Zeno Clash Dev On Abyss Odyssey’s Procedural Insanity

You certainly can’t knock Zeno Clash and Rock of Ages developer ACE Team for a lack of ambition. Their previous titles’ massive scope and rampant, otherworldly weirdness demonstrated that in spades. So how do you follow that? Why, by jutting off in an entirely different direction, of course. Abyss Odyssey is a procedurally generated, Street Fighter and Super Smash Bros-inspired roguelike-like (but not quite) um… thing. ACE Team’s first set of words about it didn’t quite do it justice, which is why we’ve given them a couple thousand more to work with. Go below to get a better sense of how Abyss Odyssey will work, which utterly mad art style inspired this one (the grand ACE Team tradition), why they’re not doing Steam Early Access, and whether or not they think procedural generation will hurt their knack for insane world-building. 

RPS: Is Abyss Odyssey in any way referencing Oddworld: Abe’s Odyssey? I know this is probably a dumb question, but I thought of it the second I saw your title.

Carlos Bordeu, founder/designer: Not at all. There are other titles like Atlus’s Etrian Odyssey that also share the word Odyssey, but our game is completely unrelated. We hope no one decides to trademark “Odyssey” now.

The truth is we got almost no credit for making a game like Zeno Clash 2 – something that was easily of the scale of smaller triple-A titles.

RPS: Real questions now: ACE is known for its hyper-imaginative art design, and I could see procedural generation getting in the way of that – especially in terms of world coherence and the general “feel” of a place. How are you avoiding that issue?

Bordeu: That was one of the main concerns of Atlus when we proposed the title and it was one of the first challenges we had with the game when prototyping it. We had to make a game that didn’t evidently look procedural while still being modularly generated for the level design. Art development has always been a strength of this studio so we needed to make sure we could create a system that was robust enough to allow us to develop something that looked great and not super ’tiled’. At this point we’re very happy with the art of the game, and most people who have played our builds are immediately impressed by the graphics and art, so I don’t think it is a concern for the title anymore.

RPS: On the flipside, do you think procedural generation provides interesting/new opportunities for you from an art standpoint? Ones that other developers might not think of?

Bordeu: I think the challenge of making a platformer that is procedural and make it look as organic as we have was a very interesting challenge itself. Procedurally generated platformers usually use 2D art, or a very ‘retro’ theme where you can get away with just repeating some elements (like blocks or ’tiles’), and that is fine – but we wanted to do something where this wasn’t evident.

The main problem you need to solve here is consistency, because the more “realistic” or complex your art design is, the less you can get away with just repeating the same elements over and over again. You basically have to create something where elements are modular and can connect with other elements – but it has to be in a more logical way. It can’t look in the end like a bunch of highly detailed lego pieces sticking together, because if the repeating patterns are evident it will end up looking terrible.

RPS: You guys always have very interesting starting points for your games. Rock of Ages was inspired by a lot of classic art, etc. Where did Abyss Odyssey come from? What’s the story of its inception? Its first moments as an idea, etc?

Bordeu: I think we’re going to eventually go into a lengthy blog post about the origins for Abyss Odyssey in upcoming content releases. For a shorter response I can say that we went through a lot of experiments to get to the final look, and our first test dates back to when we were working on Rock of Ages. We attempted some new art styles that didn’t work out at first. What finally ended up making us take a particular direction was referencing the Art Nouveau style. We started by taking references from Harry Clarke – an Irish stained-glass artist and book illustrator who made several illustrations for the stories of Edgar Allan Poe.

At first the game was being designed to be more creepy, but we felt that many other references of Art Nouveau were not so sinister and that it might not be the best for the whole design to be like this, so during the whole development process the concept has matured to what it finally is.

RPS: ACE Team is known for venturing into some pretty strange territory. Is Abyss Odyssey going to have any of that overt weirdness? How off-kilter are you aiming for it to be?

