As part of our World of Warcraft: Cataclysm post mortem series, we sat down with World of Warcraft Lead Systems Designer Greg "Ghostcrawler" Street to talk about his thoughts on World of Warcraft: Cataclysm.

Q. What were your main goals going into Cataclysm?

A: Westfall was a seven-year-old zone with seven-year-old trees and seven-year-old quests. It naturally felt old. It felt tired for players going back to level up an alt, and it wasn't inspiring for new players coming to the game. We just felt like it was time to give all of those old 1-60 zones some attention again. Beyond that though, we wanted to update the classes at low level as well. The spell flow, by which we mean the level you get certain spells, just hadn't aged well. You would get some very group-focused buffs at low level and some powerful leveling tools at high level, which would have made more sense reversed. There were some specs that just weren't functional at low level before because they lacked the damage abilities or tools to effectively solo. Similarly, we took a look at all of the quest rewards at 1-60 because some specs just didn't have adequate itemization to support them.

A second goal, from the systems design point of view, was to improve the class talent trees. We thought the trees had become bloated with filler instead of legitimately interesting talents. We also embraced the notion of class specialization to a much greater degree, by letting you choose your spec formally at level 10.

We also knew we needed to provide more content to players focused on maximum level, which meant we couldn't just re-do 1-60, but we had to provide questing zones, class mechanics, and new PvP and PvE content for players who would be at level 85 too.

Finally, we wanted to deliver all of this content more aggressively. We know players can only wait so long for something new to do before they start to get bored. This has been a goal for some time, but it has been a challenging one for us. When you compare the graphic fidelity of a raid like Firelands to an older raid like Molten Core, you can imagine how it takes both more time and more people to make a raid these days. That's exactly the opposite of what we want to be doing though, which is providing players content at faster rates.

Q. What do you think worked best in Cataclysm?

A. We're really happy with the 1-60 revamp. Each zone looks amazing, we improved their quest flow, and they all have a story that has a (hopefully) meaningful climax, often with a blue item reward. Zones that didn't have much going on before have an actual plot now, many of which are related to Deathwing's return. We also did a better job of integrating the dungeons in a zone into the questing experience for that zone, so you feel like you have a good reason to explore them.

We really like how having players choose a spec at level 10 worked out. I'd say nearly every single design decision we make ends up being at least somewhat controversial in that some players agree with them and some players disagree with them -- that's just the reality of having such a large and diverse player base. But choosing a spec at level 10 was as close to universally acclaimed by players as anything we've ever done. It just works. You get a meaningful choice early on, and powerful, useful, and fun abilities to go along with it. It leads to each spec having a stronger sense of identity, even at higher level.

We're pretty happy with the level 80-85 content that we offered. The zones looked great and the stories were good. We offered several new dungeons, raids, and Battlegrounds. Late in the cycle of Cataclysm, we introduced Raid Finder, which provided a new type of content to players who historically weren't raiders. We're at the part of the lifespan of the game where some original features no longer have the cachet they used to -- you can only roll up so many alts, and by this point you might very well be done with achievements or convinced yourself that that type of gameplay isn't for you. When we can offer a whole new way to play the game -- in this case provide raids to non-raiders -- it's a big win.

Transmogrification is another one of those features -- it opened up an entirely new avenue of gameplay. One of the great things it's done, aside from giving players more tools to personalize their characters of course, is make a lot of old content relevant again. Now players are doing old raids and dungeons looking for Transmogrification pieces, and that's really cool.

I could name a few smaller features we thought worked out as well. The Justice / Honor badge system in Cataclysm cleaned up the crazy system from Lich King. All things considered, we're happy with the healing model. We encountered issues with mana being in short supply at lower gear levels and conversely too abundant at the higher levels, eliminating much of the challenge for healers when the content is supposed to be the most difficult, but overall the model did what we wanted, and we'll be refining it in Mists.

Q. What didn't work out as planned?

Everything else! Seriously though, we tend to be our own harshest critics, so it's actually easy for us to point out things that didn't work out as expected.

