Master of Orion 3
1. Who are you and what was your role in the development of MOO3 ?
Bill Fisher, founder and President of Quicksilver Software, Inc.
I was the executive producer here. My role was a combination of high-level project management, technical oversight, and a few bits of design. As often happens in small, highly-motivated groups, we each tend to take on more than one role at various points in the process.
2. What was the most difficult part in creating Master of Orion 3 ?
The biggest challenge was controlling the scope of the game. When your mission statement is “simulate a galaxy,” it’s easy to let your imagination run away with you. I certainly participated in my share of grandiose concepts, some of which ended up in the game and some of which were justifiably killed off at various stages in the process.
What kept us grounded was that we knew anything we implemented in the game ultimately had to make logical sense from a gameplay perspective – in other words, the player had to feel that the feature was there for a reason, and that changes made to a particular setting would have a predictable result. It’s our belief that the game isn’t done when you’ve added all the features you want; it’s done when you’ve taken out all of the features that you don’t need. Thus, a lot of our efforts were directed toward looking at what we were creating (on paper and in code, depending on the situation), and taking a hard look at how useful and understandable each would be to the player.
3. What were the toughest decisions that had to be made during development ?
The toughest decisions were the places where we decided to cut something that we’d worked on for a long period of time. Several times, we had what sounded like a really cool idea, wrote a bunch of code, and then discovered that it just wasn’t going to work the way we wanted. It’s a lot easier to leave something in and keep pushing to make it work, but sometimes you have to “kill your own children”. That’s what it feels like at times, but ultimately we know we’ve made the right decisions on the game. When I look back at some of the choices that seemed so painful, I’m happy with every one. I don’t regret any of the cuts.
4. What was the funniest incident during development ?
We played a gag on Rantz Hoseley, our creative director. He had a very strong mission to ensure that the aliens in our game were more than simply “humans in furry animal suits” or anything else resembling cheesy sci-fi. And, because of the wonders of 3D modelling technology, we were able to achieve that goal rather handily. I really like the extreme variety of aliens in the game.
But Rantz was so strongly opposed to the concept of cheesy aliens in furry suits that we actually hired a person in a six foot animal costume for a party here about a year ago. We pulled a surprise on Rantz by bringing the character into the warehouse in the back of the office through a side door. The expression on his face was priceless, to say the least. We laughed for days.
5. Looking back, what would you have done differently ?
I’d have spent a lot more time on conceptualization of the user interface. We didn’t really nail down the look and feel as soon as we should have, because we were so involved with thinking about all of the zillions of possibilities in the game. I think we all agree that, if we’d done a prototype UI with just a simple galaxy and some empty menus, we would have realized a lot sooner that there were too many screens in the first design.
As a result, we’re designing our next set of products much differently. We’re spending the time up front to make sure that everything works together conceptually before we write a lot of code. That’s a little easier this time, too, because we have a complete set of tools to work with. One thing that definitely made it harder on MOO3 was that we were creating many entirely new elements, like our user interface toolkit, to replace previous generations of Quicksilver technology, and this meant that it was harder to know exactly what we were going to be capable of implementing.
The other thing we are going to do on the next titles is be much more proactive in killing off features that don’t work as early as possible in the process. Fortunately, we have a very outspoken group of folks here who honestly want what’s best, and I make sure that their voices are heard, even if they disagree with what we’re planning. Some of this company’s best ideas have come from folks telling me I was going in the wrong direction. As long as you’re willing to challenge your assumptions throughout the development process, you’ll end up with far better results at the end.
6. What makes Master of Orion 3 different from other space strategy games ? And if you were to pick one thing that’s the strongest part of MOO3, what would it be for you personally ?
Our top goals in creating this game were to break new ground in graphics, user interface, and artificial intelligence. I’m very happy with our achievements in each of these areas. I think these are the features that distinguish MOO3 from other competing products.
Visually, we’ve created aliens that are far more believable than what I’ve seen in the past. We have creatures that live in just about every imaginable environment, with complete back stories that describe everything from their political systems to their reproductive systems and preferred forms of manufacturing. Some live on energy, some live on minerals, and some live on even more interesting things. Add to this a very clean, elegant user interface that tells you what you need to know without overwhelming you with unnecessary visual detail.
The interface took a huge amount of work and a lot of refinement during the project. Our basic philosophy was what we call the “layers of the onion” approach – at the top level, you see a summary with only the information you need to make high-level decisions. You can always peel back a layer, though, if you want to find out more. And you can sometimes keep doing that for a number of levels, especially when dealing with the manufacturing or economics of a specific planet. We had a number of key goals: making sure that players didn’t get lost in all of the screens, making sure that what we displayed was exactly right for the specific situation, and making sure the game didn’t require a long read of the manual to get started. This is definitely the hard way to create a UI. It’s a lot easier to simply provide tons of raw or half-processed data and let the user sift through it all. Many business apps work that way. But this is a game, and it’s supposed to be fun. We spent a lot of time analyzing exactly how much data was needed at each point, and then providing it in as clean and attractive a manner as possible.
Good AI is an absolute must in a Quicksilver strategy game. It’s one of my biggest missions in this business, so I made sure we had plenty of AI programmers (a total of four or five people worked on AI at various stages of the game, with several others consulting). It was a massive effort, with over 20 separate AI modules performing duties ranging from placing buildings on planets to designing ships to searching out new planets. And that’s not all. We also set ourselves the goal of making the AIs “transparent” to the user – you can actually override individual decisions of the AIs, and they’ll pay attention to your selections while continuing to make intelligent decisions about other elements that you didn’t change. This was incredibly tricky, but it works really well. You can change one slider on a planet, add two items to a build queue, and know that your Planetary Viceroy will handle the rest of the details intelligently. That’s a real accomplishment in a game. Usually, AI players are pretty stupid, but these are really quite detailed.
7. What’s the next step ? Will there be extra’s going to be added or patches created in the near future or will you go directly to a new project (MOO4 maybe) ?
We have a number of interesting new projects in various stages of development. We are certainly talking with Infogrames about what might happen next, but we’re not at a point where we’re ready to finalize any plans.
Fortunately, the game is in very good shape already. We don’t expect to be needing a patch on the first day after we ship. I’m sure that the fans will have plenty of suggestions, and we will be watching the discussion boards closely to see what people like and what’s confusing, so we can make sure our next efforts are well targeted.