Off the Grid interviews Brian Reynolds of Big Huge Games

Every other week Scott Jon Siegel contributes Off the Grid, a column on gaming away from the television screen or monitor.

Big Huge Games, the developer behind the acclaimed Rise of Nations series, recently made a splash on the internet with the announcement that they were bringing the board game Settlers of Catan to Xbox Live Arcade. It's news that even palpitated the steel heart of Penny Arcade's Tycho, and for damn good reason.

I recently had a chance to discuss Catan with Brian Reynolds, CEO and creative director of Big Huge Games. Reynolds talks at length about the difficult process of adapting such an esteemed title, touching upon issues like designing challenging computer opponents, mapping moves and menus to the 360 controller, and tweaking Catan's refined rules for ranked and non-ranked matches on Xbox Live. It's all here. And we've got photos too!

Let's talk about how Catan Live was born. Who went to who with the idea? Did Big Huge Games want to make Catan for XBLA, or did Microsoft want Big Huge Games to make Catan?

It was actually Microsoft who approached us – last spring I had no idea the project possibility even existed, but Microsoft was looking for developers to bring "Euro" board games to XBLA, and they came to us early in the process. Obviously once we knew about the project we were very excited.

So you were tasked with porting a renowned analog title to a digital platform. Where do you begin? What was the absolute first step in designing?

Wow, the very first step? In the rulebook for the board game, there are some black-and-white diagrams of the mapboard. I took one of those and xeroxed it up to the size of a full page, and then used it to create a system for numbering the hex tiles, the corners of the tiles, and the edges of the tiles that would be easy for the computer to deal with. Because one of my two initial concerns was whether we'd be able to make a decent A.I. for the computer player, so I started right in on some basic A.I. and rules coding, since I didn't need much graphics work to be done to start working on that.

The word is that Klaus Teuber had a hand in the project. What did he bring to the development process?

Yes, he was instrumental in helping us create the A.I. for the game. It turns out that over the years since designing the original board game, Klaus Teuber has always wanted to have a strong computer player, and so he had put a lot of thought into what the proper strategies and tactics were, and formulas that would be useful for helping a computer player evaluate choices. Of course he'd never had a chance to bring these into play, because the developers of previous versions of the game either didn't have time to do a full treatment of his ideas, or else decided to try their own approach altogether. So he had these Excel spreadsheets full of formulae, plus a nice write-up he'd done. Being an old-time A.I. guy, I looked at these and saw gold: here was somebody (the designer of the game no less) who'd already done the legwork of collecting the tactical situations and strategies, and even done some of the work of creating algorithms to choose between them. So I was able to blast through all of that stuff in a few weeks, and use most of my time refining the really high-end game for the expert players. The result... a much stronger opponent at the top level!

%Gallery-1729%

Were you a Settlers fan before starting the project? Has working on Catan Live burnt you out on the game, or perhaps reinvigorated your interest?

Oh yes, I've been a big fan of Settlers of Catan (and the games of Klaus Teuber in general) for years. In fact I have a 1st place trophy for Settlers of Catan from our regional Maryland tournament, and the fact that I'd played the game at the tournament level was definitely a help in doing the A.I. It's funny, but if anything working on Catan Live has gotten me even more into playing the game – I played in another tournament this fall (only 5th place this time, I'm afraid), and I've been playing a bunch with my family who all love the Catan games.

How far along is Big Huge Games in the development process?

We just sent a build off to test that we have high hopes will be our "final certification" build. In other words we think that Big Huge is done working on the product. If we're right, you'll probably see Catan on Live Arcade sometime in mid-to-late March.

Have there been any design or development problems worth highlighting? Anything about the game that made it particularly difficult to move to a digital format?

Although when you play Settlers of Catan for the first time your first impression is mainly of the map and pieces, the heart of the game system is the trading – the wheeling and dealing with other players is a major driver in making the game fun. So a huge concern for us was creating a "trading" interface which would not only allow players to trade, but which would also make sure players would have fun while trading. We did a lot of versions of this screen, and spent a lot of time in one of our local University's user testing labs while getting this screen just right. About half of the challenge was guiding players naturally into how to make trading offers and easily understand the offers being made to them (so for example the arrow coming towards you from the other player shows you the resource he's offering you, and the arrow going away from you toward the other players shows the resource he wants from you, etc).

The other challenge was giving players ways to communicate as many nuances as possible in their negotiations – just "refusing" an offer tells little about why you're refusing it, and if there's something the other player can do to make the offer attractive, for instance. Obviously many XBLA players will use the voice chat, which certainly helps a lot, but we didn't think that was enough since it's not very visual and not all players use it. So we created the "Tickler" emotes, which allow players to send little graphic animations to each other – smashing the dice with a hammer if you don't like the die roll, blowing a kiss to someone who made you a nice trade, putting a target over the avatar of someone you think is too close to winning, and so forth. We have 16 different ticklers in the game, in four easy-to-navigate pages, so you can quickly and visually emote your feelings to the other players. We also wrote A.I. for the computer players to use these features too where appropriate, which definitely adds a lot to the solo game.

Let's talk about those computer players! Does Sun Tzu play Catan different than Elizabeth? Should I be worried about Frederick? What other personalities do you have planned for the release, and how do they differ from one another?

The AI personalities have several possible kinds of "quirks" which affect how they approach the game.

The first is their overall "Strategy", which can be either "Largest Army", "Longest Road", or "generating the most resources". This determines their favorite way to gain the winning edge in victory points, and will affect the priority with which they'll choose different things to build.

