[Fire] Kver's Comprehensive Guide to Fire v1.0

[Kver 65]

Introduction

The purpose of this guide is twofold, on one hand it attempts to elaborate and expand on fire mages, a spec most players incorrectly consider to be trivial and simple. On the other hand it proposes some interesting theories. The fire mage community has in due time declined in numbers, and the end result is an afflictive absence of discussion. A popular opinion is that there really are no possible alternative ways of playing fire mage, an opinion I vehemently oppose.

 

When can I safely go fire?

Fire starts to become solid when you have the 4 set and 45% crit chance. Having high crit is a requirement to smooth out the RNG. If you have a lot of haste/mastery you’re better off going frost. The reality is that Fire is currently incredibly crit dependent. If you have low amounts of crit you will have fights where even with perfect play you will simply have horrible performance.

 

 

Builds

 

Standard build

Hit to cap > Crit > Haste > Mastery

This is the conventional build. It focuses on as much instant Pyroblast damage as possible. By capitalising on haste you reduce the global cooldown and increase your cast speed, increasing the DPET of your pyroblasts by a lot and making it easier to keep up the 2 piece by generating more pyroblasts in a smaller time frame.

This build rocks on single target and spread out multi target, and is slightly better at movement due to the ability to finish casts and cast faster evocations, but loses out on AoE and stacked cleave damage.

This is the build I would advise to most players because it is simply unfeasible to go for that much haste at lower gear levels.

 

The Haste Cap Build

Hit to cap > Crit > Haste to 14100 > Mastery > Haste

This build assumes that by keeping up the 2 piece bonus, combined with meta gem and bloodlust, you can invest some of your haste in mastery instead. At 14.1k haste having 5 stacks of the 2 piece buff gives you 50% haste. Combustion's increased damage and the increased ignite values of spells are stronger than the increase haste gives past 50%, due to instant casts no longer scaling with haste beyond that point. In addition, haste increases mastery’s relative potency.

 

2 * 1 = 2

A 2.2 * 1 = 2.2

B 2 * 1.2 = 2.4

Assume that the first variable is haste and the second value is mastery. In both scenarios A and B an absolute value of ‘0.2’ has been added, however due to the relative value of the ‘0.2’ compared to the original value, the increase is largest in B.

I would personally advise this build to people with very high gear levels, who can gain a sizeable amount of mastery from this.

 

Mastery build

Hit to cap > Crit > Mastery > Haste

This build focuses on two things:

- Higher ignite value boosts individual spells’ DPET

- Higher ignite value boosts combustion damage

Single target damage is a lot lower than haste, but this build dominates on AoE fights such as TFP and Spoils. It has more issues with mobility than the haste build; invocation casts are slow and you will need to interrupt casts more often to avoid damage. Keeping up the 2 piece becomes difficult as well. Only use this build if you know what you are doing.

 

 

Something to help you choose and think

 

What do the two other secondary stats do?

Haste does a multitude of things: it increases your cast speed, the frequency at which DoTs do damage(“ticks”), it reduces the global cooldown and it increases your mana regeneration(Irrelevant for Fire).

The strength of haste lies in the combination of casting speed and the global cooldown reduction: together these two increase the rate at which you can pump out pyroblasts, making it easier to keep up the 2-piece set bonus, which in turn gives you more haste. Once you hit 50% haste(in any situation) your global cooldown is capped at 1.0 second, reducing the marginal increase of haste.

 

While it is certainly tempting to look at simple DPET calculations to justify stacking haste, there is more to Pyroblast’s haste scaling. These calculations generally assume that the DoT runs its full course, which is almost never the case. For example: you cast a pyroblast, which leaves an 18 second DoT. 2 seconds later you cast another pyroblast, overwriting that DoT with another 18 second DoT. Ultimately, you only gain 1 extra tick from this, and you have effectively “wasted” 16 seconds of the DoT. This does not matter for your DPS, because you will never delay a pyroblast cast for the DoT, but it matters tremendously for defining the efficiency of haste, because it means that haste’s value is diminished the more often you cast a pyroblast.

 

Mastery is simply a flat out damage increase in the form of a 4 second DoT(Ignite) which scales linearly. The ignite can be spread to other targets with inferno blast, making mastery very strong on cleave. Because combustion is based on ignite damage, this automatically makes mastery stronger for combustion as well.

 

 

Advanced section

Warning: some of these should only be used if you know why you’d use them. I invite people to think about what I describe here and come up with a conclusion of their own.

