Mount & Blade: Warband is one of my favourite games but I haven't played it for a long time. In part, that's because I've been waiting for the sequel, Bannerlord [official site], since it was announced four years ago. After over half a decade of development, details about the game have started to emerge and I spoke to Armagan Yavuz, CEO and Founder of developers TaleWorlds, to find out how the team are aiming to improve on the dynamic world of the original. We talked combat, historical influence, settlement management, co-operative possibilities, modding and AI.
RPS: One of the key elements of Warband, which very few games attempt, is the creation of a dynamic world, that supports both strategic play and a more RPG style experience. What are the main ways you're building on that?
Yavuz: With Warband, we felt we had a very original game with some unique elements. A lot of players and critics said that they felt it was like a rough uncut gem. There was a lot of potential but it didn't quite meet that potential, and I agree with that. So we've tried to improve on all of the various elements, as well as trying to make them click with one another more effectively. We want all of the mechanics to work togeether.
At the heart of that, there's a new scripting system, which is C#-based. It allows for many more sophisticated mechanics, from the way seasons work in the game to the functions of the AI. In Bannerlord, the AI can use any gameplay mechanic in the same way that the player can, whereas in Warband there were things that worked differently for the AI and the player. We're getting rid of almost all of that, so that the player and the AI are working on an equal footing.
For example, you can talk to your enemies' vassals and snatch them, poach them for your faction. Now, they can do the same thing to your vassals. That means you always have to be on your toes because whatever plans you might come up with, the AI can be coming up with similar plans to use against you.
RPS: Building an AI capable of working with such complex systems must be a huge challenge. How do you begin to build something like that?
Yavuz: One of the most important things is to make sure the AI can evaluate all of its options. If it doesn't know what the options are, it won't use them, just as a player won't. We use a very modular system, which works such that when we add new features to the game, the AI is automatically able to see them and use them.
We approach the design the same way when we think about the players' experience, introducing new elements that overlap with things that you've already learned. We try to make the game more transparent to players - if you don't know how to do something, or even know that it's possible, you might as well not be able to do it.
There are a lot of mechanics that make the economy and politics more fluid, and we want to make sure that managing villages and diplomacy doesn't become an intellectual load. It's very easy for a game to become intimidating when there is so much to do and so many options, so we need to make sure there is no information overload.
To do that, we try to give the player very simple interfaces to interact with. They're very rich in information but not overwhelming. The information that you need when trying to perform any action should always be visible, and positioned under your mouse pointer as soon as you need it. For managing villages, you basically have three sliders - militia resources, taxes and building resources - and you decide how much importance to place on each area.
And as a beginner player you can leave them all in the middle and not worry too much. The game doesn't force you to optimise constantly, or to care about all of this stuff just to survive. If you are part of a kingdom and you have a single village to manage, you can do that suboptimally without altering the course of the war too much. That way, you can learn the more complex mechanics and how they all tie together while working as part of a bigger system, without too many responsibilities. That's a natural learning curve and there's plenty of time to master strategies as you play.
The entire UI is much more streamlined than in Warband as well. You can see characters in the gameworld and interact with them directly, and you'll be able to access various properties and statistics directly rather than looking through menus.
RPS: Why did you decide to set the game two hundred years earlier than Warband?
Yavuz: We decided early on that we would either be basing the game either a little earlier or later, because we didn't want to revisit exactly the same time. We opted to go earlier for several reasons, one being that if we go too much later then, realistically, the combat changes a lot and becomes based on heavy armours, full plate, and firearms.
There are lots of interesting things to look at in that setting and there's a charm to it, so it's something to explore at another point maybe. But it's not what we wanted to do just yet because we wanted to retain the medievalish feel of the combat.
One thing that opens up in this setting is the use of female characters. In Warband we had female nobles but they didn't have armies and they weren't commanders, because that didn't fit with the time period. But if you go back in time a little bit, you see that a lot of societies did have female leaders. It was much more prevalent and it's something that we've implemented. It's an fascinating period to explore culturally and gives us a lot more freedom to create some new social dynamics.
RPS: How do you balance historical realism and entertainment? Medieval life wasn't all fun and games...
Yavuz: First of all, we arevery much interested in history and we've learned a lot while making the game. Steve Negus, our writer, is a super history nerd and a great source of knowledge. There are a lot of really interesting, knowledgeable people in our community as well.
