Today we continue our class Q&A series with Greg "Ghostcrawler" Street and the development team, in which we're taking a look at each class and answering some of the top questions brought forward by their communities. Next up, we take a look at the most asked questions from the druid class and find out more about the design philosophy and expectations for the class, as well as what may lie in store for it in the future.

Druid Q&A with the Voice of the Class Design Team, Ghostcrawler

Community Team: We’d like to start things off by asking a question that players often ask in regard to the very purpose of each class. In this case, we’re looking specifically at the druid, which has one of the widest ranges of play styles in World of Warcraft because of the variety offered from the different talent trees.

Q: Where do druids fit into the larger scope of things currently and where do you see them going from this point forward?

Ghostcrawler: Much like the paladin and shaman, the original World of Warcraft druid was intended to be a healer for end-game content. Bears could tank Upper Blackrock okay. But honestly, nobody took cats, bears or moonkin very seriously at level 60. In Burning Crusade, this changed a little. Bears could tank a lot more content, and might have even been overpowered in the endgame if it weren’t for Sunwell Radiance. Cats were probably underpowered. You still saw them, but they were there mostly for buffs or because they could slip easily from tanking to dps in a world before dual-spec. They were great for Zul’Aman for example. There were some great Balance druids, but still a lot of Oomkin. Our philosophy at this point in the game was that hybrid classes needed to give up a lot of their power, especially on the dps-side of things, in order to justify their versatility and massive buffs. In Lich King, we set out to change this and make sure all four types of druid had a PvE role in heroic dungeons and raiding. And yes, this meant four roles, because if we were going to let cats do credible dps or bears be able to tank anything in the game then we wanted players to have to commit to one extreme or the other. You can remain a half-bear, half-cat if you want to as well, but you won’t be as good at either extreme and my experience is players rarely choose that route.

In PvP the story is quite different of course. Once Resto druids got enough tools in Burning Crusade, they were very difficult to counter in Arena matches. Cyclone, Feral Charge and running and hotting made druids frustratingly slippery. We wanted to chill this out a little in Lich King, largely by pushing Feral Charge farther away and making Tree of Life a real option in PvP. Unfortunately, high tree armor and dispel-at-your peril hots have still kept druids difficult to counter in PvP. Feral druids have always had a place in PvP, but were never as dominant as the Resto druids. There is a little bit of a chicken and egg problem here because while some druids only want to PvP as Feral, others were happy to respec to Resto for PvP. Balance druids seemed even less viable than Feral, which is not ideal and something we want to improve. We tend to take a longer view on some of these things than do the players. We’ve had other balance issues that we’ve wanted to address in PvP first, and getting every spec of every class viable has had to take a back seat on occasion.

Q: What is it that makes them unique compared to all other classes?

Ghostcrawler: Forms is the big answer. The druid versions are more meaningful than other class equivalents in the game, such as Stances or Presences. One could argue they are too meaningful, because players sometimes don’t want to use any other form but “theirs." Perhaps the most unusual facet of the druid forms is that they use three different resource systems depending on the form, and these are not reset when they shift. A leveling druid can shift out of cat form to heal themselves, then shift into bear form while their mana regenerates.

Druids also have a couple of other interesting qualities. They actually fit four distinct roles into the class, even though two of the roles share a talent tree. They have some unusual utility spells, including Revive and Innervate. While druids no longer bring many unique buffs, they still pack a lot of them onto one character. Raids are still happy to get druids because of their benefit to the group. Druids also have unique travel abilities, from Travel, Aquatic, and Flight form, to the Moonglade teleport.

It’s also worth mentioning that druids can be only one race each on the Horde and Alliance side. They have the least racial diversity of any class in the game.

