A few weeks ago, I asked readers to post questions for a simple Q&A style bog post. Unfortunately, it took me a long time to get around to answering them, and even longer to get this post typed up. But at long last, here it is. Feel free to post additional or follow-up questions in the comments, if there are enough I’ll try and post a Round 2 before the new year.
I’d be interested to hear some more gameplay related thoughts. Not limited to the following, but just throwing a few ideas out:
– Dealing with information overload (particularly when you’re also raid leading)
Luckily, I don’t raid lead much anymore, so I don’t have that to worry about most of the time. But my strategy for dealing with information overload was generally dealt with by careful UI planning. To illustrate what I mean, here’s a screenshot of my UI during an encounter
I took a lot of time to set my UI up, and it’s generally arranged around a principle of “least eye movement.” The idea is that the stuff you need to act on quickly should be in front of you, and require little to no eye movement to see. Stuff that’s not as important, or things you only check once in a while, go farther out. This minimizes the amount of time your eyes spend searching your UI for information, and lets you concentrate on the decision-making portions of gameplay.
The most important stuff goes near the center – stuff that I don’t want to or can’t afford to miss. This includes cooldowns, DBM timers that are about to expire, and a few power aura strings that warn me about encounter-specific stuff. As you go out radially from the character, the level of importance drops. Rotational abilities, health, and debuffs are right below me, so they’re close at hand. DBM warnings are a little farther away, in the space above my character. Incoming and outgoing damage, DBM timers that have plenty of time left on them, buffs, and Recount are all near the outside where I can mostly ignore them. And of course, if there are DBM timers I can ignore completely, I often disable them. Finally, I stick a lot of informational stuff on ChocolateBar, most of which gets ignored completely unless I’m out of combat.
– What elements do you look at in combatlogs when evaluating your own (or other tanks) performance? Both cursory & in-depth check would be interesting
I don’t spend a lot of time reviewing logs these days, so Meloree might have more interesting answers here. But generally the cursory review will look at the obvious stuff: DPS/self-healing, cooldown and Holy Shield usage, and ability usage. Checking things like the average time between Crusader Strikes and number of Holy Shield casts in a fight is a fairly quick way to see if a tank is active enough.
I tend to be more interested in death logs though, because they tell me more about how a tank plays. One of the first things I do after a death is look at DeathNote to see what happened. Was it a lack of healing? Did I botch a cooldown, or stand in something I shouldn’t? How long was the death event, and did I pop any available cooldowns appropriately to deal with it? If I’m evaluating a tank, I’m looking for them to be using WoG, Scales of Life, Holy Shield, trinkets, etc. during a death event. Even if they ultimately died, it shows that they’re taking the appropriate measures to try and prevent the death. Of course, the better the tank is, the fewer death events there are to analyze.
– Targeting. It often feels like we’re fighting against Bliz’s targeting system (hi adds who spawned on the far side of rhyo’s platform) – how do y’all go about winning this fight?
I make pretty heavy use of macros for targeting. I have quite a few “/tar MobName” macros made up, and two key binds dedicated to fight-specific mechanics like that. There’s also an addon called “I Have Macro LoL” that lets you create a dynamic macro which updates based on zone or boss fight, which helps coordinate and consolidate those macros. However, I think the author has retired, and it’s usability has been spotty this expansion.
In addition, I use the nameplate addon TidyPlates to help with manual targeting. It’s one of my “must-have” addons, simply because it’s a quick and efficient way to display what I have aggro on and what’s loose. It also makes it very easy to pick up the one add you don’t have aggro on, thanks to its aggro-based color-coding and size scaling.
With the threat modifiers set up so high now along with relatively simple, technically, tanking requirements, what do you think can be done to bring a challenge back into the role?
Threat, as a mechanic, was an amplifier that increased our influence over a raid’s success. A tank who performed poorly could threat-cap their raid and cause a larger raid DPS loss than their own individual contribution. As the threat mechanic has slowly been eroded away, we really haven’t gotten anything to replace it.
Active mitigation could fill that role, but both Mel and I have reservations about how successful it will be. We’ve talked in some detail about our concerns in the past few weeks. My biggest concern is that it will be tough to balance across skill levels; if it’s tuned so that players at the top end of progression find it challenging and engaging, players of lower skill level will struggle and be much squishier. And if they tune it so that it takes very little skill to maintain survivability, players at the top may find it boring or trivial.
Nonetheless, I think there are a few things they can do to engage and challenge tanks. For starters, they could make our DPS matter, so that performing a rotation well had a measurable benefit. The fluctuations in DPS, or in other words the penalty for performing the rotation poorly, would also have to be larger. If a weak tank generated 30% of a pure DPS class’s output, but a strong tank could reach 75% of a pure class, people would definitely start caring more about nailing the rotation.
But the simplest solution is just to make more interesting encounters. I gave some ideas for this last month, but the thought isn’t limited to making sure that fights encourage taking two tanks. The jobs they give us could be more difficult, more varied, and ultimately more interesting than what we generally get. Even in Dragon Soul, we’re still seeing a lot of “taunt based on debuff/ability” re-hashes, which really aren’t that inspiring. We need more fights like Spine, with multiple adds to wrangle and positioning to care about, and fewer like Madness, which is a bit of a snooze-fest to tank.
By the way, if you’re the same Zaephod I ran into in LFR last Thursday, howdy.
there is a point where in any role one goes from mediocrity to mastery – do you remember what this point was for you?
