Warlords IV: Heroes of Etheria Wrap Report

Best Decisions The three best decisions were starting again from scratch, the combat system, and what I call our Fat Design Phase.

I think if we had not started from scratch, then Warlords IV would never have crossed the finish line. It had been in development a long time, and the codebase was a little "muddy" by the time we took it over. This is partly a matter of me knowing my own limitations - I was Lead Programming the project as well as doing design, and I knew that I wasn't going to work well with the old codebase - it was very, very heavily object-oriented, and my preferred method of coding only uses this technique in moderation.

The Combat System was important. It is, in many ways, the focus of a Warlords game, and it needs to be simple, fast and elegant. It sounds very attractive to say you've got a "full tactical combat system", but the reality of the situation was that having this system would not only have added eights weeks to the development time, but it would have made missions take 15 to 20 hours each. In today's market, a 15 to 20 hour strategy game scenario is not acceptable. Now, I'm sure there may be solutions within the game design that can solve this problem, but I would have guessed that we needed four to eight weeks more to iterate through and tweak it! So, moving to the simpler Speed Tactical System for combat that we finally settled on kept the project on track for us, and actually turned out to be a whole lot of fun!

We traditionally use a nice Fat Design Phase at the beginning of our projects. We design all major sub-systems before we touch anything except the engine code. Our theory is that if you know what you're going to code, it is a whole lot easier to plan for it. Even though it was tempting to dive straight in and start coding, we held off for almost four weeks while the game was planned out. On a normal 18-month project, I wouldn't call this a Fat Design Phase, but on a six-month one, it certainly is. Planning the game out and iterating through our design let us remove some unnecessary features from the game and saved us time towards the end of the project as we headed into crunch time.

Key Strengths There are three key areas where I feel we excelled are design, code quality and multiplayer.

The design on Warlords IV was very tight and elegant - a hallmark of all Warlords games, I believe. Everybody on the team contributed good ideas, and they were sifted for the ones that were best and most consistent with the franchise. While the game may contain certain small imbalances, they are easily fixed by tweaking some numbers... the actual systems within the game are nicely designed and work well together. I feel that we kept our eyes on the goal for the whole project and never lost sight of the game we were trying to make.

Our code quality was also very high. We achieve this by having regular code reviews and scheduling one week out of every six for just fixing bugs and problems. I know code reviews sound boring and wasteful of time, but they are invaluable to any team larger than two or three people. I have witnessed many teams leave a bug in place saying "we'll fix it in beta". I think this is insane! If you get two or three really nasty bugs happening at the same time, each one can end up making the other harder to find - it's very important to fix them reasonably soon after finding them.

Finally, the multiplayer on Warlords IV worked really well. We integrated Internet, LAN, hotseat and play by e-mail modes all into the one system. We have a lot of experience with multiplayer systems, and a few years working on RTS games has taught us many things. We made all areas of Warlords IV, even the campaign, available for multiplayer. The flipside of this was that turn-based strategy games are often not played in multiplayer, so many people will probably never experience it - it does require a little bit of patience if you have a slow partner / opponent. However, we've had a blast playing the campaign multiplayer (both via LAN and PBEM) so I speak from first hand experience when I say "give it a try" - it's a whole lot of fun.