Community Team: Let’s take a look at the Balance talent tree. Eclipse is a crucial talent for players who are looking to perform a damage dealing role and invest in the Balance tree. There has been a lot of discussion though regarding the gameplay it provides as there is a lot of randomness involved. The buffs to a solar rotation were very well received, but the lunar rotation still has to wait for a critical strike to happen as well as an additional random proc to occur which can lead to some long gaps of nothing happening. Then when these finally fall into place players have complained about threat issues which force them to slow down or having to move to avoid something which ruins the procs they have and significantly hurts their overall damage.

Q: What are our thoughts on how Eclipse is functioning and do we have any plans to change how this talent works?

Ghostcrawler: The goal of the Eclipse talent in the first place was to give moonkin a more dynamic rotation that involved the player paying attention and responding to the environment rather than falling into a monotonous rotation. The Eclipse in 3.2 should less strongly favor one half of the Eclipse over the other since the cooldowns are independent. Ultimately however the problem we are trying to solve is that Wrath and Starfire are just too similar. In PvP you get a little bit of interest out of the fact that they are in different schools, but in PvE by the time talents are factored in, the two spells just become fairly quick (but not instant) nukes and it’s easy to math out which one to use and which one to ignore. Long-term to fix this problem we need to add another spell, separate out Starfire and Wrath from each other a little more, or make one of the other spells, like Moonfire or Insect Swarm, more dynamic. I’ll give a couple of example of caster rotations that “work” in our opinion: Destruction warlocks want to Immolate before they Conflagrate, Frost mages can proc a Brain Freeze and throw out a fast Fireball.

Community Team: Another aspect of Balance talent tree that has seen a number of discussions lately is the survivability of Moonkins while in a Player vs. Player setting. Many players agree that they have a very low survivability rate and that a spell like Typhoon doesn’t provide enough help to prevent classes like Death Knights and Rogues from doing some serious damage to them.

Q: How do we feel Moonkins are doing in PvP and do we have plans to improve their survivability?

Ghostcrawler: We don’t think Moonkins are quite there yet, though we’ll see how they look after 3.2 ships. Some of their problems are not limitations with the spec so much as they are ramifications in other parts of the PvP environment, such as some classes being able to burst them down too quickly. Of the casters, and as of this writing, only Frost mages really seem to be a potent PvP force, but it requires multiple forms of CC and escape mechanisms to get there. We really don’t want to go down the road of every caster needing that many unique tools – it homogenizes the classes and makes the mage tools less compelling. We understand some players are desperate to play Balance in PvP, but our priority is on getting underperforming classes viable before we worry about the second or third spec of classes that already have a strong PvP presence. We’ll get there.

Community Team: Players have also been discussing how they feel the overall playstyle of a Moonkin is just not very exciting. The best damage dealing rotation for them requires very few spells and a deep Balance spell like Starfall doesn’t help them very much.

Q: Are there plans to make any changes to the general spell rotation for deep Balance/Moonkin Druids?

Ghostcrawler: As I mentioned above, we do want to improve the spell rotation of Balance. We think the spells are interesting when considered alone (expect for perhaps Starfire and Wrath) but they don’t necessarily play together in interesting ways. You don’t try to save up a Starfall for example for great synergy with another spell. We have no problems with Starfall itself -- it is an AE with smart targeting that doesn’t require channeling. It is basically just bonus damage. Often players with a gripe about Starfall are wishing that it still proc’ed stuns with Celestial Focus or was a stealth remover. But those uses meant players saved Starfall for only those specific situations instead of using it when they needed extra damage, which was the original intent.

Community Team: We are going to switch gears from the ranged caster damaging druids to the up close and personal Feral talent tree.

A number of players have brought up questions regarding the over-arching goal of the feral talent tree. In the past they were able to change forms regularly and cast spells and then go back into a Feral form to continue playing. Once we merged their gear with rogues this playstyle was significantly reduced in effectiveness. As it currently stands, even with the low mana pool, they feel their mana regeneration is very poor in comparison to other hybrid classes like Shaman and Paladins which in turn continues to minimize feral druids’ ability to cast spells which they feel is pushing them further and further away from the hybrid playstyle.

Q: What are our intended goals for the playstyle of feral druids?

