Fury Warrior Feedback

#1 - April 30, 2018, 1:23 a.m.
Blizzard Post
I have waited until I have extensively tested the Fury Warrior on Beta (as I didn't have alpha) to make my mind up about it. I took the time to play it before writing this, as Blizzard have requested. I am a lifelong Fury Warrior player and would like to continue playing it in BFA. Apologies for the wall of text below; here's what I think so far.

Part 1 of 4

General Rotation
While basically the same as Legion the rotation feels so much slower. To clarify here, by slower I'm not referring to the GCD change, we'll get to that in due course. I'm talking about rage generation; how many auto-attacks are required before being able to use Rampage and how long it takes to perform those auto-attacks.

Specifically the change to how Enrage works combined with the alteration of the rage given per auto-attack. To test the changes made in BFA vs live, I hit a target dummy with only auto-attacks and record the rage given per attack, until reaching 100 rage. This was done without the Endless Rage talent. I performed three repeats in each case to try to have a better sample size for variability.

My findings were that in BFA the rage generated per auto-attack is more variable than in the current live version. On live, every main-hand strike generated 5 rage and each off-hand strike generated either 2 or 3 rage, averaging out at 2.5 per hit. In BFA there is variability in both the main-hand and off-hand rage generation: being either 6/7 and 3/4 respectively. The averages worked out to be around 6.3 rage per main-hand attack and 3.1 per off-hand attack. We can clearly see that there is more rage generation per attack in BFA than live.

It took 27 auto-attacks to reach 98 rage on live, with the next attack to be a main-hand hit and put us over the 100 rage cap. In BFA it took 20 auto-attacks to reach 94 rage with the next attack scheduled to be a main-hand attack and cap us nearly exactly. There is potential for less wasted rage in BFA, but obviously abilities would be used, which means we can't say for certain.

So if we get more rage per attack, and it take us less attacks to cap on rage, this is surely a good thing, no? This should result in faster game-play, right? Well not quite. The change in auto-attack speed bonus from Enrage (100% live to 50% in BFA) must be looked at carefully. On live (ignoring haste), our weapon has a 3.6 second swing timer, which halves (1.8) when Enrage is active. In BFA, the 3.6 second swing timer goes to 2.7 seconds when Enrage is active.

So if we assume 100% Enrage up-time in both cases, we can calculate the time it takes to cap rage. 21 auto-attacks for BFA at 2.7 seconds per round of attacks (remember we attack twice as we have both weapons), results in: (21/2)*2.7=28.35 seconds to cap rage. In the live case: 25.2 seconds. Now, I would like to ascribe this difference purely down the to Enrage duration buff to 6 seconds in BFA and therefore the increased Enrage up-time it should bring. However, with the loss of our legendary effects, along with traits such as Uncontrolled Rage, Oathblood, Pulse of Battle and Death and Glory, we are generating less rage overall resulting in this slowdown in pace, which is abhorrent to those of us currently enjoying the live version of Fury.

The problem of rage generation may well be fixed by some Azerite traits, but that's what we keep hearing about everything and frankly, it's a poor excuse. We may well be able to keep our artefact weapons and legendary items for a time, but they will lapse at some point and leave a void in game experience at whatever level Blizzard decide to make them ineffective. Why should we go from an enjoyable spec, to something vacuous, to what we are promised will be an enjoyable spec again? Can't we just have an enjoyable spec as a baseline?

New players (as in leveling from 1-120 in BFA) will also be left with a bad experience of playing Fury until max level if they enjoy any form of tempo in their game-play.

tl;dr the general rotation is slow and boring at the moment and therefore isn't working well. This is because of the slow pace of rage generation. This is easily fixed by increasing rage per auto-attack slightly or increasing rage from abilities.
Forum Avatar
Community Manager
#15 - May 25, 2018, 3:07 p.m.
Blizzard Post
Hey all,

Seph, one of the Game Designers on World of Warcraft, posted a lengthy post about Fury over on the US forum. There's some interesting information in here which I feel is relevant to this discussion, so I've copied it below.

TL;DR: The spec is still being iterated on.

Work is still underway – the spec is currently being iterated on. To give some more information on the overall direction of the spec, here are some points we are looking to address:
  • More tightly adjust/control Enrage uptime and Rage income (both of which heavily impact pacing, flow, and the feel of the spec) throughout the expansion, rather than letting both scale up dramatically as gearing happens.
  • Reduce the amount of stacking damage bonuses on the spec that are available simultaneously, which reduces the desire/need to stack multiple at once, which now feels much worse to do now that virtually all of them are on the global cooldown.
  • Figure out Furious Slash’s place on the spec.

Enrage
Getting mad and being better and stronger is definitely still the fantasy of Fury, and Enrage is staying as the spec’s core base mechanic and Mastery. However, Enrage is changing a bit into 25% Haste (still being adjusted) and 10% Movement speed. It’s currently 100% melee attack speed on live, and was 50% melee attack speed previously on Beta. Haste is something that’s much more felt in terms of pacing (button presses, cooldowns compressing) compared to auto attack speed, which mostly gets covered up by ability animations anyways.

