GameCentral talks to the lead designer of the latest Total War game about next gen consoles and next generation artificial intelligence.
If you click here you?ll find our hands-on preview of Rome II: Total War, the latest strategy epic from British developer The Creative Assembly. But we?ve also been able to chat to a couple of the developers involved: lead designer James Russell and communications manager Al Bickham.
Russell was able to talk about the nitty gritty of designing the game and balancing historical realism with deep but accessible gameplay. But with Bickham we chatted about the past, present, and future of Total War and why the top-selling PC games never seem to get as much attention as the biggest console titles?
GC: I?ve always enjoyed the Total War games but even though on paper I prefer the Empire time period I have to admit that it?s Rome that works the best. It just seems to have the perfect setting and scale for the series?
JR: Yeah, I think with Empire obviously you?ve got almost the entire world and it?s all about empire-building and European powers going across the different continents. But with Rome I think there?s something? I think you?re absolutely right that it is the perfect setting for an empire-building strategy game, because it is, in people?s heads, the kind of great era of empire-building. I think the fact that it?s in one continental block means that you don?t have these distinct theatres but you?ve still got a huge amount of variety and there?s a continuity to it.
And I think the ancient world, it?s just the sheer visual variety, and cultural variety, you get just makes it really, really exciting to build a strategy game around. Because you?ve got all these cultures that play very differently; visually they?re very different? There?s something absolutely unique about Rome and how the Roman military looked. And you?ve got Greek phalanxes, the Ptolemaics? all these different cultures with their own fighting styles that have a real unique look and feel.
And we?ve also tried to reflect that, not just in the units but in things like the cities. So we?ve got an absolute crazy amount of different cityscapes that we?ve built that are unique for each culture. So we?ve got barbarian settlements and you?ve got these Eastern cities, and they all sort of grow as you build up on the campaign map.
GC: I always got the feeling that some of the earlier games played pretty loose with historical accuracy, but how do you balance the demands of gameplay with realism? The obvious thing to do if you?re just making a game is set-up some sort of rock-paper-scissors relationship between units. But if that doesn?t exist in the real world do you create it? Do you exaggerate what is there?
JR: That?s a good question. Generally we don?t find that there?s a conflict between what history says and how we want the game to work. Our kind of method, as it were, is to recreate? to immerse you in this historical world, right? So we have a very, very historically accurate starting position. So this is how the world was at this time. But from there the player has their own agency, they can do whatever they want. If you?re Rome you can just build a huge navy and decide to go straight for Scotland if you want.
GC: Had they invented whisky at that point?
JR: [laughs] I don?t know. But because you can develop your society and your civilisation as you want you change history, you write your own history. So what we want to do is recreate the dilemmas and the worries of a Roman leader at that time, if you?re playing as Rome.
But in terms of different units and tactics, what?s amazing about history is it almost is gameplay. Every society created a military tactic that its enemies would potentially fall victim to, and they would often counter it. They?d be like, ?Oh god, we got destroyed! How are we going to deal with that tactic?? So they create a counter, and then that counter ? the other enemy, they respond to that.
So the nature of civilisation and conflict is tactic, success, counter, and counter-counter. And so it works very well in the context of the game. And so what we want to do is establish these opportunities and tactics and counter-tactics, and let the player kind of explore that space. But we don?t feel we have to make it up, because this is the stuff of civilisational conflict. It?s very interesting.
GC: Whenever I talk to a console developer about a new sequel one of the questions is inevitably about dumbing down, or if it?s an old franchise of ?modernising?. Is that a problem with Total War? Does being on PC shield it from that or is there anything you?ve had to fight to include in the attempt, presumably, to broaden the appeal?
JR: What?s really important to us is to maintain the gameplay depth. It?s really important to us that the game is deep and interesting, and we add gameplay features that add depth, right? But it?s also important to make sure those things don?t make the game inaccessible. It?s the classic kind of mantra of what makes a great game. What makes a great game is something that?s easy to pick up and takes a lifetime to master.
And that?s obviously what every kind of game designer strives for, right? To create a game that anyone can understand the rules and what you?ve gotta do, but actually there?s a lot of subtle depth there that means that real expert players can use things that the other players aren?t necessarily worried too much about.
But it?s also important, in battle for instance it?s very much part of our philosophy to have real world features working in the game. What that means is that if a tactic works in the real world it should work in the game. So things like flanking, morale, we?ve got this new line of sights system ? if you?re hiding behind a hill you can?t see so you need to do scouting behaviour and so on.
These are deep features that will allow hardcore players to become much more effective as they learn how to wield them properly. But they?re also intuitive, you know? Fundamentally the battles are relatively straightforward to pick up and play. You select a unit and you tell it to attack another unit, or move to a particular place.
But if you want to play with the advance tactics you really just have to think about what would work in the real world. And I think that?s the way we try and have our cake and eat it in terms of what you were talking about? about making it accessible to a wider audience. I think we can do that, we try to do that without creating a tension between catering for the hardcore fans or the new players. We can try and do both.
And that makes the game better because we can make the game deeper and we can make it more accessible. Those two things aren?t conflicting. They?re hard to do, but I don?t necessarily see them as conflicting.
GC: The longest standing criticism by far with Total War is the artificial intelligence, which I assume you?re going to tell me is much better than it used to be. But how much of a priority has it really been? Considering it?s not something you can show off in a screenshot or video.
JR: I think the AI is absolutely a critical thing for making a satisfying, deep experience for players. And absolutely I?ll say it?s better than ever before, because it absolutely is! [laughs]
GC: But why should I believe you when you say that now? As I?m sure it?s what you?ve said before the release of all the other games too.
JR: Well, I think the proof is in the pudding, so hopefully when the game is out there people will see. But to give some concrete examples, first of all we?ve got more programmers working exclusively on AI than we?ve ever had before. So we?ve got small teams of programmers working exclusively on the campaign AI, we?ve got a team working exclusively on the battle AI. I think the campaign AI moved on a lot in Shogun 2, and it?s moving on a lot again.
But it?s also about letting the player have insight into how the AI?s thinking, and making it feel human and convincing. You want to understand why the AI might get ****ed off at you or why the AI might want to attack you, right? On the battlefield I think we?ve got a programmer who?s just been working on siege AI.
And we?ve got new features around the data, to give the AI a real good understanding of streets and open squares and choke points, so it can understand that it needs to guard that particular street or that that particular breach is really fundamental. So there?s new data actually in the maps themselves that helps the AI understand. And we?re still balancing the strengths of those behaviours, so we?ve got specific behaviour types for different things?
GC: It must be difficult as well because if you read history battles are very often lost because of an idiotic mistake, which is obviously a problem when you?re trying to make a game of it?
JR: That is true as well!
GC: It?s a bit like driving games where if they stick too close to the racing line computer-controlled opponents just seem utterly robotic.
JR: There is an element of that, but I think the main thing is we want the AIs to be convincingly human and also? the system needs to have a bit of noise in it. Because otherwise the AI can become predictable. So what you want to do is make sure that you?re kept guessing.
And another thing that?s important is that the new line of sight system means there?s a lot more uncertainty and tension in the battles. Because you don?t necessarily know where all the enemy forces are, and that creates this kind of emotional uncertainty around the way you encounter the AI that really makes it different from before.
GC: OK, that?s great. Thank you.
JR: Thank you!
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.