Q&A: Sega on Bringing Valkyria Chronicles 2 to the PSP

[Valkyria Chronicles 2's producer Shinji Motoyama and director Takeshi Ozawa talk to Gamasutra on scaling the console game down for portables, and how to design great maps in strategy games.] Valkyria Chronicles was Sega’s acclaimed entry into the tactics RPG genre from Sega, with innovative active battle sequences that incorporated non-grid movement with dice rolls and finite turns. Valkyria Chronicles was on the PS3, and sported an appealing pastel pencil storybook visual style. Sega has just released a sequel to the original on PSP in Japan, designed around a multiplayer experience for the Monster Hunter crowd, but without following the Capcom juggernaut’s playstyle (incidentally there is still a single player campaign). In the past, Gamasutra interviewed the original game’s producer Ryutaro Nonaka and director Shuntaro Tanaka extensively, but this time around the game is getting some new blood. Gamasutra spoke with Valkyria Chronicles 2’s director Takeshi Ozawa, and producer Shinji Motoyama to see just how they scaled the experience down – and up. One method not mentioned in the interview is splitting levels up into multiple areas, allowing tiered maps to save on geometry while still allowing a vast playfield. Graphically the new game looks pretty similar to the PS3; maybe just fewer polygons. How much of a challenge was it, making it work on the PSP? Takeshi Ozawa: Well, without going into the nitty-gritty, there are assorted issues that the engine has to work with. The hardware platform's completely different, of course. The PS3's a high-end machine while the PSP's lucky if it's half as powerful. You just can't expect to output as many polygons on that machine, and the way light coloring and so on works is also different. So one of the main issues we had to deal with on the PSP was how to deal with colors -- not have a shader overlay on top of everything like on the PS3, but still retain that canvas-like look that's the game's trademark. That required a lot of trial-and-error to accomplish. Do you have previous PS2 development experience, and does that help when optimizing? TO: In terms of being able to ballpark-guess the upper limits of what we'd be capable of accomplishing with the hardware audiovisually, I'd say that -- and we've all got PS2 experience, yes -- I'd say that experience is something we can advantage of. Our staff doesn't have to actively benchmark everything out; instead they can sort of eyeball it, so to speak. This is not a criticism, but simply for the sake of asking - why make a sequel and not a port? What sorts of new things did you want to put across? TO: Well, the story is completely new, set two years after the original game, and the characters and battlefields are all new and revised. In addition, there's also a new and very substantial online-play mode, with both co-op and competitive modes available; something that was possible with the PSP that really piqued our interest. The volume of the game is a lot larger than before, too. Those are the sorts of things we were challenging ourselves to do this game, I'd say. Shinji Motoyama: Multiplayer was something all of us on the team wanted, but after that, we also considered how we could improve on the BLiTZ battle system (the realtime/turn-based hybrid strategic battles in Valkyria Chronicles), which had gotten a lot of positive response. We wanted something that everyone could play together - the story of the previous game put a lot of emphasis on the bonds the characters forge, so we wanted a way for players to form those bonds with each other within the game. With the PSP, and especially with Monster Hunter dominating the marketplace, there's a demand for online games that people can play while talking to each other in real life. With the sort of multiplayer we had conceived, we thought the PSP was the ideal platform for that -- being able to call out to your friends to take out this or that unit and forming strategies with each other while you're playing. So that was one motivation behind choosing the PSP. What sort of online play types are there? TO: Well, there's co-op and competitive. In the previous game you consumed command points (CP) in order to move people around maps; here, two people can do that at the same time. When this was first announced, I said to myself that it'd be nice to have a mode like that. TO: Oh, exactly. Each character has a certain amount of CP that he or she carries, but you can pass these around between people and consume it as you work together. You can have both characters attack, or have one attract enemy fire while the other goes behind their flank. That sort of thing. In competitive battle, it turns into something very much like a chess game. There are a lot more customizable options in this game, too; you can create your own trooper and fight with him. Can you make battle maps? TO: Well, not battle maps, but we have a lot of maps available for play, and each map has its own special features - the previous game had sandstorms in the desert map, but now there's something like that in every map, which opens up the amount of choices available to you a bit. You can also set the rules of battle - in the previous game you had objectives like "destroy the enemy base" or "defeat all enemies," and now you can define what objectives you'd like to have in your competitive match, along with the time limit and the field of battle. How do you design levels with the active-time system, in terms of determining optimal positioning for buildings and obstacles and things like that? When characters have different lengths of walking, their walking ability is different, so how carefully do you have to construct these worlds so that you know a scout can go this far, a lancer can go this far, a sniper can go this far...determining placement of the crow's nest and buildings and things like that. Breakable walls for the tank. How do you go about that? SM: Well, in the original, you had your low-level unit types that could be leveled into higher-level types. In this game you've got 35 different types -- for example, you might have a character who's mainly a spy, but is at least somewhat capable in battle as well, or has more movement range than normal. That sort of customization is now available. It's not like a spy can transform into an anti-tank gunner all of a sudden, and as a result of that, making every team member play his or her specific role to the hilt plays a far more important role in this game. The level design, naturally, follows from that. It must require a lot of play-testing. TO: It does. SM: Every day. TO: Every day! At what point did you begin the playtest cycle? SM: Well, the project runs on milestones, and before the alpha - at certain points in post-production whenever we had a sizable chunk of the game up and running - we'd run playtests and make that part of the milestone. Then when we’re wrapping things up, we can look across the entirety of the game and checking the balance and "fun factor" behind each section.