GR: If you could ask yourself a question about environmental design, what would it be? (Please answer that question, if it’s interesting.)
DP: I think the question I’d ask is, “Can you talk about the design process for making a skirmish map?” My answer would be:
With Halo Wars, we did something new for Ensemble. We set our random map generator from the Age of Empires games aside and really set out to make the best looking strategy maps ever made. We wanted a more crafted experience on each map because that was better for the fast action pace that we wanted. Random maps are super cool, but they have a few visual limitations and lead to slower overall games.
So, we had to come up with a whole different process for making skirmish maps during Halo Wars. Skirmish maps are the maps you replay over and over again, over Xbox LIVE or solo against A.I. opponents, so we know they have to be very balanced/fair and have classically fun elements. That last point is an interesting one. Some maps are fun once or twice, but lose appeal after a few plays. We needed maps that would endure years of play with Halo Wars.
As it turned out, the big solution was to start with a simple map that we could prove would be fun on paper. We blocked that map out quickly in our editor. The core design team would iterate that map for awhile. We found out that many of those “Oh, that’s gonna be hella fun” maps from the paper phase turned out to not be so fun when we played them. By the end, though, I think we found a groove where we did get a lot better at predicting which elements of a potential Halo Wars map were going to fun versus which ones were riskier to attempt.
Anyhow, after we had a good map from the internal design playtests, we would do two things. We’d start bringing the art team into the loop and we’d put the map out in larger playtests. Art would begin doing concepts for how the map would really look compared to our sometimes hideous design art skills. The larger playtest would, hopefully, validate our early thoughts on the fun aspect of the maps. These two phases of the map process could take many months to complete.
Once we’d made it past both of those phases with a map, we would give it over to the Art team to begin the work of truly sculpting the map, building custom objects for each map, etc. This took a while in the first half of production, but it got better at the end as we had a library of assets built, the tools got some much needed bug fixes, and people just got more efficient with the whole process.
After the skirmish map came back from Art, there would inevitably be a set of changes that Design had built up. We’d make most of those changes, but some would get scuttled because it would cause too much re-work on the map. Once those changes were done, the map would go into “bug fix” mode and we’d just touch up errors as the internal testing at Ensemble and Microsoft found problems.
GR: How much does environment affect the outcome of combat? Does it mostly have to do with taking cover and using defensive towers or is there more to it than that?
DP: The environment affects combat quite a bit, even though it does vary map to map depending on what elements are on a given map. Units cannot shoot through terrain, so some of the maps with lots of cliffs reward the players who can manage their units better in combat. If you’re not one of those players, you might be forced into more of an air unit build-up on those maps since air units negate a lot of the terrain line-of-sight restrictions. There are also scads of cover spots on various maps; everything from crashed Pelicans to sniper towers. If you see one of those, getting in it can completely turn the tide of a battle since infantry gains huge combat boosts when garrisoned. We also have some more interesting map elements that affect how combat turns out. One of the maps has a central Mega Turret. If you fight through the rebels guarding it and take it over, you can hit just about anything on the map when it shoots. That’s obviously going to affect how combat plays out.