The 20% increased damage taken during Enrage is being removed. Along with this, the 15% passive increased maximum health the spec had is also going away – this was largely added to counteract the increased damage drawback that Enrage had. Now, Fury will simply have the bonus of ~8% more health than other non-tanks in the same gear level, due to being able to dual-wield a pair of 2-handed weapons.

In terms of Enrage uptime, our current thought is we’d like to try targeting Enrage uptime to around ~60%, and not growing wildly throughout the expansion as gear improves. I think there exists a sweet spot for Enrage uptime where it’s not up so often that it’s taken for granted and you don’t have to think at all about playing around it, but also not so rare that it feels bad.

Because Enrage uptime is a function of Enrage duration and Rage generation (which is directly proportional to the time in between Rampages), Enrage duration going up means uptime can get too high, which means time in between Rampages can get too high, which then feels bad. Currently, Enrage duration is back to 4 sec, but Rage generation has increased significantly since this round of Fury changes, so the spec should feel much faster now and the uptime on Enrage shouldn’t feel low.

Raging Blow
Raging Blow now Generates 12 Rage, has 2 charges, 7.5 sec recharge (reduced by Haste), has a 20% chance to instantly reset its own cooldown, and no longer requires being Enraged to be usable. It’s more similar to the Inner Rage version than the live baseline version, where it has no cooldown but requires Enrage to be usable. Raging Blow is doing a lot of things for the spec here:
  • Raging Blow was changed to fit together into Rampage and Bloodthirst, both of which we thought were good and didn’t need any notable changes.
  • Being usable only while Enraged, meant you very frequently pushed it multiple times in a row for 5 Rage each time, which didn’t feel super engaging.
  • Having 2 charges baseline felt better, as a version with either a 6 or 7.5 sec cooldown but no charges meant it collided with Bloodthirst’s cooldown very frequently, which did not feel good. Additionally, charges on Raging Blow is sort of the only actual resource you manage on the spec, since Rage is technically a resource but only spent on Rampage and nothing else.
  • Having a 20% chance to reset itself is the random proc on the baseline spec that adds an amount of unpredictability to the rotation.

Furious Slash
Furious Slash didn’t really found its place on the spec. it’s a no-cooldown free filler that didn’t feel great to press, largely as a direct result of being the lowest priority/impact ability by design. The version of Furious Slash you’re seeing this week is a first iteration, where we tried giving it more tie-ins to your core rotation/mechanics: Generates 3 Rage, has 10% chance to trigger Enrage.

Playing this version some more internally, we thought the button felt better, but still didn’t feel great to press. Additionally, it overlaps some with Whirlwind, which is also a baseline no-cooldown free filler. So we’re going to try removing Furious Slash baseline and making Whirlwind the baseline filler in single-target. Whirlwind has the benefit of having better audio/visual feedback, and generally feeling better to press. It also just fits the fantasy of a whirling blademaster/berserker pretty well. It will generate 3 Rage + 1 additional per target hit, up to a maximum of 8 Rage (this is intended to also speed up the AOE rotation). Furious Slash is being merged together with Frenzy (talent) and becoming a talent in that slot, without the new experimental 10% chance to trigger Enrage mechanic.

Whirlwind's Meat Cleaver effect
Whirlwind will now cause your next 2 single-target attacks to gain the effect of striking up to 4 additional targets for 40% damage. With the changes to Raging Blow, it made sense to add Raging Blow to this effect, but at that point the only thing not included was Execute and Furious Slash, so we've added them too.

Execute
Execute being a Rage spender caused it replace Rampage entirely during execute range (against enemies below 20% health). We are trying a version of Execute where it has a short 6-sec cooldown, and generates 20 Rage instead of spending it, so that it adds to your rotation and helps fuel your Rampages, rather than replacing one of your core buttons. In the current design of the spec, it feels more correct for Rampage to remain the almighty and only Rage spender, and Execute to be a more support role (rather than the spammed Rage spender), even if it deals similar damage to Rampage on paper (because Execute is available less frequently and Rampage Enrages you).

Recklessness
With Enrage no longer coming from Bloodthirst critical strikes and instead coming from Bloodthirst at a consistent 30% rate, the spec no longer has a core critical strike mechanic, and it made less sense to have 100% critical strike chance as Fury’s baseline cooldown. Thinking about it more, 100% critical strike chance would make more sense as Arm’s cooldown, because that spec is slower-paced and more about setting up for fewer, larger hits. Fury is currently much more about flurries of many smaller hits, speed/haste, and Rage generation. To that end, Fury’s cooldown turning into double Rage generation from all sources (and some critical strike chance for flavor) made more sense, both in terms of feel, theme, and gameplay.

As a side note, one of the upsides of Fury’s resource model is that it can support a huge amount of burst/temporarily increased resource income in a way that feels great (something many other specs can’t do as well), because Rampage spends 85% of the Rage bar – we’re consciously leaning into this more with the Recklessness mechanics changes.

As always, thanks for all of the feedback.