While zones like Uldum and Deepholm look fantastic, they didn't fit together as well as we'd have liked. In the planning phases, we didn't think that having scattered end game zones would be a big deal. It turned out to feel a lot weirder than expected. Players ended up teleporting to nearly every destination, and it gave Cataclysm a disjointed feeling, detracting from that feeling of exploration and discovery. We learned that giving players a land to explore, a sense of place, is valuable. Ultimately, the scattered zones and the portals both served to kind of shrink the world, when we want to make the world a place you want to go out and be in. We're definitely looking forward to getting back to a continent in Mists. We underestimated how important that was.

In addition, while we liked that each zone has a story, questing ended up being too linear. It didn't feel like you could fly into a zone, find some quest givers, and explore. Instead, you kind of had to start at the beginning and follow all the quests to the end, and if you didn't like a quest, well, you had to stick with it to get to the next one. We want zones to have an identity, flavor and a story, but we don't want to railroad players through a zone either.

The difficulty at which we pegged our heroic dungeons and raids was controversial. They were designed to be about as tough as the dungeons were back in Burning Crusade, but the game has changed since then. Coming out of Lich King, we'd gotten the message loud and clear from players that they wanted tougher challenges. They liked the convenience of Dungeon Finder, but they missed using their crowd control and survival abilities and having to strategize about how to beat a given encounter. We designed the Cataclysm heroics with that in mind, and the players who wanted challenging content were thrilled.

The problem was that we had this whole group of players who felt like they couldn't make any progress on their characters. Even if they wanted to end up raiding with their friends, they couldn't earn the gear they needed to get into those raids (especially in the absence of Raid Finder). I don't believe that the instances were too hard; it's obvious there are players who enjoy that content. I believe the problem was that there were no alternatives. With such a diverse community, the goal is to have experiences that players from all over the spectrum can enjoy. We don't want to shut anyone out. So, we're addressing that with Challenge Modes in Mists. You'll have normal and heroic mode dungeons, and then Challenge Modes, for players who are looking to prove their mettle. Likewise, you'll have normal and heroic raids, and Raid Finder for players who don't enjoy wiping on a boss week after week until they can master it.

While choosing a spec at level 10 felt great, we weren't very happy with the rest of the talent tree overhaul. We definitely pruned some dead wood from the trees and got rid of some talents that weren't a lot of fun, but players felt like they weren't getting anything out of the bargain. Having simpler trees is a good goal, but it would have felt better if players felt like they got something cool in return for losing some boring fluff. Unfortunately, as is the case with many compromises, this one didn't fully solve the original problems it was intended to solve, while it created new ones.

Fundamentally, taking into account what we've learned about talent trees over the years, we've come to the conclusion that the talent tree model where you pick up tiny performance increases here and there (and where there's, mathematically, nearly always a 'right' answer and a 'wrong' answer) is not a great model. The Mists talent design is a major revamp that should fix this problem once and for all. Talents should be meaningful game-changers. At absolute worst a given talent may be the right one only situationally, and at best, players will have a lot more customization to make their play-style stand out. Furthermore, the fact that you'll have more flexibility to change your talents should help keep gameplay fresh, even with that character that you play most often.

I feel like I should mention Abyssal Maw again. As with many cancelled features, it somehow took on a life of its own in the minds of players. Believe me, though -- you just don't cancel things that you think are going to be awesome. It was three bosses inside Nespirah, with no unique art. The reason it was originally appealing to us was because we had so many Vashj'ir assets that we could use. But by the time it was time to do the work, we felt like we (and many players) had Vashj'ir fatigue. Now don't get me wrong -- I loved Vashj'ir. I was an oceanographer, remember? Vashj'ir delivered on the promise of an underwater zone, but we feel like most players were ready to be done with it by the time they had quested through that. (Individuals will feel differently -- it's that diverse player base thing again.) Firelands received a lot of new art, from bosses to environments, and we just didn't feel like Abyssal Maw was going to compete. Who knows, though! We haven't totally given up on the idea of cool underwater experiences, so maybe there's potential we'd visit it again someday. (For my money, the zone I am personally saddest about cancelling is not Abyssal Maw; it was the Azjol-Nerub quest zone in Wrath of the Lich King.)

