Ratchet & Clank Future: Tools of Destruction Lead Programmer Anthony "Moo" Yu discusses the challenges of next-gen development.
(NB: Membership is required, thus the whole article is posted here for n4g.com)
"There has been a lot of discussion about what makes next-gen games so challenging to make. It ranged from the complicated, multithreaded hardware architectures of the next-gen consoles to the impossibility of getting a green light on non-derivative titles. The next generation is finally here and I’m starting to think that we’ve overlooked something huge: what it really takes to make a great game is a good team. That’s not just a collection of people who are individually good. It’s a collection of people who are good at working together.
I’ve only been in the industry a few years now, but something that has become abundantly clear to me is that you can’t make someone create something awesome by force. For anyone who is in the industry and has talked to a programmer, they will quickly realize that how long it takes to program a task is inversely proportional to how much the programmer likes that idea. I’ve convinced designers that changing the speed on an animation is basically impossible because I hated how it played, but on the other side, I convinced our marketing team that I could program a playable character in three hours for the Extreme Makeover: Home Edition project. What’s important is making sure each person involved in the process believes in what they are doing. As professional as we would like to be, there’s no question that we will always do a better job on things we care about. It’s just the nature of human behavior.
The challenge then is how to get people excited about the things they’re working on. A lot of people would expect simply great ideas to be sufficient, but it’s really not. Ideas are entirely subjective and there really is no idea that is universally loved. If you walk around the office and ask people what their least and most liked gameplay mode is in Ratchet & Clank, you will not only get answers across the board, but you’ll also get plenty of people whose favorite segments are other people’s most hated. What it really takes is getting people invested in the ideas. The designers get the ball rolling, but we make sure that everyone helps to determine where the ball ends up. Whether it is implicit in the fact that designers can’t specify everything gives people enough room to put their own spin on things or the times where we actually set up brainstorming meetings to make sure everyone has their say, we do our best to make sure nobody is just a cog in the machine.
The thing that starts getting in the way is that the teams keep getting bigger and bigger, and even more importantly, team members become more and more specialized. Since each person is a smaller part of the big picture, it means two things. That person has a smaller amount of control over what they’re doing and a larger number of people need to agree to get something done. This becomes a big problem. If everyone gets disenfranchised with an idea, it’s doomed to soulless mediocrity, regardless of how talented the particular individuals involved are. Also, it becomes more and more difficult to put in random pieces of flair that really are a huge part of Insomniac games. On the previous generation, a piece of programmer art might actually survive to the final product. However, since everything is so much more complicated, there’s no chance that it would pass. Thus, all secret projects require a larger number of conspirators. We are starting to see these problems and only beginning to get a grasp on how to approach them.
We definitely have had a handful of success stories so far. It’s amazing how huge and the impact arbitrary changes can have on the development process. One of the things we tried out in pre-production was to have the entire pre-production team sit together instead of having them sit with their respective departments. It was an unpopular idea at first, but ended up being such a success, that even in production, we decided to intersperse the gameplay programmers with the designers and to move QA right next to us. Despite the fact that this does lead the gameplay programmers being constantly interrupted, it’s for our own good allowing us to learn of the problems earlier rather than having them all fall upon us in the eleventh hour of gold submissions.
Another thing that brings us together and keeps us motivated is external viewing of the game. We really do care about our work and the quality of our game and the mere thought of having someone outside of the company see or play it is a sufficiently strong motivator. Every month or two, we put together a focus test disc for random members of the public to play. It’s really astounding how much progress can be made in such a short amount of time when we know exactly what the target is and everybody knows their part. For things that are actually going out to the general public, like our Ratchet & Clank Future: Tools of Destruction trailer, it’s a whole other story. For the trailer that was posted a few months ago, we had to put it together many months before the game was going to ship, but still guarantee that it lived up to the game that we knew we were going to make. It was a lot of hard work at a time where we didn’t have a ton to work with at the time, but after seeing the press coverage and fan response, it was definitely a great experience.
It is my personal belief, though, that the only way a team will truly be inspired is if they know they are making a great game. There is always a concept of critical mass that I first noticed when I was working on Ratchet & Clank: Up Your Arsenal. There was a point at which it started feeling like a game. It was pretty bizarre. One week prior, we felt like it was a series of cobbled, mangled pieces of art and code, and suddenly it felt like it was something that you could have purchased off a shelf. With over a month left, that’s when we all knew we had something good. It raised the bar that we were aiming for and every time someone fixed something and made it better, everyone knew it was their turn to step up. It’s sort of a catch-22. As long as everything is broken, it’s easy to accept other broken stuff. However, once you have something that resembles a real game, it’s time to put your money where your mouth is. For every game you’ve ever played and thought, “Why didn’t they fix that?,” you now have the opportunity to fix it such that some other gamer doesn’t have to think the thought.
I definitely think the currency of great games is enthusiasm and it’s surprisingly hard to come by with the increasing team sizes and established properties. As a lead, I face all sorts of issues like keeping designers in control, trying to keep artists from checking in broken art, and finding programmers that match our team. The biggest challenge nowadays, though, is maintaining morale and keeping people inspired. And just in case anyone was wondering, fan mail definitely helps, doubly so if it comes with crayon drawings."