 

It is incredibly important to maintain the stacks of the 2 piece buff. It is called Potent Flames and gives 750 haste per stack up to a maximum of 5 stacks. I highly advise you to track this buff; you will want to keep it up especially on higher stacks. This means that you will sometimes cast a pyroblast first, and then evocation or living bomb directly afterwards. Do NOT go out of your way; don’t delay evocation or bombs, and never cast scorch as a primary nuke to keep it up. Keep in mind that the Pyromaniac buff lasts 3 seconds longer than the mage bomb.

 

Pyro gambling

If your 2 piece is on 5 stacks and almost running out, you can opt to “gamble” for a pyro proc by using an inferno blast proactively to effectively “win” a global. The downside is that if the fireball doesn’t crit you lose DPS and the 5 stacks. However, the presence of a large intellect proc reduces the risk. You’ll need to decide for yourself when you think this is worth it.

 

Advanced bonus GCD pull

The idea is that by using a speed boost to move towards the boss you can reduce the travel times of pyroblast to make 5 pyroblasts hit within 4 seconds. You will need a large speed boost to practically end up inside the boss, and you will have to use combustion right away. You can use Alter Time to teleport you around as well: Alter Time halfway towards the boss, run away from the boss when you cast pyroblast, when Alter Time resets you will be closer to the boss, now keep running towards the boss while shooting out pyroblasts.

Pyroblast

Pyroblast

Pyroblast

Pyroblast

Alter Time

[speed boost]

Pyroblast

Pyroblast

Combustion

 

Advanced mage armor pull

Theory: the final average 450k damage from a PoM pyro proc is smaller than the large % increase to ignite and combustion. You will have a very high crit chance due to trinkets, and 2 guaranteed crits due to the 4 piece buff. Before the pull, make sure you have mage armour on already. Do a pull as you normally would by getting buffs, entering AT, spamming pyroblast, then reset AT, cast pyro, pyro, PoM Molten Armor, combustion, continue with the normal rotation.

Make sure that you keep running towards the boss, the closer you are the easier it gets.

The following macro should make it easier to cast both PoM Molten Armor and Combustion using one button. You will have to press it multiple times however.

/castsequence reset=5 molten armor, combustion

You can use the following alternative. easier macro, which comes at the cost of 1 PoM pyro. To do this you need to exclude Presence of Mind from your Alter Time Macro.

/castsequence reset=5 presence of mind, molten armor, combustion

You’ll need to press it multiple times in order to cast all of them, an alternative approach is to just mash it. With this macro you simply walk towards the boss, casting all your pyroblasts and pressing that macro once you are done.

 

Math:

Mage armor = 3000/400 = 7.5% base mastery (even more with pboi, rating*1.09 for example). Let's assume a PBoI is present and take 8% mastery.

(Mastery % + MA%) / mastery % -1 = combustion damage increase. Example: ((20+8)/20 - 1) = 40%
Average pom pyro damage = pom pyro damage * (crit chance / 100). Example: 650 * (80/100) = 520

To be worth it in this example, combustion needs to do at least 520k more damage to make up for the loss of the PoM pyro spent on Molten Armor.

520k / Amount of combustion ticks. Example: 520k / 47 = 11k per tick, so divide by 1/2/3/4/5 for 1/2/3/4/5 targets. The more targets, the stronger Mage Armor gets.

So in total you need a 11k*5 = 55k higher ignite. In this example we assumed 800k damage per pyroblast(Which is very high, BiS levels), you would get a 2.((800k*3+520k)/4)*0.2 = 328.5k ignite. The mage armor increase here would give 328.5*0.4=131.4k extra ignite.

This does not account for the loss of the crit from molten armor during your pyroblasts, but with the crit recommended in this guide and trinkets and 4 set up(As you are supposed to) this is negligible for RNG.

 

Combustion delay

Most mages use combustion on cooldown, however on a fights that last longer than 240 seconds it is a large DPS gain to delay your combustions until bindings procs. This is due to the large bonus to pyroblast damage, the increased crit chance on pyroblast, and the increased crit chance on combustion. You are not using less combustions than you would if you used them on cooldown, but you make both your combustions and pyroblast torrents significantly stronger. If a fight lasts shorter than 240 seconds, cast combustion on cooldown, otherwise you will "waste" a combustion.

 

 

 

Closing remarks

Special thanks to Dutchmagoz for surviving the torrent of ideas I threw at him and coming up with some interesting suggestions of his own. Please note that this guide will be altered constantly. If there is anything you think is poorly worded or incorrect then please respond constructively and back it up with factual information. Do not claim that B is wrong and A is correct because you “feel” it is or because you read it somewhere.

