Klasa Q&A: Mage

Community Team:Joining us today to shed some light on the many questions we’ve fielded from players within the mage community is the Lead Systems Designer for World of Warcraft, Ghostcrawler, who has enlisted the assistance of several members of our class design team to provide the most thorough answers possible.

We’d like to begin by exploring the perceived role of the mage class. A lot has changed since the days when the “glass cannon” description was applied.

Q: Where do mages fit in the current scope of things, and where do you see them from this point going forward?
A: The mage is the iconic caster -- a ranged class that wants to stay at range in order to focus on dealing damage. They can do single-target damage, area of effect damage (AoE), or crowd control. Every group should want a mage because they are reliable, powerful and flexible. Most of the mage’s spells have a cast time and a lot of the gameplay involved in improving your mage revolves around minimizing the limitations of the cast time, whether it’s lowering cast time through talents and gear, getting away from enemies so you have an opportunity to cast, or using the occasional ability to make a spell instant.

While all three of the mage talent trees focus on dealing damage, we are pleased with the different feel between Fire, Frost, and Arcane. Arguably mages even have a fourth potential style now that focuses around Frostfire Bolt. We know the stylistic differences work because there are Frost mages who just love Frost and want to see it work in Player vs. Environment (PvE) and Fire mages who want to play Fire in Player vs. Player (PvP). They prefer a play style within the same class over a different play style that would be arguably more effective within a particular aspect of the game. While understandably frustrating for those players, it also points out a success in the class design.

We used to call the mage the master of AoE damage, but we’ve since decided that’s not a great niche for anyone. The “AoE class” feels mandatory in situations where you do have large crowds of enemies to contend with, but then the AoE class gets bored when everyone else is maximizing their single-target damage on a boss. Now we try and give AoE tools to all damage-dealing specializations (specs), though we will always make extra effort to make sure mages stay good in that department.

Players sometimes wonder why the mage class has seen fewer changes than some of the other classes during Lich King. We think that’s because by and large, the class works. That’s not to say there aren’t areas we can improve, but we think the mage has all the right tools to live up to its reputation.

Q: What is it that makes them unique when compared to other classes?
A: All of the mage specs, though less-so Arcane, focus on a single spell such as Fireball. At first glance, and especially to non-mages, this might make the class appear overly simple to play, but really you can have a lot going on. There are chance-on-hit abilities (procs) such as Hot Streak and Firestarter to contend with. Mages have some great tools, like Presence of Mind and Arcane Power, to really kick up their damage on demand. Frost PvP in particular requires a lot of finesse to get the Water Elemental’s Frost Nova at the right time for a Shatter combo. Mages are fragile though (just ask a healer), so they have to make sure they’re staying alive as well and using the tools they have to do so. Even though most of their damage comes from one spell, mages have a lot going on. The damage per second (DPS) difference between a skilled and less-skilled mage with the same gear can be pronounced.

Mages still are a glass cannon when compared to priests and warlocks. While all have their armor spells, the mages also have escape mechanisms from Polymorph to Frost Nova to Iceblock to Blink. Mages should never feel “tanky” in a PvP environment. The biggest risk for homogenization occurs with the mage and the warlock, but in this case we think the mage is in a good place and it’s the warlock that we want to move slightly farther away. We’ll talk more about the locks soon, but we need to focus them even more on mechanics like shards and demons.

Mages also retain some unique tools, such as the town portals and the (ahem) food and beverage service. Their crowd control is still among the most powerful, if not the most powerful, in the game.

Community Team:A lot of initial questions and concerns we received from mages around the world were concerning itemization.

Q: In particular, a lot of the newer PvE and PvP mage or caster items seem to favor Fire spec and, to a lesser extent, Arcane. Do you feel as though mages are being forced to focus too much on critical strike rating (crit) over stacking more haste, spell power, and intellect, stats that are much more beneficial to the Frost mage?
A: It isn’t in our best interest as designers to have Frost want very different stats than Fire. In a world where we already must add so many new items to the game with every new raid tier / Arena season, we just don’t want to dabble too much with “this piece is attractive to the Fire mage, but not the Frost mage.” We think the value of different stats has just crept too far apart for different specs of the same class. It’s just never going to feel right when one stat is worth double or more of the value of another stat. We’re making a big pass at all of the talent trees and item stats to try and get this a little closer for everyone. Ideally you might be comparing two pieces of cloth and have to decide whether the haste or crit is more valuable to you, and not just write off everything without crit as junk. So to answer the question succinctly, yes mages are being asked to focus too much on some stats.