Ghostcrawler: The druid class overall is intended to be flexible, and we feel that it is. What we don’t want is a class that can do all things with a single spec -- do damage like a rogue, then tank if the MT falls down, then battle rez the MT and heal her back up, etc. That might sound like a lot of fun, but that’s because you’re fulfilling the role of half the raid all on your own (which means it’s less fun for everyone else watching you be a superstar). Every cat worth their salt will shift out to cast Rebirth or Innervate. But in order to justify cats doing credible melee damage, we felt like they had to give up some of their ability to tank, cast, and heal. Note than you can still take a more hybrid-focused build. Players don’t often do that though because they’d rather do one thing really well. Now I will say that long term something we’d love to do is get rid of shifting costs altogether. We want to see druids in lots of different forms -- more on this in a minute.

Community Team: In order to provide the best damage possible, the rotation for a Feral druid is quite complex. While this is okay in itself, it is nearly impossible to pull off in a PvP scenario so they end up using just Mangle, which they feel drastically hurts their value. To be the most effective they also rely heavily on someone else like another druid or an Arms warrior.

Q: What are our thoughts on the complexity of the Feral druid damage dealing rotation?

Ghostcrawler: If you want to do the best damage possible, you need to be able to master a complex rotation. This is one of the features that attracts players to the Feral spec. However, it’s also pretty forgiving. If you just Shred, you’re going to do decent damage. If you Shred and try to keep up Savage Roar, it’s going to be better. If you can also manage your Mangles and Rips and Rakes, then you have the potential to do very high damage (assuming you don’t have to move around much and can reach the target’s back).

Community Team: Feral druids have also complained a bit about their lack of utility in the PvP environment.

Q: Do we have plans to provide more utility for Feral druids such as a more reliable interrupt?

Ghostcrawler: Feral druids have Bash (which they can improve through talents), Maim and the Feral Charge stun. We think their tools are sufficient for PvP and there are other melee classes that can handle it in PvE. It’s not a strength of the spec for sure, but we’re okay with that. We don’t want all melee to be identical.

Community Team: Bear tanks have often felt inferior compared to the other tanks.

Q: How do we feel druid tanks are doing and do we have any immediate changes planned?

Ghostcrawler: We think bears have felt inferior because for a long time we basically said “You are designed to be inferior.” Sometimes old perceptions die hard. Bears are not inferior tanks in Ulduar and it’s possible their survivability is too high in 3.2.

Community Team: Savage Defense has been the source of a number of debates as players feel it is somewhat lackluster and doesn’t provide enough of a benefit, especially in PvP.

Q: Do we have any plans to change and/or improve Savage Defense?

Ghostcrawler: We’re pretty happy with Savage Defense for now. It accomplishes its goals, which were to make dps stats more useful on leather and to keep bears from hitting the armor cap so easily. You might consider it lackluster if you’re counting on it saving your life, but it does account for a lot of damage over the course of a boss fight.

Q: Are there any plans to change the functionality or cooldown of the Cat form charge to better align it with the Bear form version?

Ghostcrawler: We’re happy with the distinction. We don’t want to duplicate every ability since the druid has access to both forms.

Q: Can you give us the reasoning behind why the Bear form swipe is 360 degrees while the Cat version is only 180 degrees?

Ghostcrawler: The bear change was just a quality of life issue. A bear can wipe the raid if they can’t pick up incoming adds. The cat is much less likely to wipe the raid by not being able to Swipe all the adds. We don’t need for cats to be an awesome AE spec. We just want them to have something to do on big packs that everyone else is AE’ing, which is what cat Swipe gives them. On the other hand, if there is a perception that bears can’t manage adds, they aren’t going to be used on a lot of fights.

Community Team: On to the last of the talent trees now, Restoration.

Q: Can you describe for us what the intended identity and versatilities of the Tree of Life druids are supposed to be?

Ghostcrawler: This is a pretty philosophical answer here. If you only read one answer in this Q&A, you might want to check out this one.