Bordeu: This game doesn’t get into the level of weirdness that Zeno Clash does, but mostly because Zeno Clash is on the extreme end of weirdness, so anything past that point would be unintelligible [laughs]. Abyss Odyssey does have some surreal and bizarre references – it is definitely an unconventional title – and yes, there are a some “what the hell!?” moments that are likely to catch people off guard. But again – we had said we weren’t doing something as weird as Zeno Clash for the next project, which I don’t think is a bad thing.

An interesting note, though, is that there is a bunch of monsters and characters inspired by Chilean folklore. The “Camahueto” is a myth about a bull with one horn that “erupts from the earth with such force that it leaves a tremendous hole and drags down everything in its path in a rapid race to reach the sea.”

RPS: How much environmental variety are you hoping for? How many different sorts of levels and configurations?

Bordeu: The variety is not only about the levels themselves. A really big part of the variety comes from the enemies and elements within the Abyss itself. In terms of how many enemies we have, and how complex they are, I think this game beats Zeno Clash. Having been one of the two animators to work on all the hundreds of moves for the creatures I can say without a doubt that this has been our most intensive work to date with regards to combat design.

The game will feature different themes, where the Abyss changes in style and you get to meet all new monsters, as well as bosses and mini-bosses. But the structure isn’t so linear and some stuff that you get to see will be different over time or for different players.

RPS: Procedurally generated levels often have trouble matching the precision and memorability of handcrafted worlds. Are you worried that Abyss Odyssey might not stand up to, say, Zeno Clash on that front?

Bordeu: Not all games have the same objective and the type of discovery you get in a game like Zeno Clash is very different to a game like Abyss Odyssey. For starters, Zeno Clash is a first person adventure, with a pretty set-in-stone structure. Once you’ve seen it, you’re done. Abyss Odyssey isn’t trying to provide a similar experience to Zeno Clash. I’m pretty certain that one of the things that will make people excited about the game is seeing friends who have found stuff they have not seen, captured an enemy soul that they thought was impossible to obtain… things like that.

Zeno Clash was a more story-focused adventure with melee combat, but I think most people played it to see the art design rather than master the combat mechanics. Abyss has a very strong emphasis on other aspects, and combat is a key element of the game. The game is easy to play, but the depth of the combat mechanics goes way beyond traditional 2D action titles and more into the realm of fighting games.

That on itself is a huge part of this title, and I’m pretty sure people who might have ignored our previous games because they didn’t care for their art design will be interested in this because of the core gameplay mechanics. Which isn’t to say that we feel that aesthetics in this game are unimportant. We think Abyss Odyssey looks great when compared to other platforming or roguelike games.

RPS: In what sense(s) will the game be roguelike-like?

Bordeu: The abyss is a recurring event/environment and we will explain this much more in detail in our second video, but for now I can say that this isn’t a game with a traditional structure. The game is designed with replayability in mind, but we’re not even sure if we should call it roguelike since the death mechanics and some progression details differ quite a bit from traditional roguelike games. We will share more interesting details of other unique elements that are related to this concept soon. This will be featured in our next video release.

RPS: Are you worried that people are tiring of every game having roguelike mechanics? It’s become quite the trend.

Bordeu: Abyss Odyssey is very unlike other roguelike games. Again, we’re not even sure if we should call it roguelike. By itself the elaborate combat mechanics and combat design means we have a fighting game mixed with an action platformer and I’m pretty sure some people are going to care more about this aspect of the game than anything else.

We have a training room that you can access where you can choose the ‘fighters’ (monsters) you unlock, and it is pretty much like a traditional fighting game’s training room. I’ve sometimes found myself spending a lot of time practicing combos and juggles within the training room with all the different creatures instead of just going into the main adventure mode. In co-op, just taking a pause and facing off your partner is a blast!

RPS: What’s the story of the game’s world? Will we uncover it more as play and – presumably, due to roguelike elements – fail?