[Akraen 230]

If Mordent Evenshade wouldn't kick me out of his inner Kaldorei/Shen'drelar circle for using too much fire magic, I would use this guide.

[Kver 65]

This is a theory I came up with after Akraen made a joke about rotations. Varieties of this have been tried a lot in the past(Famous Mage Armour Scorch Thok parse) but as with many things Fire, it seems that nobody has delved into it. You can consider this a continuation of my haste-cap build, so if any of the details down below don't make sense to you, read the guide first.

 

Do note that I am not a brilliant mathemagician or somebody who writes everything down in nice and tidy spreadsheet. If you want shiny graphs, I'm afraid I'll have to disappoint you for now.

 

I'm looking for people to test out my new theory. Use the following stat distribution as given in my haste capped build: Crit > Haste to 14100 > Mastery > Haste, and use Molten Armor. Change the following thing in your rotation:

Always use Scorch unless the Meta Gem and/or Bloodlust are up, then switch to Fireball.

 

Using Scorch over Fireball is an 18% DPET loss, which needs to be made up for by more DoT ticks, higher 2 piece uptime(Especially on 5 stacks) and more Pyroblasts. This is only a few pyroblasts(No more than 10 on a 4.5 minute fight), not counting the increased amount of casts and ticks due to the 5 stacks of 2pc.

 

Some early testing indicates that it becomes very easy to keep the 2 piece up for a very long period of time, and that the increased number of pyroblasts is higher than the 10/4.5 minutes you need, but due to the difficulty in modeling Fire this has to be proven by actual in-game testing. 

 

I would like to urge anybody who is interested in this to send me logs of Malkorok parses(Difficulty doesn't matter) where they use this rotation.

 

This is just a theory that I would like to propose and discuss, do not take it for a 100% guaranteed DPS increase in situation [x].

[Frostedmages 3]

What's the difference between the standard build and the haste to cap build? I tried to throw together a set of heroic + random HWF gear that maximizes crit while also trying to get the 14100 rating and could only get to around 11k haste.

 

If the goal is higher 2pc uptimes, wouldn't it be better to prioritze 14100 haste > crit > mastery > haste? Or to just run the standard build and try your idea with scorch/fireball weaving.

 

I was able to throw together a BiS set that has the 14100 haste while mainintaining 16.4k crit rating by enforcing 14100 haste > crit. It was about a 2k loss in crit rating compared to just crit > x, which is around 3% I think (600 rating per crit? I can't remember off the top of my head)

Edited March 18, 2014 by Frostedmages

[Kver 65]

What's the difference between the standard build and the haste to cap build? I tried to throw together a set of heroic + random HWF gear that maximizes crit while also trying to get the 14100 rating and could only get to around 11k haste.

 

If the goal is higher 2pc uptimes, wouldn't it be better to prioritze 14100 haste > crit > mastery > haste? Or to just run the standard build and try your idea with scorch/fireball weaving.

The BiS sits above 14100 haste(16k I believe), in the haste to cap build you would use a different chest(Thok HC WF) for some extra crit and mastery. To be honest, it really only matters at very high gear levels or with a BBoY, but if you get above that amount of haste then you'll be over the GCD cap with a full 5 stacks of 2 piece. Haste above GCD cap isn't "as good" as the haste before it. 

 

In my opinion you'll always want to prioritise crit though, it is definitely our strongest stat. The entire idea of Scorch is actually fishing for more crits, so I see no reason to prioritise haste. The less haste you have, the stronger critfishing gets.

[Frostedmages 3]

The BiS sits above 14100 haste(16k I believe), in the haste to cap build you would use a different chest(Thok HC WF) for some extra crit and mastery. To be honest, it really only matters at very high gear levels or with a BBoY, but if you get above that amount of haste then you'll be over the GCD cap with a full 5 stacks of 2 piece. Haste above GCD cap isn't "as good" as the haste before it. 

 

In my opinion you'll always want to prioritise crit though, it is definitely our strongest stat. The entire idea of Scorch is actually fishing for more crits, so I see no reason to prioritise haste. The less haste you have, the stronger critfishing gets.

 

Are you sure? I'm looking at it right now and it's got 18.9k crit and 11.7k haste. I wonder what we are valuing differenty here. Do you have a profile for it?

 

Here is a list of the gear I chose.

http://pastebin.com/QBBbWd8i

[Kver 65]

http://www.askmrrobot.com/wow/gear/24acda86-2712-48fa-9f6c-38aa5bae243d

 

Is the BiS I would propose with capping haste. Switching out the Spoils chest for the Thok one increases your critical strike, too. 