The strength of the Resto druid is in heal-over-time spells. They can also do some decent single-target healing through Nourish or group healing through Wild Growth. The tree pretty much used to be a PvE mechanic since “rooting yourself” in PvP to benefit from ToL greatly limited the druid’s crowd control and escape abilities. We made enough changes to the talent to get trees into PvP in Lich King… arguably too well.

Druids overall have a strong niche. We are at a little bit of a crossroads with the Tree of Life however. We are currently wondering if druids sacrifice too much just to be as good as a healer as everyone else. What I mean is that if druids were good healers in caster form but great healers in Tree form, then there might be a decision there. However, we pretty much assume that healing druids are in Tree form nearly all of the time and balance around that. We don’t think it would be fair for them to be the best healers just for taking that talent.

In addition to having to give up utility in order to heal as a Tree of Life, we have become less enamored with druids locking themselves into one form. In fact, you really never see the basic tauren or night elf druid form (you know, the one that actually shows off the awesome armor art) because all druids are in cat, bear, tree, or moonkin form nearly 100% of the time. I’m not saying we would just cut Tree of Life from the game. It’s been around awhile and for better or worse, it’s part of World of Warcraft now. However, we could see taking the druid in a direction where shifting was much more common and easy to do. Maybe you only go into tree form for certain spells but leave for other spells -- this didn’t work previously because of the high cost of shifting, but in the absence of power shifting, we’d love to get rid of the costs completely. Another way to go would be to make Tree of Life form a cooldown, more like Metamorphosis. You shift into tree when you need a healing boost, but you don’t stay in it all the time. Now, I am totally waving my arms here. This is not the kind of change you are going to see in the next patch. But it is something we’re thinking about long term, and the kind of thinking we’d love to have more feedback on from the community.

Note to other healers: this is why you are unlikely to see any kind of “Holy form” ever. Giving up healing to do damage works okay. Giving up everything to heal is lame.

Community Team: Both Tranquility and Healing Touch are meant to be strong healing spells but are generally viewed as very ineffective due to their lack of flexibility.

Q: Do we have plans to improve Healing Touch or Tranquility?

Ghostcrawler: Tranquility is a fine spell. You can use it in situations where high amounts of damage are coming in on one group. Ultimately, we’d like to get away from the concept of groups within raids, but that’s not super high priority. Having a talent that promotes a situational spell on a long cooldown though is a problem (for any spell, not just Tranquility).

Likewise, we think Healing Touch is fine. You use it at low levels and you use it with Nature’s Swiftness. Maybe you use it when healing a 5-player dungeon. Druids have a lot of heals and if we made Healing Touch more usable it would come at the expense of another heal like Nourish or Regrowth. The problem again is the talents. It doesn’t make sense to have so many talents support a situational spell.

But GC, you ask, if you know the talents are lackluster why haven’t you improved them? The answer is that Resto druids are in a really good spot now. If we changed the HT and Tranquility talents to boost other kinds of healing, then Resto druids might become too powerful. If we nerf Resto healing to compensate for these talents, then druids might have less flexibility in their talent builds. We would also suffer a lot of player backlash if we nerfed spells just to make talents (which you may not be able to afford anyway) any better. We’ll suffer player backlash when we think the change is important. I’m not sure this one is. We’ll definitely rejuggle the talent tree long-term though.

Community Team: The next set of questions are related to the itemization for druids which was the source for a wide variety of questions. For one, Restoration druids feels that critical strike rating is not very helpful for them yet it is often present on the leather casting gear.

Q: Do we have plans to improve the usefulness of critical strike rating for restoration druids?

Ghostcrawler: Yes. We’re fine if some stats are slightly more attractive than others. The problem comes when some stats are considered junk because they are valued so much less than other stats. We don’t have a ton of stats to work with when you consider that things like Int and Spell Power basically improve at a linear rate with item level. That means we need for say Haste and Crit to be attractive to all casters. This is a problem with several classes, not just druids, and was largely caused by having too many scaling talents that over-inflate the value of some stats relative to others.

