He’s one of the most famous game designers in the world, but that doesn’t mean he’s always right. Peter Molyneux talks to Develop about the mistakes and lessons learnt that have put Lionhead back at the top of its game…
Let’s start with the ‘Justice’ concept artwork that leaked on the internet last month – how does that tie in with your GDC talk about experiments?
This talk I’m giving is all about how we access creativity through doing experiments. These aren’t just wild crazy ideas I have in the bath; they’re the result of real due diligience experiments that we’ve done, and we’ve gone through a lot of those.
You know, I think… you don’t always work on the first thing you think of, and it’d be inappropriate to do so. The better thing to do is to lay these things down and say things like ‘how would this look’, ‘how would this feel’, ‘what would I experience if I was playing this’, so that before you end up spending huge vast amounts of money – which is the way we used to do it – you try and get at least get half-way to knowing if you’re making a good experience.
We do some art concepts, we do gameplay concepts, and we sort of mix those together to see if something’s going to work. It’s funny that it got out there – I understand how it did, but it’s a shame, because there are confidentiality issues.
Is it something you’ve always practiced, or is it a new initiative?
The truth is, I think I’m famously awful at developing games. Before, I’d walk into the office, wave my arms and say ‘I’ve just had a cool thought’ – usually after severe alcohol abuse – and that lead us to spending a lot of money very foolishly on things that weren’t going to get anywhere.
Quite a while ago now, we sat down and thought, well, this is ridiculous – we can’t keep this notion that game development is a purely creative process, and that you have to build it to be able to see it. There’s got to be another way.
The first thing that we did is say, right, we need to do more work upfront it design and concepting, and that means less iteration further on. Because when you’ve got a team like Fable, which was around 100 people, you can’t experiment with that many; you’d be spending mad amounts of money. You just can’t do it.
So now, a lot of our design decisions are made when the team is small so that we can change our minds, because the thing about design is that you need to change your mind. You can’t come up with one idea and then just expect that idea to be perfect – it’s never going to work.
I’m a great believer of getting the ‘wrapper’ right up front – how you describe the game to people, how people describe the game to each other, and getting the mechanics and look and feel of it right – and that’s a lot to do with experimentation. We do that a lot. In fact, we do it so much that people get a bit panicky and ask when we’re actually going to go into production because we’ll keep on going around in that loop until we’re happy with it.
Was this process applied to Fable 2?
Yeah: the big things – the breadcrumb trails, the dog, the one-button combat – are all the result of real iteration, experimentation and concepting that we did. If it wasn’t for that, it just wouldn’t exist. So, we had a one-button combat demo three months after we started thinking about the game.
I’m not saying it’s the perfect way to work, but one of the things that Lionhead really wants to do is innovate, and challenge the fundamental foundation stones that we think of as a given in the industry – things like death mechanics and pause screens and minimaps – and if we’re going to do that, we can’t do it in a purely creative way.
If you asked me what I was most proud of in Fable 2, it’s not actually any of those: I’m most proud of the process. If I’m honest, on Fable we just burnt people’s lives; we destroyed the team. Week after week, month after month, they worked 50, 60, 70, 80 hour weeks. It destroyed their lives and destroyed their marriages. You just can’t do that anymore. You can’t do it.
How much of that ‘destruction’ was down to pressure that you put on the team?
What I used to say is, ‘Look, I’m working 120 hours a week – I’m only asking you to do half of what I do!’ But you can’t do that anymore, because you’re asking people to make a choice not between whether they go to the pub tonight or not but whether they go home and see their children.
And so because of me, and the way I used to sort of lead from the front and work harder than anybody else, it used to semi-destroy people’s lives. So I’ve tried to structure Lionhead so that creativity is really important, but it isn’t obsessively important any more. That the voice of production and the reason of finance can be heard at an equal level as an idiot like me shouting from the rooftops, which is very much what I used to do.
So looking back at Fable 2, yeah, we crunched at the end, but it was only at the very, very end that we used up people’s whole weekends. There’s a few exceptions, but a lot of Sundays this studio was empty, which I thought was great.
Everything’s important when you’re making something big like this – the team, the morale, the quality of their life, how much money you’re spending, the game features you’re working on – all of that comes together to make a successful product. It’s this holistic thing. I used to think that, if it’s not good enough and you’re working 60 hours a week, you should be working 70 to make it good. But that’s just such a wrong way of working, because you’re going to end up making huge mistakes.
At around about the same time, there seemed to be a consolidation inside Lionhead as you went to focus solely on one product. Why did that come about?
Well, we had this absolute nightmare of a time around the latter half of 2005 because of mistakes that we had made – and when I say ‘we’ I mean ‘me’ more than anyone else – which meant we had to slip Black and White 2. We’d made a mistake with the game, and essentially we were building a game that was far too big.
We also had to slip The Movies because of other design problems that were there, and Fable: The Lost Chapters for PC and Xbox were scheduled to be coming out too. It meant that within three weeks we had four SKUs to get out of the door.
Now, in order to achieve that – which we had to do because of lots of financial and legal reasons – we had to swell Lionhead to about 240 people, and that meant that we were in about seven different offices around this research park. The amazing thing is that we actually managed to get those three games out on time, but if I’m critical I would say that the quality did suffer, and we never want that. When we got through it, we sat down and asked ourselves if it was important to make three or more games, or to make one or two games to the best that we could. The answer was pretty easy.
I really, really hate letting anyone go at any time, but we did for logical reasons have to say ‘this is not why we established Lionhead; this is not what we stand for’. And so we had to shed some people. That was this awful, terrible time, and going through meeting after meeting having to talk to people about why they weren’t staying was really, really tough. I wouldn’t want anyone to go through that.
But when you come out of that, you realise that, yeah, the reason we did this was because we want to make great products in the future. But it was tough. It’s good now, in that it sort of modifies the way I think about things on a day-to-day basis, because I never ever want to go back there again.
Part two of this interview will be published tomorrow.