Bordeu: The story does progress and get unveiled more as you replay the game. The game is actually set in Chile and the basic premise is about a mysterious abyss that opens in the surface of the city of Santiago and you play the part of one of three heroes – each with their own backstory – that have to deal with the creatures coming from inside. The heroes aren’t just a plain character classes like warrior, mage, etc. They have their own identities and where the abyss comes from and who is at the bottom is something you get to learn when playing the game.

Abyss Odyssey isn’t a super complicated story that no one will understand like maybe Zeno Clash (hahaha), but it has more depth than many platformers where you basically just have to rescue someone or defeat an evil being.

RPS: How difficult are you aiming for Abyss Odyssey to be?

Bordeu: We’re still in a heavy phase of balancing, since we need to hit a good balance between challenge and progression. We need this to feel like something you can get into and play easily, but that you are compelled to master.

Currently we have a sort of global AI variable which we can raise or decrease to make the AI more aggressive, and when pushing it way up I’ve seen just a couple of basic enemies destroy a capable player. That’s how powerful our combat engine allows our regular NPCs to be. This is something unique to the game because you can’t really imagine two basic skeleton enemies beating the player in a Castlevania-like game. In Abyss Odyssey they can.

But if we don’t balance this properly the enemies will beat you pretty bad. So we still have to hit that sweet spot where we feel that a player who is mastering the fighting game has a sufficient challenge as he moves forward, and that the player who is going into the game for the first time doesn’t feel frustrated.

RPS: How much of your knowledge from Zeno Clash’s combat is going into the combo-based system of Abyss Odyssey?

Bordeu: None. This plays a bit like Street Fighter meets Super Smash Bros. Just learning the combat mechanics in Zeno Clash was hard. Here it is easy. But being good at playing Abyss Odyssey is done by mastering the gameplay. We studied fighting games like these ones extensively so we could produce our own fighting mechanics. We had to start from scratch to design the new combat engine.

RPS: You can turn into various enemies you encounter. How does that system work? Can I level up creature forms and stuff?

Bordeu: The player starts off with a base skill level and he can capture enemies that are of his skill level, or lower. As you progress and raise your own level, you will be able to get better monsters. The real leveling up for the monsters is getting better ones, as you can swap between them while you are progressing, so it wouldn’t make all that much sense to level up a monster if maybe soon I’ll capture a better one. Basically, you’re shifting into new characters, so you’re keeping your level/stats.

RPS: You mentioned co-op multiplayer. Will there be any other kinds? Competitive? Something else that’s more unique?

Bordeu: Yes, we have local and online coop. The game begs for a competitive fighting mode, and it is something we’re considering to do in the horizon, but it is a feature outside of the time and scope for the initial launch. ACE Team has a very good history of updating our titles with new game modes and content and this will be no different.

RPS: The game will be on Steam, but are you going to do Early Access or anything like that? Roguelikes especially have found tremendous success on Early Access in the past.

Bordeu: No, this isn’t an early access game, but something it will definitely share with early access titles is the level of support we’ll include based off the community feedback that will translate into relevant new content and updates like the one suggested in my previous answer.

RPS: Are you looking into mod support or anything like that on the PC side?

Bordeu: It is still too early to decide on mod support or things like that. But we haven’t discarded it.

RPS: When are you hoping to release it?

Bordeu: Abyss Odyssey will be available in summer 2014 as a downloadable arcade title for Xbox 360, PlayStation 3, and PC. We’re actually late into starting the promotional efforts for the game – we should have done this a while ago – so the game is deeper into development than other titles we’ve made before by the time they were announced.

RPS: Is this ACE’s main project right now? Are you working on anything else alongside it?

Bordeu: We’re always working on more than one thing, but this is the main project of the moment. I think some people have been misguided by our previous games into thinking that we are (or have become) a large studio. In the reveal trailer, the first pan of the office space that appears in the trailer includes pretty much most of our company. We have a second smaller room with just a couple of other people, and that’s it. That is the team that made Zeno Clash 2. We can’t do much more with such a small team.

