Feed RSS RSS feed
2008-08-22 15:12:27
Raid stacking in Wrath of the Lich King
As we've suggested a few times, we are currently taking a serious look at all of the various buffs and debuffs classes can bring to parties and especially raids. A few of these changes have shown up in the recent build, and more are on the way. I'll first explain why we're doing this and then what generally we are doing.

Why we think this change is necessary:

1) We are adding a new class in Wrath of the Lich King as well as improving some specs that were previously viewed as inferior for PvE. Yet because the raid size is remaining at 25-players (and we expect many people with raid with 10), it may be hard to fit more people into your raid.
2) We are moving almost every buff to affect the entire raid. There are only a few exceptions, and these tend to be short-term, bursty abilities. Most buffs from Battle Shout to Leader of the Pack will affect the raid.
3) While we are likely to increase the debuff limit on bosses, this is a problem we keep running into. Furthermore, it asks a lot of players to be able to parse bosses with 30, 50, 100 debuffs on them. At some point, we need to stop the madness.
4) Stacking a raid in order to get the right buffs has started to feel a little like a crutch, much like stacking consumables felt not so long ago. Because some of the buffs scale so well and have so much synergy with other classes, you may sometimes feel that you should pass over a really skilled player in order to pick up a buff that will bring more to you group. We'd rather get back to bringing good players or, gasp, even your friends.
5) We want the challenge of the encounter to be the fight itself, not collecting all of the buffs and debuffs you need to succeed. Buffs are fun. We don't want to cut them or nerf them significantly. But we do want to make benefiting from them less of a burden.

Some changes you're going to see:

1) Ease and flexibility in getting all of the major buffs in the group, while still having a few spare slots to take the people you want to take.
2) More parity in which classes can raid. While it's probably not realistic to get 2.5 of every class in a 25-player raid, we can get a lot closer than we are today.
3) More overlap in buffs and debuffs and very few unique buffs. If you can't get player X to bring the melee haste buff, maybe you can get player Y.
4) No two classes should have the same set of buffs. This is to make sure that one class can't completely replace another. If you have two players both bringing melee haste, there is a good chance one of them also brings something else you need. Again, the goal is flexibility.

A few examples:

1) To get a spell damage vulnerability debuff on a boss, you can bring a a warlock for Curse of Elements or a death knight for Ebon Plague. They don't stack, but both classes also bring additional benefits in case you want to run with both. The lock brings Curse of Recklesness, Demonic Pact, Curse of Weakness, Blood Pact and Fel Intelligence (depending on spec). The death knight brings Abomination's Might, Improved Icy Talons, various Auras, and a new Str + Agi buff you haven't seen yet (again depending on spec).
2) To get a 5% melee crit bonus, you can bring a Feral druid for Leader of the Pack or a Fury warrior for Rampage.
3) While having a mana battery feels essential in many cases, you can now bring a Shadow priest, Retribution paladin or Survival hunter to fill the role.
4) You can improve health of the entire raid with Commanding Shout or Blood Pact. You don't need both, so perhaps the warrior can Battle Shout or the lock can bring out another pet instead.

I want to stress that none of this is set in stone. A change like this is going to take some iteration until it feels right. While we are pretty happy with the plan, it is certainly possible that some class or spec is going to be hurt by a change more than we anticipated. We don't want to hurt anyone's viability while soloing, in small groups, or in PvP with this change. We aren't trying to slap anyone in the face -- we're trying to make the game more fun.

The goal is to get more people into raids and to let you bring the people you want. Ultimately, that should benefit everyone. Please try and keep that in mind as you start to see the changes.

And of course, please share your feedback on them with us. :)
Avatar of Ghostcrawler
Ghostcrawler
2008-08-22 15:42:24
Re: Raid stacking in Wrath of the Lich King
Q u o t e:
The implication here is that their abilities will likely not stack. Is this the correct?


Mana batteries are a slightly unusual case because you probably want 1 for a 10-player and 2 for a 25-player raid. Assume that you can stack any 2 of the mana battery classes (2 Shadow priests, or 1 Shadow Priest and 1 Retadin).

And, again the point is NOT that you're penalized for having too many Retadins. It's more that you're NOT penalized if you lack a Retadin (because there is some redundancy in buffs and roles).

[ Post edited by Ghostcrawler ]
Avatar of Ghostcrawler
Ghostcrawler
2008-08-22 18:23:33
Re: Raid stacking in Wrath of the Lich King
The goal is to try and boost Affliction, and any other spec that has been traditionally under-represented. We haven't made much of a numbers pass on classes yet -- we're still in the stage of building new talent trees.

Without getting into too many specific examples, the 5 warlock stacking for Sunwell is exactly the kind of thing we'd like to avoid. Warlock 1 boosts Shadow damage, so warlock 2 does more damage, and warlock 3 does damage and... hey why did we bring a mage, we'd do more dps with another lock!

If I need to come right out and say it, stacking a class for a single ability such as Bloodlust to be used in rotations is not good for the raiding game. I think most players suspect as much.

