Lhivera’s Library

Adventures in World of Warcraft, Dragon Age, the real world, and beyond

Solving Frost's Haste Problem: The Brainstormenating

Tuesday, March 12 2013 at 9:44pm CDT

Some good discussion happened on this topic both in my own previous post on the subject, and in a forum thread I posted today. I wanted to summarize some of the ideas that came out of it to prompt further discussion. I have not reproduced every idea; there were a few that seemed likely to be deemed too complicated to put on a tooltip, and there were a few focused on lifting the Crit cap, something GC has said they don't want to do. Most, however, are represented here in abbreviated form.

Most of these ideas are not mine! Please read the linked posts to see the many people responsible and some more detailed discussion of the ideas.

Excess Haste Conversion

One common suggestion has been to cap Frost's Haste at 50%, and convert any Haste above that point into Mastery or a straight percentage damage multiplier. This has the advantage of being simple and solving the Haste problem completely — it becomes impossible to exceed the cap. But it may make gearing somewhat oversimplified: you don't really have to think about Haste at all. You just take as much as you can find. That may not matter; working around the Hit and Crit caps may be enough.

Have Haste Reduce Cooldowns

If most of our casts can't benefit from haste, perhaps more of our cooldowns should:

  • Icy Veins
  • Frozen Orb (this is one of the simplest and most popular suggestions)

Have Haste Increase Fingers of Frost Procs

This idea came from a couple of different people, but the most completely formed idea was that FOF should work with the RPPM system. Haste thus increases the frequency of our hard-hitting casts, helping make up for the fact that the casts themselves can no longer benefit past the cap.

Improve Mastery

Another idea is simply to make Mastery more attractive so that carrying so much haste won't be as appealing. Some ideas include:

  • Improve the Water Elemental's benefit from Mastery
  • Change Icy Veins to cause bolt spells treat targets as frozen, to allow more spells to benefit from Mastery
  • Apply bonus damage to targets affected by the Frostbolt debuff and/or chilled, rather than to frozen targets

Change Brain Freeze

More correctly, the suggestion would be: remove an instant cast from rotation. But out of Ice Lance and Brain Freeze FFB, the latter seems the more likely. The idea is simply to reduce the amount of damage coming from instant and 1.5 second casts, so that more damage can benefit from more haste. Some ideas include:

  • Replace Brain Freeze with a hard-hitting, short-cooldown rotational spell with a cast time.
  • Instead of proccing an instant FFB, proc an instant Frost-damage DOT. The cast wouldn't benefit from extra haste, but the DOT ticks would; it avoids the problem of extra burst, and gives Frost some multidotting capability, though subject to RNG. (Also makes Frost Bomb a viable option even when the best trinket is triggered by DOTs.)

Change Icy Veins

This is almost certainly in the cards already; Ghostcrawler hinted as much in the Totemspot interview:

The Glyph of Icy Veins helps with that, but there will pretty much always be some instants in your rotation, leading to some haste soft capping, but it is potentially too easy to reach that with the number of temporary haste effects we’ve given out lately. We don’t really like that that Glyph becomes (unintuitively) “mandatory” once you’re at a certain gear level, but it’s a necessary evil for now.

But how to change it? Some ideas:

  • As mentioned above, Icy Veins could make bolt spells with a cast time treat the target as frozen.
  • A variant on that idea would only treat targets as frozen if they were permanently immune to freeze effects.
  • One idea I had in beta was to reduce the haste bonus by half, and compensate by increasing the crit damage bonus. To smooth out the RNG, it could still launch triple bolts (at 1/3 damage), each with a separate crit chance.

Baseline the Icy Veins Glyph Effect

This is one relatively complicated idea I'm including here because it generated a lot of interest and discussion. The basic idea, if I understand Akraen correctly, is this:

  • Frost will always cast various spells (like Frostbolt, Ice Lance, FFB) in triplets, as it currently does with glyphed Icy Veins.
  • Haste, rather than reducing cast times, will instead increase the number of spells cast. So at a certain amount of haste you start casting four bolts per cast, and with more you get up to five, etc.