Speaking of raids, we also weren't particularly happy with how accessible legendary items became in Cataclysm. Multiple characters in a single raiding guild were getting, and worse, expecting a legendary weapon. Legendaries are supposed to be rare and exciting, not a bar you fill up like some reputation grind, and certainly not something you feel entitled to get because it's "your turn." Dragonwrath in particular was usable by a large variety of class specs, which coupled with the guarantee to completion, just made them too ubiquitous. In the future, legendaries will be more legendary, perhaps so much so that not every raiding guild will have one. In that model, there might be those who almost, but not quite, complete one, but there will also be those who finish one and feel truly honored.

I have mixed feelings about Archaeology. I feel like it's a good addition to professions and offers more, and more varied, gameplay than our existing professions. Still, it's clear that some players wanted more. We wanted Archaeology to be hard to complete. We didn't want it to be one of those professions you can max out by buying up mats at the Auction House. But random reward systems whose long-term goals are more interesting than the short term ones can feel grindy. Archaeology had too much travel time. It could be punishingly random, especially for players who imagined that it would be a guaranteed delivery mechanism for Zin'rokh (which was never the intention). Players missed a lot of the lore, which was delivered in the Archaeology journal and not as part of the survey or digging experience. We think the Mists of Pandaria expansion will be really good for Archaeology. Players will be focused on a couple of new races on a single continent, so travel and randomness will be reduced automatically, and leveling Archaeology should be a bit more convenient since there will be more opportunities to dig at a single site. We have other tricks up our sleeve too.

Q. What lessons have you learned and what are some of your top goals for Mists of Pandaria?

There are four big goals for Mists:

Get players out into the world. We don't want to totally eliminate convenience, so it's fine to queue for some features from capital cities, but we also want players to see other players out in the world, questing, trying world bosses, engaging in PvP, and just travelling from place to place.

Give players plenty to do. It's a sad feeling, and a real failure on our part, whenever someone says "I want to play WoW this evening, but I just don't have anything to do." Like I said above, achievements and alts were great in their time, and we'll continue to support them, but we understand the need for new ways to play as well. The new expansion will have entirely new systems, like scenarios and challenge modes. We are designing the initial zones to have features similar to the Molten Front daily area, so you don't feel like questing is something you finish at level 90 (and so you don't feel like daily quests are synonymous with 'boring' or 'grind'). We want to make the Pandaria factions interesting. We want Exalted to be something you earn for bragging rights, not something every player has. We are adding a lot of mounts that will be hard to get, and awesome-looking armor that you'll want just for transmogrification. We're considering ways to let you increase the number of Conquest points you can earn per week or a way to translate questing into bonus loot from instances. We want to hide lots of cool little things all over Pandaria. Some will offer your character more power and some won't. And if you really like achievements and alts, well hopefully we've got you covered there too, with account-level achievements and a new race and class.

Appeal to a broad audience. I'm always surprised by the number of players who want the game to be easier and the equal number who want the game to be harder (and can't understand why anyone would disagree with them!) We approach the issue in a different way -- we think that what all of those players are really saying is that they want content for them. Message received. We'll be offering Raid Finder versions of all of our raids going forward. We'll be offering brutally difficult challenge modes of dungeons for players who thought the Cataclysm heroics were too easy. We're experimenting with some tricky boss encounters for players who loved the hard-mode Ulduar achievements. We want to provide more cross-over between PvE and PvP, for those who are interested, so that it doesn't feel like you have to play two different games to progress your character. We want to continually add new Battlegrounds, so those players have fresh experiences to look forward to. We'll provide players with ways to upgrade their gear incrementally, while reserving tier sets for actual boss kills.

Get great content out faster. Enough said.

************

We hope you've enjoyed this entry in the Cataclysm post mortem series and that this has proven to be an enlightening opportunity to take in our perspective on what worked, what didn't, and some of what's coming. If you missed the chance, you can join us in looking back at Cataclysm by checking out the other entries in the post mortem series with Lead Encounter Designer Scott Mercer and Lead Quest Designer Dave Kosak.

Now it's time to look forward, since we have more to share about World of Warcraft: Mists of Pandaria coming on March 19. Stay tuned!