I don’t want it to feel as a complaint, but the truth is we got almost no credit from the press for making a game of the scale of Zeno Clash 2 – something that was easily of the scale of some smaller AAA titles and selling for a third of the price. All this considering our indie nature of course. What simply happened was that people said “Oh, they are a big studio now… they have money” …but that was far from the reality.

I think with Abyss we can focus on making a more polished and smaller game and have more time to iterate on it rather than making something so huge like our previous title. And if that reason isn’t good enough; not getting murdered by our wives for taking so much work home might be another thing to take into consideration.

RPS: Hurrah for not getting murdered! Thank you for your time.


  1. Lukasz says:

    heh. Brooke is in it? awesome.

  2. Viroso says:

    I hope roguelike is less of a fad and becomes a standard thing. For a long time, before roguelike became a thing, I wanted games to be more like that. It’s what makes sense I think.

    Going roguelike, and going procedural, really mean emphasizing systems. Games can become really interesting when the developer’s presence is weaker.

    • tangoliber says:

      Completely agree! I would prefer for procedurally generated/permadeath be the standard. Static, hand-crafted, save-anywhere games should be a niche. But what to call them?

      • Viroso says:

        Another good thing about permadeath is that every time you play it’s fresh. You know what sucks, is when you go back to a game and you remember there’s an inventory full of crap you’re saving for some reason and it keeps interfering with every decision you make.

        Procedural content gives the game a long life while permadeath makes each life count, instead of making a long game that never reaches climax and has a ton of stuff slowing it down.

        • jrodman says:

          Eh, permadeath doesn’t really keep things fresh. Combined with a strong, well curated, procedural content system, it provides the edge to care about that procedural content in a way you might not otherwise. But it’s really the procedural content that makes it work and is difficult. Permadeath is trivial to implement but doesn’t make a game.

          • Viroso says:

            Permadeath keeps the time you’ve spent playing the game from having too much of an influence in your future playthroughs. Like I said, inventories. But there are also a ton of things that may weigh the game down. But when permadeath is implemented, that doesn’t exist. So when you get into the game, you play it with almost no baggage. Like in the 16-bit days, each playthrough is a fresh one.

          • jrodman says:

            Nah, the old nintendo games were a very different feel. It was all about memorization and rote execution.

            Meanwhile a game with procedural content and no permadeath still can be fresh. There’s no requirement that the game be a long continuous experience, even with saves. What permadeath brings is the tension to dig deep into the possible resources afforded.

          • Viroso says:

            You’re like totally skipping over the part I talk about the baggage you carry. Old console games were designed to be short, 1 hour long things. Sure there was memorization, obviously, but that doesn’t have to do with what I’m talking about here.

            There were no saves in a lot of them, maybe some passwords. They were designed to be short. That only worked because you could lose the game permanently, and vice versa. Whenever you picked the game up, even if the levels were all the same, it’d be a new adventure.

            Now, comparison, let’s look at Terraria. Totally procedurally generated but meant to be played for a long time with the same character. Inevitably you get to some point where things you’ve done before have an influence on what you’ll do next.

            You’ll get to a point where you need to look for a specific material, or you commit yourself to build something, now your play time will be defined by things you’ve done before.

            Inevitably a procedural game, or any game that relies a lot more on systems than pre designed levels/narrative/progression, without permadeath is a game that will go for as long as the player wants it to go. Dwarf Fortress, Terraria, Starbound, Minecraft, Sim City, The Sims, Spore, Crusader Kings.

            When a game has permadeath, and when we say permadeath it’s more than just death but also lethality, that forces the game to be short and each playthrough has no past. So when you pick the game up, you’re always starting from zero. That’s what permadeath is all about, starting from zero. Starting from zero is part of what makes the game fresh.

            Even if permadeath isn’t solely responsible for a game’s freshness, it has an enormous influence on that.

