Developer Newsletter 19 – An Interview with Leif


Greetings Commanders!

We’re doing something a bit different this newsletter, something that I’ve always wanted to do, it’s an interview that I recently recorded between myself (Josh) and Leif, World Designer on Total War: ARENA. 

Josh: Thank you for joining me Leif. We’re going to be talking about maps if that’s okay?

Leif: *laughs* that’s fine by me.

J: Alright then, so let’s start with an easy one, what’s your favourite map in-game?

Leif: That’s always an interesting question because it’s like picking your favourite child. There are things about every map that I’ve liked or haven’t liked so much. That being said, there are a few things I can pick. I’ve always had a soft spot for Salernum. It was pretty much my first fully realized map of my own design, so it’ll always have a special place in my heart. Otherwise the latest map we’re working on always becomes my favourite because you get enthusiastic about the new ideas. It does always feel like we’re raising the initial quality bar for every map we make.

J: That raises a good point, what’s the balance like between looking at old maps and working on new ones?

L: We’re always looking at things like the balance of the maps, and to an extent their popularity. When the earlier maps have been left behind a little bit in the previous realm of quality, we look into doing a visual or gameplay update.

J: Like that GFX update?

L: Do you mean the VFX update?

J: …

L: *laughs* Yes, very much like the VFX update. But those updates do add technical challenges, we have to make sure we’re not reducing the technical performance. We have to maintain the minimum performance requirement.

J: Okay, so that covers old maps, but what’s the first step in making a new map? I’m sure there are lots of possibilities and suggestions, how do you settle on an environment?

L: It may seem like we throw a dart at a world map, but it’s a combination of things. We look at historical battlefields or places where there were historical moments. We look at those in relation to the commanders and factions currently in-game, or coming up, and create a shortlist. Then I’ll do further research, and if a battlefield had particularly interesting topology, that eases me into choosing it. It’s also a bit of ‘I like the idea of this location’, I’ll find images that look really interesting and striking, that’s especially the case for Oasis. We knew we wanted a desert map linked to a commander, so I looked at loads of desert battlegrounds and we found one with one striking rock feature. We took a little bit of a liberty with how close it was to the real location, but that allows us to create something really cool.

J: What is a blockout?

L: Essentially it’s the first stage of a map’s development, it can be called a blockmesh, a whitebox or whatever else, but that’s what it boils down to, it’s the first version of a map that’s built purely for the high-level gameplay. We don’t need it to look good, we need it to figure out how it’s going to play. We use placeholder boxes to create the layouts, and then we figure out the gameplay aspect before getting into the grittier details.

J: What then? Can you tell us about the overall production pipeline for map development? Do you start with concept art or a blockout?

L: So sometimes we start with a blockout, sometimes we start with concept art. Typically I’ll start with research and put together a brief. Then I’ll make top-down drawings and sketches to get ideas out of my head, maybe it’s going to have a little river in it which will affect things in this way, and stuff like that. Then typically I’ll go straight into blocking it out, but I’ll use the brief, theme, sketches and references to get the concept art started, usually for a capture area, because that’s really useful for the artist to work out how the map should look and feel. Then between the designer and artists we come to an agreement on a certain image of how the map looks and feels.

J: And where does it go from there?

L: Then there’s the blockout process and the series of playtests and iterations on the design and once we’re design confident, the artist will start doing an art pass, although they would have already started on any new assets. Towards the end we’re working together to beautify it, making sure the art and design complement each other, sometimes the design will inspire the art, but sometimes the art will inspire a new part of the design.

J: How many iterations would you usually go through?

L: Ooh, anywhere between a couple and about seven. It varies based on the map, some are more difficult to figure out than others.

J: Depending on how complex the gameplay is?

L: Yeah, and sometimes you have an idea that hasn’t quite worked yet, and it takes a few iterations to get it working. We went through this with Capua and the flow of battle. We went through a lot of iterations to make sure that players didn’t end up in the corners of the map and had clear paths to the enemy. Sometimes it’s as simple as opening up a road or closing a road or even moving the watchtowers, it all has an effect on the overall flow.

J: Okay, well you said earlier that you hit a point where you were ‘design confident’, how do you know when you’re confident? What clicks for you?

L: Ooh… Err, I think a lot of it is intuition that comes from practice and experience. You get to a point where you understand the game and what plays well, and you can appreciate the map as both a player and a designer. Then I’m confident that it’s a functional map that plays well. It’s hard to clarify it in a clinical way, it’s about making that judgement call.

J: And we do see a lot of love for the maps, so we can rest assured that your judgement calls are good.

L: *laughs* thanks.

J: Well, I think that’s all we’re going to have time for, I have more than enough to write up, thank you for joining me.

L: Thank you for having me.

Hopefully you found this interesting! Let me know below if you’d like to see other Developer interviews.

See you on the battlefield.


Next 22/11/2018 ARENA is ceasing live service February 2019
Previous 07/11/2018 Patch 3.1.15 is live!