately. So with Rogue Legacy 2, there are more… if you look at the trailer you’ll notice one of them is “Pure Diva,” which has nothing to do with a disability. Because [focusing on disabilities] was never what we wanted; we just wanted to have general “Traits” that weren’t supposed to have a specific identity to them. So we do still have some scientific Traits, some of them are disabilities. But we have a lot more general ones, plus this idea of negative/positive are much more vague now. Hopefully people won’t feel that way as much with Rogue Legacy 2 because we spend more time broadening the idea of what a Trait is.

Teddy: Also in Rogue Legacy 1 in general it was like, 80/20 in terms of negative to positive Traits for gameplay. In this game we have a lot more straight up positive benefits, less negative, and a lot more ephemeral sorts of Traits.

Yeah, it’s weird. I was recently diagnosed with adult ADHD, and going back to Rogue Legacy to prepare for this interview, just seeing those letters in the game, despite already knowing that was in there, stood out to me more strongly. So I can see why it’s such a delicate balance either way.

Kenny: Yeah, it’s more recognizable.

Teddy: That’s why we’re putting a lot of thought into it. We actually have something in place, but I don’t know if it’s going to be in the first release.

Kenny: That’s part of the challenge with Rogue Legacy 2. Do we just add all of the Traits from Rogue Legacy 1? But what does that mean; we don’t want to repeat ourselves to where Rogue Legacy 2 feels stale. So that’s why we’re adding new stuff, but at the same time what if we remove one? We don’t know so we’ll have to see.

Because Rogue Legacy 2 is a 2.5D game, some things we just can’t do anymore. For example there was Stereo Blindness that would do like a Paper Mario effect. With the way we set up Rogue Legacy 2, it’s a lot harder to Paper Mario a 3D object. So maybe that one should just go. So hopefully that won’t be too much of a problem for people.

Can you talk about the creative and development side of what motivated the art style change?

Kenny: [laughter in an oh boy here we go sort of way] That’s a long discussion. The thing is, historically we’ve only made (commercial) pixel games. The response to it is really up and down. People either like pixel art or they hate it because they think it’s cheap or a cheap way of making video games. It’s not true by the way; pixel art is not easy. So when we started Rogue Legacy 2 we knew we wanted to go to Unity. Unity offered us a lot more in terms of what we were capable of doing. For example with the previous games I was the engine programmer, and I didn’t know any 3D programming so they had to be 2D anyway. Unity brought in the 3D opportunities, so we were like, “where can we take this?” That naturally led to a non-2D style.

For technical reasons why we went to 3D, it’s much easier to do things like attach different parts of equipment and weapons, whereas with pixel art or hand-drawn art you have to manually redraw for each piece of equipment. We knew that we wanted Rogue Legacy 2 to be more customizable, so 3D felt like a natural progression. And as far as how we got to this particular art style, well that’s up to our art director.

Teddy: I just wanted the characters to be cute!

So what kind of impact does that have when figuring out things like hitboxes or hurt boxes?

Teddy: A ton!

Kenny: Huge ramifications! Because this is in 3D we can’t do a lot of the things we did in 2D. In Rogue Legacy 1 the hitboxes were sort of integrated into the metadata of the sprite sheets. There’s no more sprite sheets in Rogue Legacy 2! So we had to come up with a completely new system for doing hitboxes. The other big issue was the animation system. In 3D everything is sort of automatically interpolated, so you put a position, and then the guy just sort of tweens to that position. That’s how a lot of animation works in 3D.

We didn’t want that animation system. We wanted the chunky, nice-feeling, hand-crafted 2D animations that you see in like Spider-Man: Enter the Spider-Verse, or games like Dragon Ball FighterZ where each frame is actually steps, so it’s not like a smooth animation. It still keeps that 2D chunkiness. So we had to revamp the system to support that. It was a lot of work! I think it was successful; a lot of people look at it and don’t even realize it’s a 3D game because the animations are stepped, it’s the cel-shading and outlines. It looks 2D. Whether that’s a positive or negative seems to be up and down on the forums and stuff, but I think we achieved what we aimed for.

Teddy: One thing that always irked me in Rogue Legacy 1 was when we made hitboxes, they were super generous. You see your character, and you see his hitbox and it’s like half his size. So the bullet has to go really deep into your chest before it will actually hit.

Like a shmup!

Teddy: Yeah, identical to a shmup. But because of a bug we had, or rather not a bug but a problem we had when we made hitboxes, your feet were pixel perfect. That drove me crazy; people would complain because when you jumped onto a bullet you’d get hit the moment the pixels touched. We couldn’t fix that because of spikes on the floor, so we couldn’t create different hitboxes. But we were able to fix that in Rogue Legacy 2, so it’s way more generous. I’m super happy about that.

rogue legacy 2 interview 3

One more question! Is there a new feature or change for Rogue Legacy 2 you’re particularly excited about, that you hope gets attention in coverage and whatnot?

Teddy: Well, we’re turning it into a Metroidvania. That’s actually what Rogue Legacy 1 was originally gonna be. So you’re gonna get abilities, the game does change as you get further in. That’s gonna have consequences to how we do NG+ which is gonna be really cool. But it isn’t just, “we’re doing a Metroidvania” because who cares, there’s so many Metroidvanias. Rogue Legacy 2 is a Metroidvania in a world that’s still procedural. I have to backtrack in a world that’s changing, so where does the logic for anything come in? So we have to do all that, and I’m excited for that! It’s already kind of working; as we flesh the world out we’ll see how it scales. The only thing I’m afraid of is right now it feels very natural. So I’m hoping people don’t gloss over it. It’s hard to do!

Kenny: I’m most interested in the biome generation system. We spent a lot of time developing a comprehensive system for generating different-feeling biomes. So it’s no longer biome 1 and biome 2 where biome 2 is just a harder version with a palette swap. There are significant changes to the worlds, and I’m really hoping that shines through for people. It is challenging, because every single world has to be not only unique to play, but randomly generated. And connected! So we’re trying to be as creative as we can with handcuffs on because the worlds have to follow a lot of specific rules in order to connect properly with one another. I’m pretty excited about that and I hope it pays off. I hope people really do feel like, for example, “the tower feels so much different from the castle now, and it’s fun to play and it’s fresh.” I don’t know if we succeeded or not but we spent an incredible amount of time trying to get the biomes to generate in a unique manner.

Teddy: I’m afraid it might be a God of War thing; you know how they did all that work for that one camera? And then nobody notices [laughs]. A lot of the trouble we’re running into is that we’re still building a single world. So you can walk, and if you open up your map you can see the entire biome all at once. Most people would be like you just enter a door and you’re in a new biome. It’s a trillion times easier to just teleport you into a new biome. But I’m just like, “no we gotta do this!” I hope we’re not shooting ourselves in the foot. You know what I mean? We’ll see.

Rogue Legacy 2 enters its Early Access period on July 23, 2020.

Lucas White
Lucas writes about video games a lot. Sometimes he plays them. Every now and then he enjoys one. To get on his good side, say nice things about Dragon Quest and Musou. Never mention the Devil May Cry reboot in his presence. Backed Bloodstained on Kickstarter but all his opinions on it are correct regardless.