          • Josh W says:

            I personally love those long form procedural games, on the one hand, yes having all kinds of random stuff in your inventory can slow things down, but on the other hand, it can mean that the games reward memorisation in a completely differently to the classic rote repetition, instead working in terms of an expansive time frame of potential possibilities and “oh, now I’ve worked out how to do this, I can finally do something about that boss I had to run away from 8 levels ago”, or in the way that in more build-y games you can see the effects of far earlier choices impacting what happens now indirectly. These games, by their chaotic openness, can do long time frames better than those games that would start looping the experience because their systems are less advanced.

            Procedurally generated but static content allows continually changing experiences, but progressive reactive procedural alteration of the world allows the game to keep going for ages longer, assuming there’s some stability in how it reacts to the player. If you don’t have that, you have to keep resetting the world anyway, because it’s that initial world generation that provides all the interest, and so you need permadeath or moving on to new levels etc. These provide their own interesting hardcoreness, but for me they are means to the bit I like.

      • Oasx says:

        Interesting, those two things are the completely opposite of what i want in a game.

  3. Moraven says:

    Makes me think of Odin Sphere a little bit.

  4. Wedge says:

    Unfortunately they didn’t seem to learn anything from how to time animations from a fighting game. Even if there are solid mechanics at work here, I couldn’t deal with those clunky lifeless animations =/

  5. Shooop says:

    “Procedurally generated roguelike” is becoming a checklist akin to the industry’s love of describing their AAA titles as “immersive epic”.

  6. Messofanego says:

    God, this just made me even more excited. A fighting game within an action platformer and an evolving story (maybe like Binding of Isaac) with a lot of replayability?!

    Hearing that it’s based in Chile is awesome, just like how Papo and Yo did Brazil. Seems like ACE Team got a good partnership with Atlus. Zeno Clash 2 absolutely shamed many AAA games for scope and variety, my mind was blown hearing it was made by 8ish people!

    • Carlos Bordeu says:

      It was actually 15 of us who made ZC2 (with a couple of people doing some few things externally). 7 people made Zeno Clash 1. But still… 15 is a tiny number for a game that size.

    • Carlos Bordeu says:

      You also might be interested in taking a look at this video (shows the full world scale of ZC2). Spoilers warning!

      • ncnavguy says:

        Keep up the good work and know that you are building a loyal fan base. I had gotten Rock of Ages on steam without even knowing it was you guys that developed that game also, just bought it on the merits! I cannot wait until Abyss Odyssey comes out. I know you said you are not going to do Early Access on steam but have you considered either a beta or demo to help drum up interest + getting free QA/ bug reporting? I would be on it in a hot second!

        • Carlos Bordeu says:

          A beta would need to be coordinated with Atlus, and it is something we’ve never done (and I guess is also more complicated since QA has to go through console versions too).

          Though we’re not doing early access, there’s been a ton of talk within the studio of how much we want to support the game post launch (with updates and new content). One of the reasons for the extended support will be a feature of the game we haven’t talked about yet, which will probably be the highlight of our next trailer/video.

          There’s also some stuff we want to put in the game that we just cannot manage to do before launch, but are really interested in adding later on. The game would definitely benefit from having a competitive vs mode, and that’s something we’re aiming to hopefully be able to include afterwards.

  7. Lestibournes says:

    This game sounds really interesting and I’d like to play it.

    I hope you not only port Zeno Clash 2 to Linux but also Zeno Clash 1.

    I’m also hoping for a Zeno Clash 3. The story was begging for a sequel, and I’d like to see how you expand the game with more stuff to do besides the main activity, which is the fighting, and making a real open world.

  8. Josh W says:

    This sounds a bit like what I was hoping valley without wind would turn into, and the idea of being able to replay the game as the enemies is really clever!

    For one thing it makes sure that the enemies really will be a match for the player, they have to be, because the player might be playing as them! Also it does that great thing of not separating what’s fun in the process of making a game from the fun the player will have when playing. There’s such a temptation to play games half in the debug menu, particularly in games with complex dynamics, and it shows a good amount of awareness that the team didn’t do that.

    I also really like the idea of running around attacking things as a landslide bull!