Postmortem: Lionhead Studios' Black & White
June 13, 2001 Page 2 of 3
What Went Right
1. It got finished. This sounds stupid, but we encountered some big problems, and there were times when we doubted that the game (as it ultimately ended up) would get released. As a new company, we not only had to work out the game we were going to create, but we had to write the tools and libraries, create everything from scratch in software, and also gel together as a team.
We couldn't have a dress rehearsal for this, so we learned by trying things and then changing them if they didn't work. As time rolled on, we couldn't afford to make any mistakes or pursue blind alleys. For example, we talked about updating some of the graphics at one point. It didn't seem a big job, but once we'd changed some of the buildings in the tribal villages, they showed up any unchanged ones and made them look less impressive, so we had to assign time to do them all. We got a much better set of buildings out of this, but if we'd known that we'd have had to do all of them, we would have said, rightly, that there wasn't time.
The programmers were likewise coming up with neater and neater ways of coding, and thus trying to do more and more with the code they had. It says a lot about the talented and single-minded development team at Lionhead that everybody always wanted to make every element that little bit better.
And as we fixed the bugs and sent the game to QA, we felt like people who'd run a marathon and could see the finish line, but it didn't seem to be getting any closer. Perhaps this is a function of not getting enough sleep over a period of several months.
2. All the risks paid off. We wanted to do some pretty groundbreaking things in Black & White. One example was doing away with the panel of controls and using the Gesture system for casting Miracles. We tried and tried to get this feeling just right, and if we'd had to dump it, I'd have been so disappointed. But after research, testing, and simple trial and error we got it working beautifully, and we now have a feature no one else does.
Also, integrating the story line into such a free-flowing strategy game was a risk. We thought it would sit quietly behind the game, popping up to direct you if you hadn't moved on, but the story came alive and started to draw the player through the game in a way none of us, apart from perhaps scriptwriter James Leach, had envisaged. It also gave us characters such as Sable, the Creature trainer, and those advisors whom we hear people now quoting lines from, and who exist outside the game as recognizable characters.
The huge, learning, intelligent Creature was also more of a gamble than he now seems. To go into AI in such an in-depth way required Richard Evans, our AI programmer, to consider what learning was, how practice works, and how the reinforcement of ideas comes about. Then he built all this into a character which appeared to live and learn like, say, a clever puppy. AI is always a minefield, and I'm always disappointed by great strategy games which appear to have the most simple, easy-to-predict AI running your enemies. We just wanted to advance the technology to its extreme.
We also wanted to do more with graphics and animation blending. The world changes depending on whether you're playing as a good or an evil god, and things take on subtle new looks. The Creature, the player's hand, and many of the buildings change, and we used more animation blending to achieve smooth movement and changes than anyone else has ever done, I believe.
We're also the first game (apart from Microsoft's Flight Simulator) which enables you to import real weather in real time into the world. We are also the first to enable unified messaging, whereby you can send messages to the web from the game, or receive them, using e-mail and mobile phones. This integrated two-way messaging as well as the ability to take your Creature out of Black & White and onto the web is brand-new. Also, the game can import names from your e-mail package and assign them to unique villagers in your tribe in the game. I expect lots of games to do similar things in the future, but we took massive risks and devoted huge amounts of effort to being the first and to making it work properly.
3. The game looks so stunning. When we started, we used a wireframe test bed and a couple of conceptual screenshots to provide some atmosphere. I first showed the test bed and these mocked-up screenshots to the press at E3 in Atlanta in 1998, and I could see on the assembled faces that nobody believed we could accomplish anything like it in the final game. I was complimented on the depth and beauty of preliminary efforts, but the compliments had a slightly hollow ring. I could almost hear people thinking, "Yeah, it looks great, but anyone can draw pretty screens using an art package. What's your game really going to look like?"
Concept art featuring Horny from Dungeon Keeper—a great deal to live up to.
Not only did we manage to pull off the look we wanted, but we exceeded it by some margin. The sheer beauty of the lands is something I hope won't be matched for a while, and the fact that you can move, zoom, and rotate to view it from any angle, anywhere in the game, is again something we got spot-on.
Looking back, I don't know whether we were insanely ambitious, because at the time we started, you couldn't have done what we did. We needed so much custom-written software, and we also needed the minimum specification of the PC community at large to get better before this would be viable. When we started Black & White, most people had 32MB of RAM in their PCs. The game requires 64MB, but that's commonplace now. So, if you like, we aimed beyond the horizon, and the world rotated and caught up with us so we still hit our target.
I still have those original screenshots, and I still like looking at them. We wrote a book called The Making of Black & White, and from reading that, it's clear that we went from a bunch of bizarre ideas linked by the concept of supreme control to the best game I have ever seen.
The final product.
4. The artificial intelligence. The Creature AI, as I have mentioned, is absolutely spot-on. Richard Evans worked tirelessly on this, and it became something that surprised even him with its flexibility and power. The AI isn't just restricted to the Creature. Every villager in the game has it as well, and they are all different in their wishes, desires, motivation, and personality. Because there is no upper limit to the number of villagers you can have, we had to cap the AI slightly by giving some of the villager control to the Village Center, which acts like a hive and farms out some of the cooperative elements to the people. We couldn't have them interrogating each other, so this central control means that they do work as a unit but can retain their individual characteristics. This makes the game much faster and still gives them minds of their own.
The Creature himself is an astonishing piece of work. Once he starts learning, he forms his own personality as he goes, and no two players will ever have the same Creature. The complexity is kept to a minimum to keep him fast, but we managed to steer completely clear of using random elements to make him seem like he has a mind of his own. And there is nothing in the game that you can do which you can't teach your Creature to do. It's true to say that the Creature mirrors you and your actions, so in Black & White we've got a game in which part of the game itself learns from everything you do and tailors itself to you.
5. The way the team came together to make Black & White happen. This is Lionhead Studios' first project, and everything started from scratch. The people, the software, and the working environment were all new. Although this was exactly what we needed to do a game so fresh and diverse, it also created problems which I was delighted to overcome. The lack of any precedent meant that things took a lot longer than they should have, and the open-ended nature of the game throughout much of its development meant that team members were limited only by their own imagination.
But the nice thing is, every member of the Lionhead team gelled brilliantly, and although I know we picked the very best people, there is an element of luck in whether they can all work together so well. We certainly lucked out with the team, and every one of them contributed massively to making the game what it is.
The last few months of the project were the hardest any of us has ever had to work, but thanks to the people, they were also some of the most fun months we ever had. If nothing else, we'll always remember the time we spent closeted together making Black & White.
And I'll never forget that without the right team, this game never would have happened. It's as simple as that.
Page 2 of 3