Next is their "Favorite Resource", which can be either Brick, or Ore, or neither. Brick and Ore tend to be the rarest resources in the game (and are important for roads/settlements and cities/cards, respectively), so personalities which tend to prefer settling on one or the other resource will tend to play somewhat different games.

Then there's "Expansion" which controls whether they like to expand by grabbing a lot of settlement spots quickly or "concentrate" on improving their existing settlements to cities first.

Then comes "Adaptability". A high score here means that if they don't have enough resources to build the thing they really want right now, they're more willing to go ahead and buy their second choice if they can afford it; this decreases their risk of getting nothing because of the robber, but can make it take longer to achieve their primary goal. A bot with a lower adaptability will persist in its primary strategy even at a higher risk.

Finally of course, there's the Difficulty level itself, which affects both trading and map tactics, but especially trading. Easy AI's will trade you anything in their hand (they are, after all, easy), whereas a Hard AI is like a seasoned player who will only trade when he sees "advantage". Easy AI's will avoid blocking your development with settlements, where Hard AI's will have no such compunction (and if you're winning will actively seek to thwart you). They don't "gang up" on the human player specifically, only on players who seem to be winning, so you'll sometimes see them start massive robber wars with each other which is fun.

In terms of the personalities, the ones that beat me personally the most are Alexander and Joan of Arc. Of course your mileage may vary!


There have been a few other official online adaptations of Settlers over the years. Were those used to inform certain design decisions, or did the team start more from scratch?

We looked at each of the previous official (and unofficial) adaptations, as useful guides to "well this worked but that didn't" sorts of questions. For each version we'd come up with kind of a "good, bad, and ugly" list, and see if it made sense to try something from the good list ourselves, and definitely avoid the bad list, etc. Of course the other online versions were all PC versions and we were doing the game for a console, so we had a bunch of different concerns. Also since this was for XBLA, it's a somewhat different audience. Finally, we had a bigger budget than the previous online versions probably had, so we had many opportunities to do things that hadn't been done with the franchise before.

Were there any unique design or gameplay implementations made in the course of development? Did you stray at all from traditional Settlers interpretations?

When you play a Ranked game on XBLA, it's the standard classic game rules of course. But as a devoted Settlers of Catan fan something high on my list of stuff to put in the online version were "House Rules", because of course some people like to play with event cards instead of die rolls for resource production, some people like to make rules about how the robber works at the beginning of the game, and stuff like that. So we do indeed have a bunch of custom rules you can set if you're playing a Custom match. We also included a "turn timer" to facilitate quick online play: in a Ranked match you have 60 seconds from the roll of the dice to complete your entire turn. That extends to 90 seconds in an unranked Quick Match, and in a Custom match you can set the clock however you like.

Looking at the screenshots, one thing that jumps out almost immediately is the minimal presence of the HUD elements. Online Catan games have always been plagued with a ridiculous amount of overhead, based solely on the amount of information being presented in Settlers at any one time. How did Big Huge Games reconcile that to create such a subtle interface?

It's the magic of the console controller, baby! Something which became very quickly apparent when we started creating our XBLA game is that this game is made to be played with a console controller. It's so much easier to snap a new settlement or road into the right position using the joystick and then press (A) to build it than it could ever be to find "the right pixel" to click on with a mouse. And likewise the triggers and bumpers on the controller perform great service in making the information you want available (but only when you need it). For example, pulling the right trigger always shows you the uncluttered map (even if you're on the trading screen for example) and shows you the players current score totals; pulling the left trigger shows you the prices of the things you can build (again, this is accessible from anywhere in the game). The bumpers bring up charts about how many resources are in play, scoring breakdowns, and so forth. And there are all sorts of fun "advanced player" charts available with combinations of bumpers and triggers – like if you want to see how the die rolls so far are mapping to the "bell curve" (I call this the "whining about the dice screen"), or if you'd like an overlay of where all the best settlement locations are! All of these can be pulled up at any point in the game, from any menu, but the beauty is you only see them when you actually want them.

Is Catan Live an indication that Big Huge Games is moving away from the Rise of Nations series, or is this simply a one-time thing? Any plans for more XBLA titles? Or more analog adaptations?

Big Huge Games has gotten big enough and huge enough that we can work on more than one project at the same time. In fact right now I think we're working on at least three(!). So we've got all sorts of things in production, and any one game you see from us doesn't necessarily say much about what the other projects are about, and certainly shouldn't make you think we're changing directions in a radical way or something like that. Obviously I can't talk about future unannounced stuff in much detail. As far as this new venture into XBLA, obviously time will tell – and it should go without saying that the more successful we are with this game the more likely you are to see more stuff along those lines.

Finally, any other tricks up your sleeve regarding the presentation or gameplay that you'd like to mention? Any subtleties that gamers should look out for?

Particularly players coming from the board game will be happily surprised at how quickly you can play a really fun game of Catan in the XBLA version. Here's a tip: when you first play the game the resource cards are dealt out rather slowly while you're learning to follow the game and the interface. But once you've played a few games you should go to the options and turn the "Card Dealing" speed up much higher – because once you can follow the action you'll love how quickly you can negotiate, trade, and play!

Thanks for the great interview, Brian! And good luck with Catan Live!

(All development photos and screenshots appear courtesy of Brian Reynolds and Big Huge Games. Thanks, guys!)

Scott Jon Siegel is a fledgling game designer, and fancies himself a bit of a writer on the topic as well. His words and games can be found at numberless, which is almost always a work in progress.