There are all sorts of potential issues and problems here, but I think there's a strong idea there that's worth exploring.

Water Elemental Enrage Generated by Crits

Another very popular idea! Depending on the strength of the effect, it could happen after every crit, or after a given number of crits. Improves interactivity with the pet. Various possibilities suggested:

  • Simply have it do more damage while enraged
  • Increased casting speed while enraged
  • Enrage enables a new spell that must be cast manually by the Mage and deals high damage
  • Enrage enables a new spell that applies a debuff to the target that amplifies the Mage's damage

Increase Cast Times

A simple suggestion: Increase the cast time on Frostbolt, with a damage buff to compensate. More time spent casting Frostbolt means more value from haste.

Reduce the Global Cooldown

A common suggestion, but probably not practical; Ghostcrawler has mentioned in the past that the game simply doesn't handle very low GCDs very well.

Any of these ideas spur some new thoughts?

Comments

Submitted by Binkenstein on

Similar to a suggestion I've made for Elemental (where Lava Burst would get changed to a 2 second cast time), Brain Freeze could simply go the other way from what it currently does. Rather than providing an instant cast, it increases the damage of Frostfire Bolt so that it's worth casting. A 2-2.25 multiplier would be the rough ballpark, and it also means that the FFB glyph would be useful (although that would likely be changed)

Submitted by Lhivera on

Heh -- I actually made that suggestion myself! I completely forgot to include it.

Submitted by Tiga on

I think a lowered GCD during IV would be justified. It appears that rogues can have a 0.7 GCD. I found a topic on Arenajunkies that has some discussion and references to developer thoughts on sub 1 second GCDs:

http://www.arenajunkies.com/topic/236265-rogue-gcd-1-second/

I'm only proposing this as a quick fix for IV causing the capping, so the lower GCD would only apply about 11% of the time for Frost mages (20s/180s).

Submitted by Citizenpete on

Hey,

this is coming from a progression raider PoV.

1. Switch effect between Icy Veins and glyphed Icy Veins, because glyph of IV has been mandatory for me since the second raid week of mogushan and will be for the rest of the entire addon. If you need haste for questing or the beginning of an expansion, you can simply glyph it then. 

2. I really like the idea of "Have Haste Increase Fingers of Frost Procs", because at a certain point, mastery and haste will play well which each other (haste already increase BF procs), it would make sense and on top of it, it sounds "doable"


Reducing GCD looks pretty dangerous to me, because I´m already casting around 1s with my char (BL, Trinket proc, Troll racial) from time to time and it already feels pretty clunky.

Reducing

  • Icy Veins
  • Frozen Orb (this is one of the simplest and most popular suggestions)

would be a real nightmare for me (and every decent raider), because Icy Veins and Frozen Orb lies on a perfect timeline with Alter Time, some trinket procs, engineerings synapse springs etc. I can afford to wait for Lightweave Embroidery or a trinket for 2nd 3nd or whatever Alter Time occasionally, because I have those stable cooldowns like IV and Frozen Orb. Taking them away would support mindless play (like rppm trinkets already do).

Submitted by Nathyiel on

(it's more like an aggregation of previous idea than anything new)

For me, frost mage game can be defined like this : fast and nervous, control and survivability.
If the last two are always true, the others aren't. If frost is fast with a lot of instant and more than 40% haste, it isn't nervous with the frostbolt debuff and the powerless feeling of haste proc.

I don't think the answers will come from one big change but more of a little bit of this and that.

>> mini-bolt everywhere
i really like this idea but I think it need some change :
if the number of mini-bolt is variable, the haste ratio need to be fix.
if the number of mini-bolt is fix, the haste ratio can be linear.
We can also have a maximum number of bolt with a variable ratio that will change for each cap.

Example:
+1 bolt every 12%, with a maximum of +5 at 60% haste.
damage increase by 1% every 2% haste, reset at every step .
the damage ratio can be calculate easily : (100 /number of bolt) + (%haste /2) - 6% * (number of bolt -1).
doing : 100% to 106%, 50% to 56%, 33% to 39%, ..., 16.6% to 22% then more

Another things, I think this system need to be limited to Frostbolt and Ice Lance. We need a rare big number spell and Waterbolt have a long cast time.