We also understand there are some items in Naxxramas that are superior to items in Ulduar. This isn’t ideal, but is partially fallout from our decision to not have the final boss of Ulduar drop better loot than the rest of the instance, which is a design change from the previous tier. We are looking at the items on a case by case basis as the feedback comes up. While it isn’t our goal to ensure that every drop is automatically an upgrade, it also isn’t our goal that you try and get your group to keep going back to the old content because it provides more upgrades for you.

Community Team:There are some funky cloth legging designs out there that are difficult to truly appreciate unless one wears a tunic.

Q: Even so, will mages get robes, or at least the option of choosing robes, over tunics going forward?
A: To be totally honest, this is not a huge priority for us at this time. We embrace some level of player visual customization in World of Warcraft, but it’s just not in the design vision to give players as many controls over how their character looks as some players would probably desire. One of the distinctive visual qualities of cloth is that it often looks like long, flowing robes, which is pretty consistent with the iconic fantasy wizard. No doubt some players would prefer to change the look of their weapon or weapon enchant if they could without having a game play effect, so this is just a slippery slope for us. We will keep the feedback in mind though.

Community Team:The next few questions concern the number-one issue raised by mages on the forums as of late: mana efficiency.

Mana Gem and Evocation are commonly referred to as outdated mechanics. Many players feel the Mana Gem does not restore enough mana and should not be placed on the same cooldown with a warlock’s Healthstone, while Evocation has too lengthy a cooldown and is typically not a reliable means of acquiring mana during boss fights.

Q: How do you view these mechanics, and are there any intentions of updating mana recovery capabilities for mages in the future?
A: Our general philosophy, in a very broad sense, is that healers risk running out of mana if they aren’t careful or are in over their heads, but that damage-dealing specs generally have enough mana to do their jobs. That doesn’t mean that you never need to burn a gem or use Evocation, but it does mean that if you are being reasonable about what you’re doing that you should have enough mana except perhaps on very long or unusual fights. What we are more likely to do is just lower the mana costs of the main nukes: Arcane Blast, Fireball, Frostbolt, and Frostfire Bolt.

Community Team:Compared to many other damage-dealing counterparts, mages feel their AoE damage is less reliable and way too costly.

Q: Do the developers feel that the cost to mages of doing AoE damage is appropriate?
A: It’s close. We don’t want say the Blizzard spell to ever look really attractive to use against a pair of creatures or a single target. It’s taxing on your mana bar to do many Blizzards, but it doesn’t feel inappropriate for the amount of damage you’re doing during that time. The efficiency is still good in cases with a large number of targets, which is the whole point. Now some of the other mage spells could definitely use some improvements to make them as competitive as Blizzard (the spell) in terms of usability, damage or efficiency.

Clearly it’s in our best interest to make sure a spell with the name “Blizzard” kicks some major posterior.

Community Team:Spell Steal is a very costly spell, especially considering it can be resisted, an unnecessary buff can be stolen unintentionally or the stolen buff can be dispelled.

Q: Are there any plans to reevaluate the mana cost and functionality of this spell?
A: We think the core of the problem is that a spell that was designed to let you steal cool buffs from an enemy has sort of fallen into the niche of a general dispel. Rather than make it cheaper, we’d be more likely to let it actually only steal spells that would benefit the mage. This would be a buff in some cases and a nerf in others though, so it’s not a quick and dirty change. We have considered a glyph to let Spell Steal take two buffs at a time.

Community Team:Now let’s jump to some questions about specific talent specs.

Firstly, the Arcane tree is widely considered too bloated. It seems that, over time, the talent trees of all classes have really evolved to provide plenty of different options with fewer five-point talents to allow for greater customization. There are several flavorful talents in the Arcane tree (i.e. Student of the Mind, Magic Absorption, Magic Attunement, Incanter’s Absorption, etc.), but many players feel that they cannot afford to spend points in such places since many of the most necessary damage-dealing talents require five points.

Q: How do you feel about revitalizing the Arcane tree to thin out some of these five-point talents?
A: Arcane is a little bloated. If you take all of the damage and mana talents there aren’t many left to spend on the more fun or cool talents. We recognize that it’s hard, for instance, to have a single Arcane build that can work in both PvE and PvP. To be clear though this is a problem with several of the talent trees and not a problem with Arcane alone. If you look at say the warrior Protection tree or the paladin Retribution tree, those provide a model for where we’d like to take talent trees in the future – fewer talents overall and plenty of points to spend on fun play-style choices that really do feel optional rather than talents you need to make your spec function. Also note that fixing some mage mana issues might make some of the mana talents feel less mandatory.

Community Team:To expand upon the last question, Torment of the Weak is used in several of the most popular mage talent specs for both PvE and PvP, however, the Arcane talents prior to this one are of very little use to Frost mages – and Fire mages to a lesser degree -- in PvE.