I'm not so certain raids will fill their remaining slots with rogues and hunters. For one, those classes aren't going to be taking a lot of advantage of all those nice spell buffs you've got going in the raid. Two, if they are bringing buffs, it isn't fair that their dps is greatly superior to everyone else. Higher dps is fine, especially for classes that don't have the option to respec into tank or healing. But it needs to be a modest dps boost, the kind of thing that will be swallowed up by variance in player skill for all but the absolute best 1% of raiding guilds.

Now we're not [very] naive. We know you guys are smart and will come up with ways to min / max a raid. But we're hoping this change will reign in that in to some degree. Even today not every Sunwell raid is stacked the same way, and with more flexibility on who is bringing the buffs, raids should be even more diverse going forward.

Q u o t e:
Again, I ask you, why not reduce the effectiveness of shaman utility (and utility in general for that matter) but increase their personal performance?


Yes, totally. It works both ways. If everyone is bringing buffs, then classes that have always brought buffs need to be better at dps (or healing or tanking or whatever) so they aren't eclipsed. We want you to have options, and if some specs or classes are just demonstrably worse, they really aren't an option.

Technically the DK has to spec for Ebon Plague as well, but we are cognizant of the potential for "wasted" talents and discussed that in detail when hatching this plan. To be fair, that potential exists today and players typically work around it just by knowing the people they raid with and how they are spec'd.

[ Post edited by Ghostcrawler ]
Avatar of Ghostcrawler
Ghostcrawler
2008-08-22 21:53:46
Re: Raid stacking in Wrath of the Lich King
I will try to answer as many questions as I can. Please, though, when I cannot, keep the conversation going.

Q u o t e:
If it comes down to needing one of nearly everything... raiders would either have to be happy with getting less playtime in raids or you would have to hope your raiding members are skilled enough to repec to a missing spec... and also geared enough... to fill that role.


This is actually what we are trying to avoid. Our feeling is raids work too much like that now where you have to bring X class to have a reasonable chance of success. There are some buffs that most players would consider mandatory, or at least very powerful. You'll want to get those. You'll want 2 or so tanks and 7 or however many healers works for you. Then just take good people.

Q u o t e:
What if, for example, my best buddy is a Feral Druid and I'm a Fury Warrior. Should one of us be forced to respec to get raid spots together? Should we be encouraged not to run 5-mans together because our buffs don't stack, etc?


No, that would be a fail in our minds. You have 23 (or 8) other slots to get the buffs you need. And realize when I say need that everyone's needs are different. Haste is pretty valuable, as is mana return. Is Arcane Int? Is Imp Prayer of Spirit? Your guild will have to decide that. In our design we made sure it was possible to get all the big buffs with as few as 11-12 people. That gives you a lot of room to bring your buddies even if they are the same role, class or spec.

Q u o t e:
while it all sounds nice, i kinda feel like i and others are being patronized with this post and the juice therein


I'm so sorry. It is not my intention to patronize. We thought we would try and explain our logic and engage the community in an open discussion. Otherwise we thought there might be a lot of chaos when you got the next patch and didn't understand some of what you saw.

Q u o t e:
here's an idea: lets just keep that cruise control for "easy mode" on and just speed up a few more years down the road...*all classes now have every possible buff in the game as soon as they enter an instance - we want it to be fun and easy for everyone to enjoy...screw brains and effort LETS ALL HAVE CAKE!*


It isn't our intention to dumb down the game. In fact, we want the challenge to be more about the encounter itself and not the buffs. When you think about it, it's not a big mystery what buffs you need for a good raid. The challenge is in finding good people who can fill those roles while explaining to good players that you just can't use them tonight. That's a logistical challenge I guess, but we are making the assumption that most players would rather face strategic or tactical challenges. In a game as large as WoW there will be exceptions of course. We view this similarly to actions we took when raids ran around getting the Zul'Gurub or Onyxia buffs before going on a raid. That wasn't particularly fun, but people felt like they had to do it or they'd be surrendering a big potential advantage. The flask stacking and consumable stacking got to be a similar burden. I'm not sure a lot of people wish for those days again, but no doubt some do.

Q u o t e:
when i was progressing to dethrone nefarion i felt damn good and accomplished and it was all in part of the endless preparation and strategy efforts involved. so unless you can make a fight that is a constantly changing encounter for everyone so we cant cheat by watching videos and reading strats...then i just say "thanks but no thanks. next idea?"


Do you feel the victory was sweeter because you had to farm consumables and get just the perfect mix of class and spec in order to beat him? I would be interested in other opinions on this topic. Like I said, this is not set in stone. I am here specifically asking for feedback.

Q u o t e:
Every non-stackable raid buff needs to be the exact same for it to be that great. Like Rampage/LotP. Rampage doesn't heal/give it to ranged. So if you already have battle shout, druid trumps warriors. Etc etc. Its like my original post where I was asking if some things were the same, but now I'm looking at it, they are not.