>> Brain freeze change
Brain freeze can put as proc on short cooldown (like Early Frost in cata) : reduce the cast time of FFb by 1s and allow it to be cast while moving.
Your mage bomb can add a 15s debuff : your FFB spell consider the target as Frozen.
It keep bomb interaction and remove an instant spell with not impact for movement.

>> Water Elemental enrage
I really like this one. After X spell crit, your Elemental enrage for 10s, allowing it to Cast a powerful spell.
It have tow purpose : prioritize crit capping and add interaction, with mastery include.

It could be interesting to have a word of GCabout all of this. Just like saying we have read it.

@Citizenpete : haste already increase the proc rate of FoF since FoF proc from Frostbolt. more Frsotbolt = more proc.
 

Submitted by Dagmagi on

Increase frostbolt's cast time to somewhere around 2.2 to 2.5 sec (adjust damage accordingly), should make it difficult to obtain the soft haste cap with current gear.   (It's a bit counterintuitive that the haste spec (frost) starts out with such a low cast time on frostbolt.)

Change Instant cast spells to get a spell damage increase if haste reduces the GCD below 1sec.

(Alternatively, change the haste rating conversion so that haste is less effective.  This is usually reserved for level cap extensions.)

Submitted by Nathyiel on

Increasing Frostbolt cast time won't change anythings as it need 100% haste for cap it.
changing haste conversion can be a way to do it but it can be difficult with healer since haste = mana

Submitted by Lhivera on

Before Cataclysm, Haste being less expensive than Crit and Mastery made sense because while hey increased efficiency, Haste reduced it. Now, however, that is no longer the case. Haste's cost should probably be increased to match the other stats.

Submitted by Millenia on

I like the water elemental enrage on crit thing. One possibility, I think, is to give the WE something like Water Jet, but it only causes the target to take damage as if frozen, rather than give FoF. Of course, there could be a cooldown on the enrage.

Frozen Orb could also get shorter (and more) ticks with higher haste. Could definitely help.

Submitted by Tiga on

Looking at an LFR log from today, a significant part of my damage (about 30%) was from Nether Tempest and in a PvE setting, it doesn't really benefit from mastery, so it made me think about ways to make mastery a bit better (assuming we want that).

One way to modify mastery would be to add an additional effect to it. The effect would be to give each damaging event (dot tick, spell hit) an additional chance to consider the target to be frozen. Of course if the target is already actually frozen (PvP and solo play), it wouldn't increase the damage, but on a root-immune boss, a few % of all hits would benefit from mastery. This keeps mastery from becoming any better for Deep Freeze combos in PvP, but makes it more interesting in PvE.

Submitted by Fröstler on

Here a short collection of ideas. I do not claim that my ideas are complete or balanced but maybe they trigger somebody else to get to a more complete solution.

The problem with shatter is, that as long as shatter remains, it cannot be solved because it is the nature of shatter to have a haste and a crit cap. Turning nobs is only moving the cap in a direction, but the problem still exists.

If we want to keep playstyle with procc'd instant casts we need to remove crit multiplier on frozen targets.
E.g.
* remove crit multiplier but shatter makes more damage for
    a) crit and non crit casts or
    b) for crit casts only.

Or we need to redesign shatter completely.
E.g. to follow the idea of frost bomb. Doing damage on the end of a debuff.
* Shatter can apply a stacking debuff to the target. Whether it can be simply 1 stack (no stacks at all, only a debuff) or multiple stacks.
    a) If the maximum stack size was reached the debuff explodes and does damage.
    b) Maybe we can define a spell that triggers the debuff explosion on demand for multiple stacks or simpley the next cast triggers te explosion automatically.
    c) Debuff can be dispelled (PVP)
    d) To increase mastery scaling, mastery should increase the debuff explosion damage.
Only one stack would feel better, otherwise we need to deal with a ramp up phase. Ramp up is always bad for trash and changing targets.