[Frostedmages 3]

I found where I made my mistake in gearing, makes sense.

Edited March 18, 2014 by Frostedmages

[Magi 0]

I am currently in a newer raiding guild that is currently 11/14 and planning on finishing SoO this monday. So i have noticed my damage in frost tapering down a little bit and have tried fire previously with an ok amount of success. I am currently 547 ilvl and once again am thinking back to switching to fire. My big question is that I am currently using the Spaulders of Kor'Kron Fealy at ilvl 561. I could have the 4 piece bonus but i would have to downgrade to the tier shoulders to do so. I was wondering whether or not i could still see considerable success without the 4 piece or if i should wait until i get either the shoulder/legs/helm from regular or flex SoO to start going fire. 

 

Armory Link: http://us.battle.net/wow/en/character/magtheridon/Mag%C3%AD/advanced

[Kver 65]

I am currently in a newer raiding guild that is currently 11/14 and planning on finishing SoO this monday. So i have noticed my damage in frost tapering down a little bit and have tried fire previously with an ok amount of success. I am currently 547 ilvl and once again am thinking back to switching to fire. My big question is that I am currently using the Spaulders of Kor'Kron Fealy at ilvl 561. I could have the 4 piece bonus but i would have to downgrade to the tier shoulders to do so. I was wondering whether or not i could still see considerable success without the 4 piece or if i should wait until i get either the shoulder/legs/helm from regular or flex SoO to start going fire. 

 

Armory Link: http://us.battle.net/wow/en/character/magtheridon/Mag%C3%AD/advanced

The 4 piece is essential, it's incredible strong, equip it(sadly).

[Methusula 17]

So Frosted and I were discussing the horrendous state of SimC regarding Fire's action list. It is embarrassingly poor, and I can't help but think there is a significant disparity between the stat weights simC gives us, and what is actually optimal.

 

Fire scales with crit, we can all agree on that. And as Kver pointed out, the value of haste and mastery is pretty cut and dry. The concern I have is given the values we have, Fire shouldn't be scaling as well as it is in an actual raid. I get the sneaking suspicion that simC severely undervalues int.

[Kver 65]

So Frosted and I were discussing the horrendous state of SimC regarding Fire's action list. It is embarrassingly poor, and I can't help but think there is a significant disparity between the stat weights simC gives us, and what is actually optimal.

 

Fire scales with crit, we can all agree on that. And as Kver pointed out, the value of haste and mastery is pretty cut and dry. The concern I have is given the values we have, Fire shouldn't be scaling as well as it is in an actual raid. I get the sneaking suspicion that simC severely undervalues int.

Scaling compared to a raid dummy, or? First thing you'd have to do is actually make sure the default APL contains correct combustion building and heating up camping, since these are the 2 main points it currently sucks at.

 

As long as there is no HU camping, it'll always severely undervalue haste, and it'll never actually have a 2 piece uptime worth mentioning. The core issue with SimC at the moment is that it's almost as if it is a completely different spec; it bears little resemblance to how it's played in the actual game.

[Methusula 17]

Scaling compared to a raid dummy, or? First thing you'd have to do is actually make sure the default APL contains correct combustion building and heating up camping, since these are the 2 main points it currently sucks at.

 

As long as there is no HU camping, it'll always severely undervalue haste, and it'll never actually have a 2 piece uptime worth mentioning. The core issue with SimC at the moment is that it's almost as if it is a completely different spec; it bears little resemblance to how it's played in the actual game.

 

 

Yeah, it wasn't banking pyro's for AT, was canceling AT for ANY movement, no 2pc interplay whatsoever.

[Kver 65]

Yeah, it wasn't banking pyro's for AT, was canceling AT for ANY movement, no 2pc interplay whatsoever.

The usage of the past tense here concerns me greatly. 

 

But yes, basically once you actually make that APL flawless for a regular patchwerk fight, that's when you can look at stats. As long as all the core mechanics are actually baked in correctly(You know, how heating up works, ignite's "renew" mechanic etc) then it won't undervalue int at all. What you could look into is the gear for which it rates stats though, for some builds it might relatively undervalue int, but that is in comparison to other stats. Then again, we have since 5.0 been postulating as a rule of nature that 2 secondary > 1 int.

[Methusula 17]

The usage of the past tense here concerns me greatly. 

 

But yes, basically once you actually make that APL flawless for a regular patchwerk fight, that's when you can look at stats. As long as all the core mechanics are actually baked in correctly(You know, how heating up works, ignite's "renew" mechanic etc) then it won't undervalue int at all. What you could look into is the gear for which it rates stats though, for some builds it might relatively undervalue int, but that is in comparison to other stats. Then again, we have since 5.0 been postulating as a rule of nature that 2 secondary > 1 int.

