Shang's Arcane Guide (updated for 6.1)

[Soulcrafty]

I meant for when you are already at 4 stacks of arcane charge, how do you best handle missiles? Do you weave missiles or just dump them all?

Example: missile AB missile AB missile AB barrage VS. missile missile missile barrage

[nickseng]

I usually check my mana level.

So at 4 charges:
1. Is my mana above 93%
- if so cast Arcane Blast
- else cast Arcane Missiles
2. Repeat 1.

I am a bit shaky on the burn phase though. Most people say to forget about AM under 70% mana unless at 3 stacks, but I think I read Frosted saying that we should still use AM regardless (don't quote me on this, I'm still trying to find the post he made)

[adrift]

I personally don't think saving AM under 70% is optimal. Basically, its done to keep your burn phase lengths consistent and not de-sync your CDs. IE, you get tons of procs in your first burn so Evo is delayed - then when you burn with the second PC and don't get the same AM procs the Evo no longer lines up like you need it.

However, you can achieve the same thing by simply using all AM procs during burn and Evoing based on time and not mana %. You will be at higher average mana during the burn, and simply Evo at 60-70% mana while burning all AMs instead of banking.

I haven't actually simmed this since I'm not amazing with customizing on SC, but I can't think of any reason that it wouldn't be better.

[Wilderness]

There was a sim done, but iirc it was just comparing holding AM under 70% vs completing a full burn and potentially de-syncing CDs, so it may be possible that simply using Evo after a set time rather than mana percentage would be better. It would depend if the damage you gain from those extra AM procs beats out the damage from the last part of the burn with its AB spam.

Its an interesting idea, but I'm also not comfortable enough with SimC to do much more than run basic sims.

[Kurii]

I'm not sure if it's optimal because I don't have the Sims, but I have WeakAura graphic that shows a 30 second "burn" timer after I lay down a PC (which also has a 12 second PC timer).

So basically...

AB to 4 stacks
PC / AP / Trinket (12 second timer)
Initiate burn sequence (30 second timer)
Evocation at the end of burn sequence
Conservation phase
Repeat (every other rotation will grant you an AP buff if glyphed)

[Cam]

I don't understand why Missiles are used during the 2nd, 4th, 6th burn phase (Without AP)

Surely, because of our mastery, casting Missiles at 70-80 percent mana during our burn is a DPS loss since we don't have Arcane Power up. Wouldn't it be better to keep the AM procs and use them during our Conserve phase when we will be closer to 100 percent mana so they have a higher DPE?

I find, with my very limited experience, that Missiles can cause us to extend the burn phase too long... sometimes I get 5-6 missile procs during the burn and all of a sudden I have been in burn phase for 30-40 seconds and my mana is still over %70.

Also besides hitting like a truck, Missiles also perform the job of stabilizing our mana in the conserve phase letting us get extra AB casts in at 4 charges. If you end your burn phase with a few stacks of Missiles! then your next conserve phase can last for ages allowing you to get 6+ AB casts in at 4 stacks.

Anyone got any maths to support/rebuke what I just said?

[TLTeo]

Mastery or no mastery, burning is always more dps than conserving because of how Arcane Blast works. This has been the case since its introduction in TBC. This means that AM extending the burn is actually a good thing - you are spending more time doing more damage. If this wasn't the case, the burn phase wouldn't exist in the first place.

The reason for not using AM below 70% mana is to sync our cooldowns more efficiently; think of a burn as a cooldown, just like every other spec, you want to stack however many cooldowns together as you can, meaning you want to PC (if talented) and AP during a burn. This doesn't mean that when AP falls off, you want to stop burning; it would be like clicking off DMC's 20 second proc after the 15 seconds of AP are over, if that makes sense. If your burn phase is lasting too long then yes, ignore AM, cast 1-2 AB's and Evo (that's what the 70% treshold for AM does in practice), otherwise there's no reason not to cast AM.

Also, you will build AM naturally during your ramp up conserve phase, which is better than going into conserve with 0 charges and 2-3 AM stacks, which will probably result in munched procs. I can see why having a longer period at 4stacks weaving AB/AM might be good with 2t17, but it still isn't a dps gain to neglect the burn.

I simmed it on 1 and 3 targets for both the 680 default profile, as well as my own gear (679 with only 2t17), and yeah it's a dps loss overall:

[Wilderness]

Just to tack onto what TLTeo said - just try to remember why we have the "70% rule", and that its not a hard rule at all. Its there to help keep everything synced up, and that has been shown to be a dps-gain. But sims are looking at the average results over thousands and thousands of iterations, so you will absolutely run into pulls where you get great AM RNG and that leaves you sitting at 80 or 85% mana well into your burn phase. When you find yourself in those situations, then you will probably want to start ignoring AM earlier than 70%, to keep your CDs synced up. Managing your mana and making those kind of decisions on the fly is what really separates average, good, great Arcane mages, so its something you have to just get a feel for and decide when your burn has started to go on too long.

[Malthouse]

Hey, just wanted to mention that breaktheice's frost guide talks about banking Ice Nova charges for PC that should work perfectly for Supernova. The explanation is pretty simple to follow -- http://altered-time.com/forum/viewtopic.php?f=5&t=1110" target="_blank (search for "Prismatic Crystal and Ice Nova") -- and it should map 1:1 to the prismatic crystal/supernova mechanic.