Q: Are you concerned at all that Torment of the Weak is considered to be so important to mages, regardless of specialization, that a minimum of eighteen talent points must be spent in the Arcane tree to reap its full benefit?
A: We don’t think it’s must-have for Frostfire builds and it doesn’t strike us as weird that Frost or Fire would subspec into Arcane, since that is generally going to offer them more than say a Frost mage who subspecs into Fire.

Q: Are there plans on the horizon to improve Fire mage representation in PvP?

A: Yes. It is more important to us though to fix classes that have no viable specs than it is to bring options to classes that already have a reasonable Arena presence. We are more focused on improving hunter and warlock representation than making sure Fire has a PvP role. It’s still something we would like to do, but in a game of this size there are a lot of things we’d like to do. Dragon’s Breath is one spell we think we can improve for PvP. With a lower cool down it could be more like Scatter Shot. It’s not necessarily that Fire is terrible at PvP, just that Frost has a lot more tools.

Q: Is threat generation from Fire mages a concern at all given their burst damage is controlled mostly by proc talents and critical chance?
A: Threat-generation is a concern. One way we’d like to fix this is through Invisibility. We’ve always been a little cautious with making sure the spell wasn’t too powerful, but we think we have plenty of room to improve it. In PvE for example, it’s really hit and miss whether you’ll take damage that will prevent the threat wipe. Do remember that Mirror Image is quite useful as a threat-reduction spell. Your threat is divided among the images while it’s active. Sometimes it makes sense to blow the spell right at the start of a fight, and other times when you get a spell buff or are otherwise able to go into really high damage mode for a few seconds.

Q: Do the developers still consider it an objective to improve Frost damage for PvE?
A: Yes. The challenge as always is to make sure we don’t over buff Frost in PvP just to make it viable in PvE. While it would be ideal for all specs to be viable in PvP and PvE, having different PvP and PvE specs at least keeps those specs alive rather than having one tree which is good at everything. We’d like to buff Frost through Ice Lance. Currently another Frostbolt is always better than an Ice Lance in PvE. We experimented with improving this through the glyph of Ice Lance, but it turns out the glyph would have to improve Ice Lance’s damage by x6 or something ridiculous like that.

Q: What considerations are being made for combating the use of totem-killing macros in PvP, and is this seen as a problem currently?
A: It is a problem. Totems are fragile by design, since they essentially can do their jobs while the shaman does other things. We’re fine with this except in the case of pets being able to target and kill totems via macro without any input from the player. We want the player to at least have to target the totem first before commanding the pet to attack. We have a technical change in 3.2 which will prevent macros from doing this while still giving macros the functionality that players currently enjoy elsewhere. Players have made many suggestions for how to accomplish this technically (a common one is to rename the totems), but the solution is a little more complicated than that. Whacking a totem with your weapon or wand is fine because you are making a choice and spending valuable combat time (similar to dispelling a buff or debuff). Using macros requires no player interaction, and is not working as intended. One totem that is just too easily destroyed is Mana Tide, for which we could see bumping up the survivability. Again, remember that while totems are fragile, they aren’t particularly expensive (especially if we get the change in to drop more than one at a time) and the shaman can do other things while the totems are doing theirs. We’re sympathetic to warlocks and hunters being the two classes most affected by the removal of totem stomping, and we do think their respective representation is low, but we don’t think the way to buff them in Arenas is by letting them be ultimate totem stompers. On the other hand, a reason to just not hotfix in more health to totems is that that change would primarily nerf warlocks and hunters, who need nerfing the least in Arena at the moment.

Community Team:Finally, this wouldn’t be a mage Q&A without a question about Blink. It has been discussed in the past that it’s the terrain that can cause the spell to fail and not necessarily an issue with the spell itself.

Q: While mages do recognize this issue, has there been any discussion about reworking Blink so it’s more intuitive and could recognize a mage failing to teleport any distance forward, wasting only a global cooldown rather than the mana and spell cooldown?
A: Blink is a movement spell, and anything related to movement can be a little dicey on a client-server game like World of Warcraft. That’s not an excuse for it bugging out, but an explanation for why you can get into situations where it doesn’t seem to work. In the 3.1 patch we made some technical improvements to the spell working on slopes. It used to fail a lot in the portal area of Dalaran for example, but that has been much improved. One of the places where it still seems to struggle the most is entering or exiting the tunnels in Warsong Gulch, which ironically is one of the places where it’s also the most useful. Anywhere there is a change in terrain, such as entering a building, could be problematic. We are working on this issue. If you run into a problem with Blink, the most helpful thing you can do in the Bug Report forum is specify where exactly you had the spell fail. That will let our engineers zero in on solutions.