 

 

Yeah, that's a concern I have. I'm not sure that every secondary is better than int. I'm not sure what you want out of my use of past tense. I was using in as it related to Frosted's findings yesterday, so past tense. Nobody has changed anything to my knowledge, so it still holds true, I was just specifying the interaction we had yesterday.

[Dutchmagoz 30]

From my personal experience and calculations, crit and hit are the only stats for fire where 2*seconday > 1*int.

[Methusula 17]

From my personal experience and calculations, crit and hit are the only stats for fire where 2*seconday > 1*int.

 

 

That's what I'm thinking. I see a lot of results from simC and especially from AMR with default values, that show all 2*secondaries above 1 int.

[Frostedmages 3]

The usage of the past tense here concerns me greatly. 

 

But yes, basically once you actually make that APL flawless for a regular patchwerk fight, that's when you can look at stats. As long as all the core mechanics are actually baked in correctly(You know, how heating up works, ignite's "renew" mechanic etc) then it won't undervalue int at all. What you could look into is the gear for which it rates stats though, for some builds it might relatively undervalue int, but that is in comparison to other stats. Then again, we have since 5.0 been postulating as a rule of nature that 2 secondary > 1 int.

 

The core mechanics for SimC are done pretty well. And all of the actually coefficients and stuff are taken directly from the game client files, none of that is input manually. The main problem with Fire is the APL. There may be some improvements that can be made with regard to optimizing ignite mechanics, but SimC employs a general "ignite-like mechanic" (because it's a general mechanic many classes employ), so any major problems with it should not exist. Once we have a better APL we can start looking at stat weights more critically. I've started work on a Frost APL update and hope to have it completed within the next day or so. After I finish that and have it coded into the SimC syntax, I'll start working on a Fire one.

 

If you'd like I can have a look at the code for how HS/HU are coded in and dealt with, if you think there is a problem there.

 

What do you mean by "What you could look into is the gear for which it rates stats though"?

Edited April 16, 2014 by Frostedmages

[Kver 65]

Yeah there are a few big, big issues with the Fire APL and how it banks Pyros.

 

What I mean is that stat weights influence stat weights, if a specific gear set heavily undervalues int then perhaps it has something to do with the set's stats rather than Simc.

[bugmenot 0]

At 14.1k haste having 5 stacks of the 2 piece buff gives you 50% haste.

 

I cannot convince myself of this breakpoint:

 

# Back calculate from 50% and remove spell haste buff

a = (1.5 / 1.05 - 1 ) * 100

 

# Calculate haste rating using 425 haste rating / % haste

b = 425a

 

# Subtract haste gained from 5 stacks of 2 piece and ceil

ceil(b - 5 * 750) ceil(b - 5 * 750 * 1.09)

 

I am calculating the breakpoint to be 14465 instead of 14100. Where did I go wrong?

 

Nevermind, I forgot to factor in the 9% haste amp from heroic warforged pboi.

 

I now am calculating the breakpoint at 14127 which I assume the theorycrafters are rounding down to 14100 because it's close enough and easier to remember.

Edited April 28, 2014 by bugmenot

[Kver 65]

We picked 14100 because with the HWF PBoI I got 14105 or something silly and 14.1k seemed easier to write down than 14105. We're at the levels of gear where 14105 rather than 14.1k means only 1 in a million people won't hit it when they go for 14100.

[Liquidsteel 279]

What's the situation with Inferno Blast and Pyro Chaining when you have the 4 set?

 

Queuing Fireball into Pyroblast with Pyro! and Heating Up can often lead to quite a long period of GCD's where all I'm doing is Fireball - Pyro - Fireball - Pyro. Sometimes refresh bomb in the middle and it's here where I get a second to assess if Inferno Blast is needed.

 

Leaves Inferno Blast off cooldown for prolonged periods of time. 

 

Don't want to munch any Pyro! or Heating Up procs, but am I wasting the 4 pc by not casting Inferno Blast during these periods for the guaranteed pyro crit?

 

Gut feeling tells me to just keep stringing Pyro's until a window presents itself to use IB, but would love confirmation that this is the way to do it.

[Liquidsteel 279]

Frost mage only forum?

[Berlinia 168]

Gut feeling is correct... Just saw this post xD

[Liquidsteel 279]

Gut feeling is correct... Just saw this post xD

WTF why necro old thread nab.

 

Seriously, it took 5 months for a response... arcane is better anyway frost idiots.