Chilton started off the talk discussing the philosophies behind rejuvenating sequels and updating MMOs, saying that it is relevant to look back at an existing IP and freshen it up. As this was a GDC talk and aimed more towards learning from and understanding game developer's trial and error, Chilton wanted to use World of Warcraft examples of reintroducing and updating content in a broader way -- help developers understand how to keep the heart of the original while successfully iterating on the franchise.
There were three different spotlights during the talk -- the Desolace redesign that Chilton didn't particularly enjoy, the Westfall redesign that Chilton loved, and a mechanics discussion about revamping the talent system. The philosophy behind creating Cataclysm began during Wrath of the Lich King development and aimed to move the product forward with the same team of people. Multiple teams was not something Blizzard wanted to employ, as a single team ensures consistency from product to product.
WoW's content was aging, and by the time Wrath rolled around, the game was four years old and it was showing. The content from The Burning Crusade was better than the vanilla content, and the Wrath content was better than The Burning Crusade, but the old stuff still remained in full force. Chilton discussed the donut theory, in that you have to capture the hardcore demographic while still catering to the broader audience outside. (There was a picture of a donut on the screen.)
Blizzard holds true that leveling content is important -- you can't just create a new expansion for the highest level players and expect a great result. Leveling content affects and benefits most players regardless of their play patterns, and Chilton said that if you played casual or hardcore, the leveling is still part of the experience, just at a different pace. The game becomes inherently replayable, then, as many players re-roll new characters, making leveling content meaningful for even max-level characters.
The core WoW systems were becoming bloated and too complex over time and needed to be pruned. The trend in sequels is to be more complex and "up" the last iteration. Chilton wanted to move away from this idea. He discussed the "one-third" rule, where you have one-third old content, one-third improved content, and one-third new content -- of course, these are hard numbers, but he's speaking in terms of the design philosophy. You have to capture the original's special feeling and improve and add on intelligently.
Changes and updates to zones
The world aspect of World of Warcraft is, in Chilton's mind, WoW's greatest and strongest asset because Azeroth is a charming place that draws people in. The greatest area of dangers to Blizzard were that it had to retain the soul of the original game and that when players came back to Azeroth after the Cataclysm, it had to feel like fundamentally the same world with changes. Cataclysm could not make Azeroth feel like Azeroth 2.
Zone content and mechanics were showing their age. Chilton copped to the fact that the genre was new and, in the beginning, all the quests were a little dreary and fell into one of three categories -- kill, collect, or Fed Ex. Breadcrumb stuff was important to get you from place to place, but the quest hub dynamic just wasn't there. Blizzard wanted to mix in the Lich King stuff, with cutscenes, vehicle mechanics, and more fun things to do.
The flow of the original game was also lacking. Quests would send you on immense journeys all across the world only because it sounded good on paper or a good idea at the time. Within each zone, the designer was just exploring whatever he wanted to add in, in a fairly haphazard way. This was the beginning, after all. Quests ping-ponged you around the world because quest design was missing that crucial top-down design approach.
The philosophy, then, had to be revamped. We all remember the now infamous shot from BlizzCon of the zone revamp progression and how Blizzard was going to make changes to zones based on priority. Some zones would get more changes than others. Eventually, Blizzard realized that development issues continued to cascade, as one thing changed, everything came with it. Also, redesigned zones had to be handled by people who remembered what made the original zone so memorable. You couldn't just hire a designer off the street to revamp something so important and have it retain the heart and soul of the original place.
Quest and zone flow design changed dramatically for Cataclysm. New flowcharts and design storyboards were created before pen was put to paper. This top-down approach found inconsistencies more quickly and more efficiently and allowed parallel teams work in concert. Chilton said it was difficult for an item team to come in during a design phase or the creature team to deal with a zone that didn't yet have its flow set up. There were many new advantages to a visual depiction of the flow.
Chilton was also worried about how to represent the Cataclysm best in the revamped zones. He told an amusing anecdote about how Blizzard wanted to avoid the "volcano in every zone" issue -- no one wants to wander into a new zone and say, "Look, there's the volcano and there's the big crack in the ground."
Chilton's first zone example was Desolace, once a monotonous, boring, and barely accessible run-fest that needed a huge revamp to be cool, fun, and relevant. The idea originally was to have a Burning Legion presence and have the Cenarion druids begin Desolace's healing in the center of the zone.