We aim to keep the game balanced between fun and realism. What's important is that whatever mechanics we use are believable within the world we've created. We try to use history as a source of inspiration rather than a script to follow. But we're always surprised by the number of inspiring ideas that come from historical research. As we dig into it, we find so much that we can use.
RPS: What changes have you made to combat?
Yavuz: The idea was to keep the basic mechanics that worked really well but to evolve them and to add mechanics so that everything feels more natural and polished. We've redone almost every animation with mo-cap and thrown in some physics-based calculations as well, which actually work to balance the speed of animations.
As you play, you realise that doing certain actions feels faster or slower according to the situation of your body. You don't need to learn lots of combinations and controls, but you'll come to understand the tempo and the rhythm of it in a naturalistic way, and you'll catch certain methods and be able to use then more intuitively.
RPS: Can you talk about how minor factions will work?
Yavuz: We can't share too much information on minor factions yet. They're a way to make the game world richer, as well as the lore. The exact mechanics aren't quite as rich as with the core factions but they open up new possibilities.
RPS: You've mentioned elsewhere that modding support is important to you. What are your plans for Bannerlord modding?
Yavuz: The modding scene is something that we are really fortunate to have. We're blessed with a great community, and the modding tools are one of the things that we want to develop as much as possible and to the highest possible standards. We were able to draw a lot of lessons from the things that we did and didn't do well with the previous games, so we're fixing some of the mistakes and coming up with a much better system.
There is a very rich scene editor that people can use and we'll also have very powerful scripting tool, written in C#. The goal is to make sure that even after the game is released, we'll be able to patch and change things, with as little effect on the mods as possible. And further to that, we're trying to come up with a system that makes it possible for different mods to work together.
The community also creates lots of fanfiction - the world is important to them and we're making it richer this time around. The lore isn't at the forefront though and part of the reason that it's not so pronounced is that the game also works a bit different each time you play. Part of the lore becomes your own personal history. In Bannerlord, there's more dialogue, more small stories and more interesting characters, but each player's experience of the game will be unique. We build algorithms that enable narrative rather than scripts.
RPS: A cooperative campaign is a holy grail for a lot of Mount & Blade fans. Is it a possibility?
Yavuz: It's very difficult to do, not just because of the technical difficulty, but also to make things practically playable when we have two people doing wildly different things in real-time. One player might be trying to have a very exciting battle that is the climax of a very important experience, and one player just beforehand decides to go to town and look at the marketplace. These people have to be in the same gameworld and it's very difficult to make sure that they're both enjoying themselves and all having a great campaign experience simultaneously. It's almost impossible without cutting down on what the game offers.
There may be another way to manage all of those things, by limiting the co-op to one kind of campaign. Let people play together as a party and have them always be together. That might be possible and that may be the the only kind of co-op that we can deliver. It's something we're experimenting with and that we have worked on. We'll only officially announce something if we can make it 100% efficient and fun to play though.
RPS: If you could point to one thing that has improved since Warband, what would it be?
Yavuz: It's the way that you're involved with the game world. In Warband, the player didn't have enough ways to interact with it. There was, for example, no way to hold proper diplomacy and conversations due to very limited dialogue choices. I think that's one of the most important things that we tried to address.
Whenever you feel that you'd like to do something in the game, you have a much better chance of being able to do that thing in Bannerlord. Say you're cornered in a castle and you have your enemy's son as a prisoner. You might want to give him his son back so that he'll leave you alone - those things will be much more possible in Bannerlord. You'll be able to interact much more with the game characters and the game world than was possible in Warband.
RPS: And finally, how close are you to release, will you consider Early Access, and have you been building foundations for the future as well as for this one game?
Yavuz: We're still not too close to release unfortunately. We are considering Early Access, or perhaps an open beta of some sort. We definitely want to involve players at some point so they can dig in and help with final touches and game balance.
And, yes, the big challenge was to make a platform for the future with this game. That's one reason that it took so long. At the beginning, some of the choices that we made - technological and design - were suboptimal. They became limiting factors when we wanted to add more things, which led to having to do things two or three times. We've improved in that regard. But the most important thing is to make this a great game, which can hopefully then be a great foundation not only for mods and expansions, but for future games and projects.
RPS: Thanks for your time.