Community Team: Tanking druids have regularly expressed complaints on how their tanking gear was homogenized but statistics like Block and Parry do not help them and defense provides a very marginal benefit.

Q: Do we have plans to make these statistics more helpful for them since they are on a wide variety of tanking gear?

Ghostcrawler: We think it’s interesting that a bear and a warrior tank might look at the same piece of gear and place different values on it. That’s one of the elements that makes looting interesting and rewards players who understand their class. You shouldn’t take a ring because it says “TANK” on it. You should take it because it benefits you. And really, when players say “wide variety of tanking gear” they really mean rings, neck, cloak, and possibly trinkets. Currently we’re in a world where tanks emphasize Stamina and Armor as much as they possibly can, which makes other stats feel lame by comparison. But that will likely not always be the case, and we kind of doubt it will be as much of an extreme in the Coliseum.

Q: Have we considered providing more tanking leather and to prevent extra loot clutter possibly finding ways for Balance and Restoration druids to use solely cloth item since they often use them already?

Ghostcrawler: No. Druids are a leather-using class. We are just going to have to make three types of leather (melee, ranged and healing). You have to understand that even though we have pushed bears and cats farther apart, we still consider them to be part of the same spec. We can’t get into the business of itemizing for niches within a particular spec or we’re just going to have too many items per tier. I can see the argument for having tanking and dps leather and making the casters use cloth. That’s just a different design and we currently like for certain classes to be associated with certain types of armor. We like that druids look different from say priests or mages (even ignoring the forms thing). We like that we can kit druid tier piece armor to look a certain way.

Also note that if we buffed bear mitigation through more tanking-oriented leather that we’d just have to nerf them in other ways. In my experience, most bears end up with “tanking leather” anyway because they want to gem and enchant their bear gear differently. Having one set of gear that you wear as cat or bear isn’t really feasible in Ulduar.

Community Team: Many players have been discussing relics and how they feel they add a lot of clutter to loot tables since they can have a wide margin of effectiveness and always have a small number of players able to pick them up.

Q: Do we have any plans to improve how players obtain relic items such as a relic token?

Ghostcrawler: The alternative to “clutter to loot tables” is that they go on vendors. We view vendors as an absolute last resort. They are there as a hedge against being very unlucky with drops and to give players motivation to do bosses even when that boss no longer drops any upgrades for them. When the best relics are available on vendors, then every druid will have those relics quickly. They essentially just become part of the core identity of the class rather than an upgrade that you get at some point along your progression. The best solution is probably something where a boss has a 10% (or whatever) chance to drop a relic in addition to its normal loot table.

Community Team: To wrap up this Q&A, here are a couple quick questions from the far corners of druid minds.

Q: Are there any plans for a 310% speed flight form?

Ghostcrawler: At this time we want to keep the 310% flying speed very rare – maybe 5% or less of all players. If we made a flight form that less than 1% of players had access to (since probably less than a tenth of those 5% would be druids) it’s hard to argue that’s a good use of art time.

Q: What do we think about making spells like Warstomp and Innervate usable while in forms?

Ghostcrawler: We like the fact that Innervate requires shifting. We want druids to shift more. Warstomp I could see an argument to allow in forms since it’s nice (but not mandatory) that racial benefits are useful to a variety of classes.

Q: Do we plans to alter how the GCD works for form changing so that shifting into a form is as easy as shifting out of one?

Ghostcrawler: As I mentioned, long-term we’d love to get druids shifting more often, which means shifting has to be less painful. I don’t know for sure that changing the GCD needs to be a part of that, but it could be.

Q: For the official word, do we have plans to update more druid form models at some point in the future?

Ghostcrawler: I know for a fact that the current Travel Form and Aquatic Form are loathed by the artist who redid bear and cat. We do have plans to update additional forms at some point in the future.