The problem was that Desolace lost its heart. The place was supposed to be a barren wasteland, obliterated by the first Sundering and never healing. The centaur war in the zone felt shallow, and the landscape's settlements were oppressive. Also, the demons were there, with portals that they had emerged from, but just sort of stood around doing nothing. Chilton said that the demons came out of the portals, stood around, and would remark, "Maybe I should just go back through?"
The new version of Desolace fixed the questing issues and travel but killed the Desolace charm. But Chilton lamented that the redesign to Desolace wasn't the best thing. Rather, he now would have wanted a more extreme Desolace, one that had its terrible aspects accentuated by the Cataclysm, rather than a regrowth. The soul of Desolace was gone, and now the zone was green and happy, just like Feralas to the south. The transition from one green zone to another was not the best choice, he mused.
Changes in Westfall
Westfall, on the other hand, was one of Chilton's favorite redesigns because the heart of the zone was intact while changing the flow of the story and the zone for the better. Cataclysm was felt all over the zone, and while the terrain did not change very much, the play-out of the story was perfection. The Defias storyline hearkened old players back to a Westfall of old but engaged new players with a mystery to solve.
As one of the first zones Blizzard designed, Westfall has some of the oldest design flaws and mechanics. Sentinel Hill was an inadequate quest hub, only having a few quests, an inn, and a guy to sell you milk and bread. The monotony of the zone's environment was not as big of a deal, since it was a smaller zone, and you were in and out faster.
Chilton feels that the Westfall approach was the better one. A more streamlined experience retaining the soul of the original was a huge feat.
Chilton's final example of an issue that Cataclysm had to solve was the talent trees. In the original beta of World of Warcraft, talent points were the #1 concern of players because they wanted more character customization to feel different from other characters of the same class. A rudimentary system of adding attribute points and damage was transformed into the talent system we know today, all 6 months before shipping. Choice was important for players for longevity of play, endgame tinkering and experimenting, and to make you feel different from other characters.
The talent system became bloated over time as expansions added more to the system. Blizzard knew it would happen in BC and Wrath and made the decision to revamp the whole system for Cataclysm.
The problem was an illusion of choice. More talents didn't necessarily mean more options, because cookie-cutter builds became the norm. Rather than have 10 choices, with 7 being suboptimal and 3 being viable, Blizzard pared down the talent system to the viable builds only, giving players the choice of role, rather than the cumbersome fake choice that accompanied the bloated talent system.
New players were scared when they opened their talent trees, and returning players just gave up on them. Chilton showed an amusing slide of the priest ability Shadowform's tooltip, which was over 5 sentences of explanation of what the talent did. Things had to change. The large possibility of space was deceiving to players.
One idea Blizzard had was to let the game choose talents for the player until they were comfortable with the large trees, making their choices manageable. This was nixed, however, when it became less of a choice and more automated, making Blizzard feel like they were creating content for nothing.
One of the most interesting moments of the talk came when Chilton discussed talent systems that do WoW's talents better than WoW. If he had to go back and do it again, the system would look like something resembling Modern Warfare 2. Modern Warfare 2 gives players a constrained set of easily understood choices, yet with a huge range in possibility, playstyle, and customization. The confusion of moving down a skill or talent tree is removed. You select your equipment, gun, and perks -- that's it.
Chilton also warned other MMOs that are doing talents in the WoW-way: You will inherit our problems, so think twice.
A word of caution to developers
Chilton wrapped up his development talk with a word of caution to developers, telling them to have a deep-seated grasp of what was great about the original content and iterate only where you have to. Don't stray too far from the rule of one-third, and pick your battles carefully -- optimal design isn't necessarily the optimal design.
Finally, Chilton answered one excellent question afterwards. An audience member asked whether or not Blizzard thought about rolling in every expansion with the retail box of Cataclysm, making the game more accessible to players instead of having them go out and purchase three games to be up to speed. Chilton responded that he was a fan of removing the barrier to entry and rolling in all of the expansions into Cataclysm, but it wasn't his decision to make.
Thank you very much, Tom, for giving such a meaningful talk and intelligent discussion about sequels, iteration, and pitfalls along the way.