Un blog pour découvrir ensemble le Moine !

ATTENTION : Le blog n'est pas maintenu ! Désormais ça se passe sur http://www.monkspirit.com ! ;)

vendredi 20 avril 2012

Intervention de Ghostcrawler sur l'Activte Mitigation

Coucou à tous !

Aujourd'hui, un ptit commentaire rapide sur l'intervention de Ghostcrawler à propos de l'Active Mitigation. Tout d'abord, voici l'intervention, en anglais pis je vous fais un p'tit résumé derrière pour les anglophobes désespérés ! :P (Et pour la peine je ponctuerai ça de commentaires à moi, na.)
I worry a little that "active mitigation" is getting used by various players to just mean "how I want tanking to be." The designers use it for a pretty specific meaning, so let me redefine it here.

Active mitigation is using your resources for defensive purposes.

Having a bunch of free cooldowns isn't active mitigation (using our strict definition) because you could still be an effective tank by standing there not attacking a boss and cycling free cooldowns. We don't think that would be fun. Hitting a boss with your weapons (or bear claws) is more fun. However, for many tanks, hitting a boss just to contribute 30-50% of the damage of a DPS class isn't very engaging either. Active mitigation converts your rage, runes, energy or holy power into survival. (Not guaranteed survival, because then you won't need healers at all -- but it should provide enough granularity that great tanks survive more than average tanks.)

As an aside, having some free cooldowns is fun and useful. However, it's not our goal to make sure each tank has the exact same number and character of cooldowns. In fact, we think it's a little dull. How disappointing it would be to get your Brewmaster to 90 only to discover she had all the same abilities as your paladin, with different names and icons? As long as the various tanks can handle every encounter with about the same effectiveness and relative skill, we think the system works. We haven't always nailed that, especially in early Lich King, but we think it has been the case for more recent content.

Some of you are characterizing active mitigation as "I spam my finisher." You could probably do that and be a somewhat effective LFR tank. For more challenging content, you'll often want to save your abilities a few seconds for when you need them most. Waiting too long doesn't make sense, because then you're just wasting your resources, and besides, you have the free cooldowns for the really big boss attacks, as well as plenty of healer cooldowns.

For example, if the boss is about to buff himself with something that makes his swings bigger and he does that every 15 seconds or so (too frequently for Shield Wall) and your health is relatively stable (meaning you're not about to die, but you also don't want to just soak the attack) then it makes sense to hold your Shield Block for when a block mitigates 50,000 damage instead of 30,000 damage. Now, if you are about to die, Shield Block RIGHT NOW might be much more attractive. In fact, you might also use Last Stand or Shield Wall, but then of course those abilities won't be available for a couple of minutes. Those kinds of decisions are the ones that (in my experience) tanks find enjoyable and a test of experience and skill.

I think it's simplistic to say that hit / expertise and finishers work against each other. That's like saying that avoidance and mitigation work against each other, because if you avoid an attack, then the mitigation is wasted. It is typically only a problem if you can stack a particular stat to infinity, reach a hard cap, or if one stat is dramatically more valuable than another. A warrior with low hit won't have enough rage for Shield Block. A warrior with low mastery will block a lot, but not mitigate enough damage. A warrior with low avoidance will take predictable damage but drain healer mana. A warrior with low Stamina might not be able to survive a big hit that lands at the same time as a magic attack. Ideally, players can focus on various stat allocations to find out what feels right for them or even tailor gear for particular fights (within reason -- we're not going back to resist gear anytime soon).

If you think a particular stat is undervalued, by all means let us know, but you're probably going to have to provide some math to make your argument. We also don't spend a great deal of effort balancing all of the numbers in beta until we're happy with the abilities -- there is no point making Shield of the Righteous play really nice with mastery if we decide to redesign the ability.

I'm honestly not that worried about our team being able to balance all the tanks. The tank classes were all close enough in Dragon Soul that most raiding groups were able to use their existing main tanks, even on heroic fights, yet there were still some situations where various tanks shined and they felt different enough that a Blood DK wasn't just a warrior blocking with a sword instead of a shield. If balance is something we can solve, then the big thing to worry about is whether or not tanking is fun. We don't think standing there doing nothing, or standing there trying to maximize DPS is going to be fun for tanks, so we want the attacks to translate into some amount of tank survivability. That's the intent behind active mitigation in a nutshell.

I know I have spent a lot of effort discussing and attempting to explain the design intent here, but we really want to get it right.

D'abord, GC rappelle le principe de l'Active Mitigation : c'est le fait de dépenser ses ressources à des fins défensives. Si je peux me permettre, il a oublié "de façon régulière dans la rotation". C'est vachement important cette partie car l'Active Mitigation (on va noter AM, si vous le voulez bien) a cette caractéristique super importante qu'elle fait partie du cycle. 'Fin bref, l'idée est là ! :)

Il explique que simplement taper le boss et enchaîner les CDs n'est pas forcément fun si ceux-ci sont gratuits, alors que le fait que les attaques comptent (de par leur génération de ressources), si. Là encore, il prend un peu un raccourci et omet (car cela lui paraît évident, mais je pense que face à l'ampleur des changements, un peu de pédagogie ne fait pas de mal) le contexte qui est celui d'une aggro extrêmement simple. Or comme je le disais dans un précédent article sur l'AM, c'est vraiment un bon choix de la part de Blizz car une menace tenue est beaucoup trop mise à l'épreuve ou, à l'inverse, biaisée selon le stuff ; alors que quand l'aggro est easy et que les tanks gèrent leur cycle pour booster leur survie, leur performance dépend bien plus de leur skill et forcément, ça procure plus de plaisir de jeu et donne un environnement de jeu plus intéressant.

Il rappelle que spamer son finisher d'AM, dès qu'il est disponible, ne sera pas forcément la bonne solution et devrait certes suffire à du LFR, mais les tanks se targrant d'une vraie maîtrise de leur personnages devront apprendre à utiliser leur finisher quand nécessaire, par exemple au moment où le boss se buffera pour taper plus fort, quitte à attendre 3-4 secondes avant de lancer la technique.

De même, il récuse le fait que les stats offensives et les finishers fonctionnent mal ensemble, et compare ça au fait de dire que la maîtrise et l'avoidance ont une mauvaise synergie. En fait, il faut avoir assez de chacune des stats : un tank devra avoir un peu de stats offensives, pour générer des ressources pour mettre en place son AM, de l'avoidance pour ne pas être une éponge à mana, de la maîtrise pour que l'AM soit efficace et un pool PV suffisant pour ne pas être trop vite tué.
On pourrait quand même ajouter une pointe de réalisme en disant que même s'il faut de tout, on aura probablement une stat prioritaire et qu'on aura du coup un ordre du genre "Cap expertise > Maîtrise > Avoidance > Toucher" (exemple purement hypothétique, pour illustrer).

Il explique enfin qu'il n'est pas très inquiet quant à l'équilibrage et rappelle qu'à DS, les tanks étaient plutôt sur un pied d'égalité et que les guildes n'ont pas eu à switch de classe de tank pour avancer.


En espérant que vous aurez trouvé intéressante cette intervention du développeur le plus connu de WoW, je vous invite à poser toutes les questions ou faire tous les commentaires qui vous passeraient par la tête. N'hésitez pas à demander des articles et/ou vidéos sur des points particuliers de la bêta !

A très bientôt !

Aucun commentaire:

Enregistrer un commentaire