How do you know I’m not still mediocre? 😛
I don’t think there’s a single point at which that transition occurs – an “I know Kung Fu” moment, if you will. It’s a gradual process. I was certainly mediocre when I started tanking in BC. And while I might consider myself an “expert” now, I still recognize areas where I feel I could improve further. But I can identify the point where I finally felt like I had passed the boundary between mediocrity and mastery. It was in Ulduar, where we were finally seeing difficult content for the first time since becoming full-fledged tanks in 3.0. And it was the first time where I felt like the time and care I had put into optimizing my play was paying off in progression. It was also roughly the time I was gaining prominence as a theorycrafter, so it was a transition on multiple fronts.
How are you able to get so much content downed each week only raiding 2 nights? What are a few steps that your average raid can incorporate that will make them more efficient?
The reason that I think Cadenza manages to clear content efficiently on a limited schedule is the players. We all know that we only have 8-9 hours a week to kill bosses, and we all want to kill those bosses, so the motivation to perform well is there. We don’t have many (if any) people that are showing up for the ride and some free loot; we’re all there for the challenge of the kill.
We’re also fairly strong players (though I may be a biased observer), and tend not to make the same mistake more than once or twice, which is critical to minimizing the number of pulls it takes to learn a new boss. And we’re not shy about pointing those mistakes out when someone is repeating them; if someone’s becoming an obstacle to getting a kill, we make it clear what they have to fix, and expect them to fix it quickly (as you famously say on the podcast, “fix your shit”). When one of us screws up, we’re generally more angry with ourselves than anyone else is with each other. That leads to a strong sense of mutual respect between the established players, because we all know the level of play that each player can demonstrate. And in general, the players live up to that expectation.
For most Cataclysm bosses, the biggest hurdle to killing a boss is generally getting a pull where nobody makes a glaring mistake, which could mean anything from dying to an encounter mechanic to forgetting to switch to a burn target. In a statistical sense, you want each player to have as low a probability p of making those mistakes as possible, because your success rate is (1-p)^25. If everyone has a 10% chance of screwing up (p=0.1), your chance of killing the boss on any given pull is around 7%. Drop that to a 5% chance, and your success rate goes up to around 28%. Drop it to 1%, and you have a 78% chance of a kill. So it’s important to minimize those sorts of mistakes. As a result, the best advice I can give to other raid leaders is to stress that concept when attempting new bosses. It’s far better to see the enrage with everyone alive than it is to limp along through an attempt with 2 or 3 people dead. Not only do you learn more from those attempts, but each player gets more familiar and comfortable with the mechanics, which leads to indirect DPS increases across the board.
Another trick we utilize to keep things efficient is to not waste much time looting. When you only have 4 hours, spending even a few minutes after every boss waiting for loot to be decided adds up to a significant portion of your raid night. We use a Loot Council system, so we avoid most of the overhead that comes standard with any DKP system. The GM, who is also the primary raid leader and my co-tank, hands out loot after every boss by taking whispers and distributing as he sees fit. While he’s doing that, I’m taking the rest of the raid group and clearing trash towards the next boss. That way, there’s almost no wasted time, and nobody gets bored enough to wander AFK during loot.
Also, I would be remiss if I didn’t mention that one of Mel’s earlier blog posts has some great tips for efficient raiding.
For beginner pally tanking, how important is gearing? Got a message today from someone wanting to run 5s to farm VP so they could have their 2piece before they started tanking normal Shannox. I told them they were over-thinking it, but had no math to do with it.
I don’t think any of the set bonuses this expansion are necessary for tanking normal-mode raids. The gear requirements for normal modes are pretty lax, especially if the tank plays well and uses cooldowns intelligently. It’s certainly important to collect some amount of gear, simply to make sure you have enough health to handle the boss’s output. But you can do perfectly well with just off-set pieces if you have to.
Dan Paladin asks:
There has always been a struggle to balance the effectiveness of the various tanking classes without sacrificing what makes these classes unique. Several places where blizz has done this are flavor, mobility, raid utility, buffs, damage intake method (sword and shield v. death strike v. savage defense), among others. How well, in your opinion, has Blizzard balanced each of these, how would you suggest Blizz tweak their methods of balancing around these areas, and what other areas do you think Blizzard can explore to ensure class homogenization among tanks does not become a further issue, especially with the creation of a 5th tanking class?
I think that they’ve been getting better at preserving each class’s flavor while still giving them the tools they need. That said, I think the class homogenization complaint pops up far more often than is warranted. Playing a warrior still feels different than playing a paladin or bear, even though we all have the required tools. As long as the classes retain significant rotational and resource differences, they’ll always feel different from one another, even though they might all have identical Taunts or similar cooldown complements.
As an example, consider the first tier of Paladin talents in 5.0. We’re getting mobility skills, which you could deride as “homogenization” since we’re getting something that warriors and druids already have. But our implementation is distinctly different from either of those classes; we get boosts to run speed, warriors have teleport-like effects (charge, intercept, and leap all basically teleport you to the mob), Druids get a mix of both (Charge and Stampeding Roar), and Death Knights bring the mob to them. So despite all 4 classes having mobility skills, they look and play completely differently.
Looking forward, I think Blizzard needs to be careful with active mitigation. If they go out of their way to give every class a different implementation, they run the risk of making one class too strong (see, for example, DKs in 3.0, Paladins in 3.2 or 4.2, potentially Warriors and DKs in 4.3). Or worse yet, one class too weak (DKs in 4.2) because their active mitigation plays a larger role than the other classes. This was one of Mel’s main points in his latest blog post. On the other hand, nobody wants to see all 4 classes get clones of Death Strike. I think this is one situation where gameplay has to trump homogenization though; it’s more important that the classes are balanced well than that their active mitigation implementations are vastly different.