I will admit that is a logical extension of our argument, but in reality I don't think it will get to that point, at least for 99.9% of us. If Ebon Plague was a 12% buff and Curse of Elements was a 10% buff do you think that 2% means the difference between success and failure? What if one buff has a longer duration or costs less mana? We think you just hit a point of diminishing returns, where as long as you have the big buffs you need, and maybe some of the minor ones, the rest is going to be in the noise. One or two percent may seem a big deal for the guilds going for world first on Kil'Jaeden, but that kind of player is going to go to extremes almost no matter what we do. For most of us, our players being focused and getting lucky with a string of crits probably has a much bigger impact than a 2% difference. Or put another way, a buff that increases your dps by 1300? Yes please. Would I boot that player to get the buff that offers 1330? Probably not.

Q u o t e:
As the "only true hybrid" class, what extra value are we bringing to a raid over, say, a Druid or a Death Knight? More specifically, why would a raid leader consider an Enhancement Shaman over a Feral Druid or Death Knight when both of these classes can DPS, provide excellent bufs (LotP, icy talons, the "str/agi" buff you mentioned earlier), provide utility (interrupts, silence, battle rez), as well as fulfill a crucial tank role (and possibly main tank role) in a pinch without respeccing?


Well for one thing, flexibility. If you end up just not being a stellar healer or decide you hate healing, you can respec to two kinds of dps, while the rogue who can't dps just has to go home. One of the strengths of the shaman class is flexibility. While someone else might offer a haste buff, you do have other totems you can drop. Even if there are enough players to cover all of your best buffs, if you're a good player, most guilds would take you anyway. They might have trouble doing that today on live because bringing you might mean no armor debuff or spellpower buff. It's easier to have coverage with the new plan. If you're sad because you thought it was awesome that your group got to bring 6 shamans, well then it might be harder for you now. But I don't get the sense that is your concern.

Q u o t e:
The dungeon designers are number crunchers, and they often design bosses around the most optimum raid setups.


The dungeon designers are down the hall from me. They design the encounters around the stats and abilities of the classes. If we tell them what buffs they can expect, they can use that information to balance fights and make them interesting. If we tell them they can’t count on there being a MT who can Shield Block, then they won’t make an attack like Shear.

Think about some of the later BC encounters – they were designed to challenge the current status quo. Nobody runs out of mana? Okay, we’ll hit the tank really hard. Paladins can heal the tank forever? Okay let’s make them run around a little bit. The raid can achieve X dps because Heroism / Bloodlust (and drums, and chain chugging pots) is up a big chunk of the time? Cool, they know how much health the boss can have. If we thought it worked out better for the game for the MT to have 16K health, do you think Patchwerk would hit for 19K? While we misjudge and make mistakes sometimes, the design of the encounters is to challenge whatever players can currently offer. It’s a sliding scale.

[ Post edited by Ghostcrawler ]
Avatar of Ghostcrawler
Ghostcrawler
2008-08-23 10:13:45
Re: Raid stacking in Wrath of the Lich King
I can't respond with every single change to specific classes. That will all come out over time. But briefly:

Scaling is enormously important to us. We are trying to make every ability scale that we can, otherwise class balance just falls apart at certain gear levels.

We know the numbers are off on some classes. We understand numbers can completely make or break an ability or a whole class. But they are comparatively easy to change compared to implementing abilities in the first place, so they haven't gotten a solid pass yet.

If you look at your class and your buffs seem lacking, uninteresting or underpowered, it's probably because they haven't been updated yet. In the past our design was more that there are classes that bring buffs but have lower dps, and classes with high dps but fewer buffs. While not every class will end up at exactly the same dps, we can bring things a lot closer together. Rogues, mages and hunters can end up with good buffs that fit their kit. Poisons are a great angle for rogues, though not the only one.

As far as role balance goes, it would be very elegant if a 5-player instance took 1 tank and 1 healer and a 25-player instance took 5 tanks and 5 healers. But we've designed a number of boss encounters now, and for a variety of reasons, some creative and some mathematical, it just doesn't work out that way.

We have a plan for mana generation that lets about 2 players with that ability benefit a 25-player raid. Having a little redundancy beyond this is fine since those classes all have additional buffs and could pick up the slack if someone dies or is an encounter that prevents them from doing their job for a time.
Avatar of Ghostcrawler
Ghostcrawler
2008-08-23 15:59:11
Re: Raid stacking in Wrath of the Lich King
This is a very informative discussion. Please keep it going!

Remember the goal is to give you flexibility. The goal is to make raid stacking less uber so that more classes can get into a raid. It's easy to come up with potential problems with implementation specifics, but the goal is not to chase rogues or anyone out of the raid. If your group has 3 awesome rogues, or 2 fire mages that manage to pull out crazy dps, you should be able to bring them without hearing: Nope, we only raid with exactly 2 rogues, 1 mage, 2 hunters, 1 arms warrior, 3 destro locks, etc. etc.

And if I didn't say it clearly enough, rogues and any other class that feels they don't currently bring enough synergy to the raid will get be getting new or modified abilities to fill that role. The idea is to back down a little from: I have crazy buffs but that's really it, or I have mad dps, but that's pretty much it.
Avatar of Ghostcrawler
Ghostcrawler