Some more ideas to soften PVP impact.
E.g.
1) The mage need to buff himself by a spell that can be dispelled (Freezing aura?). With this buff shatter would make more damage.
2) Place cold cloud below targets (e.g. same size as blizzard?). Within this area frost mage makes more damage on shatter, otherwise shatter does less damage.
 

Edit: Sorry, was intended to be a comment for topic "How to Solve Frost's Soft Cap Problem?". But eliminating one cap it also softens the problem with the other.

Submitted by Citizenpete on

Hey Lhiv, as a heroic raider I wanted to give you an update. My Item level is currently 534, I play raid buffed with ~52% haste, ~ 24% crit and 40% mastery. We are currently on lei shen heroic and the whole problem with the scaling is worse then ever before. We are currently practicing P1 and first transition with Bloodlust on the pull. I need not tell you, that I pass the 100% haste mark with Legendary Meta and Troll Berserking very easy and often in the first 46 seconds. The burst difference compared to fire mages is in a 100 - 250k DPS range. Frost up to 350k (420k with mage armour) and Fire up to 600k.

After about 3 minutes, the boss moves to the next phase. At this time, the Fire Mage DPS is at about 260k and my Frost Mage DPS is about 190k. Of course, the other players asks, if I want to switch to fire, because we would then make it possible to survive P1 without Bloodlust and could save it for P3.

To item level 530 I was relatively satisfied with the performance of frost. But now forced in another spec at the final boss is really annoying. Especially since I do not even have the right gear to do so properly.

I expect 5.3 (very cheap 8 item level upgrades) soon and the one or another TF loot. The gap between frost and fire and the scaling issue will become only larger with more gear. Are there any plans or comments on this particular subjects from the devs? Not even imagine how it would run the next Tier (16).

Regards,

Pete @ EU-Anetheron

Submitted by Nathyiel on

On the same subject, I have run a Primordius with 4 haste buff. So I found my-self with 80% passive haste and 132% with Sinister/BL. I have obtain a +200% haste for a few second (BL+Sinister+haste buff) and I think +250% isn't irealist. This fight is good for testing huge haste build.

Having a meta proc feel bursty but it began to feel clunky.

All spell feel like a lost of time as every things is instant and capped to 1s. Frost bomb is the most strange : with a 5s Cd, it was cast FB, cast FFB, cast FB, etc, with 1 spell in between.
From all test I have done on, only 2 spell feel "good" with huge haste: Evocation and Nether Tempest.

And in all this test, my DPS haven't "skyrocket", I have only obtain 10k or 20k more than usually.

My simulation show that Mastery=Haste around 60% passive haste.
It's not that Mastery value more, it's Haste value that drop to Mastery point.

Reflexions :

  • After 40% passive haste, gameplay feel bad as proc don't feel like a burst.
  • Mastery value need a huge boost.
  • As far as I have test it, in T16, Frost will be implayable (worst than in DS).

Submitted by Nathyiel on

I have made some prospective simulation.

I have use the 3 T15H profiles and boost all stat by ~20% for an equivalent ~555 ilevel.

I have fixed intel at 22500, spell power at 12800 and hit at 5100. All 3 profiles use 32500 stat point between crit, haste and mastery, with frost's crit at ~26%.

  • Fire : 322k
  • Arcane : 305.5k    (-16.5k -- 5.1%)
  • Frost : 293.5k       (-28.5k -- 8.8%)

The situation isn't totally out of control but not far.
Fire :

  • Pyro's crit is at 80% for 33.5% of the DPS.
  • ignite, combstion and firebal count for 16% to 19%.
  • burst is around 1.25M dps

Arcane (haste) :

  • Missille and Bast count for 30% of the dps, each.
  • Arcane's blast cast time is at 1.15s in average.
  • burst is around 0.73M dps

Frost :

  • Frostbolt have 24% of the DPS, IL and FFB are at 20% and 17%.
  • Tempus Repit uptime is around 46%
  • Frostbolt's cast time is at 1.13s in average
  • burst is around 0.81M dps