[Tumporai]

First of all: Great Guide! I really enjoyed reading discussions on the forum while learning how to play my mage.

I tried to do some calculations to compare the damage of Nether Tempest and Supernova on a different number of targets. The formulas I was using for the damage are:

Nether Tempest at 4 Arcane Charges:

Code: Select all

0.4092 * Spellpower * Number of Targets * 3

(4 * 50% more damage from Arcane Charges = 300% -> * 3

Supernova:

Code: Select all

1.9 * Spellpower * (Number or Targets + 1)

(+ 1 target for the 100% additional damage Supernova does on one target)

To get DPS out of this I divided the total damage done by 12 for Nether Tempest (damage ticks over 12 sec) and by 25 for Supernova (cd of each charge of Supernova is 25 sec).

It appears that NT pulls ahead after just 3 targets (provided they are stacked). Is this possible or am I doing something wrong?

[Wilderness]

Comparing talents in a vacuum isn't that useful, because what other talents you pick and how they interact is what will determine what is strongest in different situations. Its pretty important how well NT and AO work together, as well as SN and PC, so you can't simply say that one talent is better than the other in a row. That's why sims are so useful, since they let you test various combinations with different conditions and see how they stack up.

I didn't really look at your math, but I can tell you that 4 x 50% damage is 200%, not 300%.

[Seprest]

A 200% increase gives you a total of 300% base damage. 100+200=300...

[Tumporai]

Comparing talents in a vacuum isn't that useful, because what other talents you pick and how they interact is what will determine what is strongest in different situations. Its pretty important how well NT and AO work together, as well as SN and PC, so you can't simply say that one talent is better than the other in a row. That's why sims are so useful, since they let you test various combinations with different conditions and see how they stack up.

I didn't really look at your math, but I can tell you that 4 x 50% damage is 200%, not 300%.

I guess you're right about simming. Anyways, I had fun doing the calculations. By the way, who maintains the arcane mage APL?

Seprest already wrote it but I will repeat why 300% is correct. The tooltip of Nether Tempest says "[...] Damage increased by 50% per Arcane Charge at cast time." Thus 100% of the base damage at 0 charges, 150% at 1, 200% at 2, 250% at 3 and 300% at 4.

[Komma]

Nether Tempest at 4 Arcane Charges:

Code: Select all

0.4092 * Spellpower * Number of Targets * 3

(4 * 50% more damage from Arcane Charges = 300% -> * 3

(Number of targets + 1) because it ticks double on the main target.

That said, I think Wilderness is right that it's really hard to compare them in a vacuum. Between the different number of GCDs spent, mana regained (since they're both mana positive), potential AM procs and the difference in haste scaling, there's a lot of factors that create uncertainty. Which kinda explains why we use simulations for comparison!

That said, I think we'd all appreciate in-depth analysis on these issues, if just to understand better their strengths and weaknesses.

[aznxk3vi17]

Does the Arcane Blast/Missiles alternating still take place during the burn phase? That is, when Prismatic Crystal is off cooldown, given the DPS of missiles, would it be better to spam all available missiles before using Blast, or should we still aim to Blast until below 93% mana before using Missiles?

[FPSFrost]

Is getting haste to 900 an important factor for playing arcane? Im currently sitting at 346, with a greater haste gem socketed. Without it, im at 296

[Komma]

Is getting haste to 900 an important factor for playing arcane? Im currently sitting at 346, with a greater haste gem socketed. Without it, im at 296

Arcane has no haste breakpoints at all.

[Laneil]

Presence of Mind DOES work with Polymorphism.

[Lammer]

The reason for not using AM below 70% mana is to sync our cooldowns more efficiently; think of a burn as a cooldown, just like every other spec, you want to stack however many cooldowns together as you can, meaning you want to PC (if talented) and AP during a burn. This doesn't mean that when AP falls off, you want to stop burning; it would be like clicking off DMC's 20 second proc after the 15 seconds of AP are over, if that makes sense. If your burn phase is lasting too long then yes, ignore AM, cast 1-2 AB's and Evo (that's what the 70% treshold for AM does in practice), otherwise there's no reason not to cast AM.

but there is an opposite point of view on the length of burn phase in Dutchmagoz guide:

There's no length where it's too long, as long as you're not burning at like 60-70% mana. If you stay at 85%+ mana constantly never dropping stacks, that's great! Just watch your evocation timer for the next burn and don't start it too early.

so to summarize:

Opinion 1 (TLTeo & Wilderness): Burnphase is a CD and must be synced up with other CDs. We don't care about mana percentage, 70% is just an indicator to help, so u can use Evocate even at 80% for sync.

I suppose it lasts 22-30 sec depending on AM procs during burn phase

Opinion 2 (Dutchmagoz): We are burning endlessly, but not when mana is below 60-70%.

TBH I played like in the second scenario and had a stable feeling, that I was doing wrong, that is: burn phase lasting up to 1 minute and using AM at (50:70)% interval, because I did not want to lose a potential AM proc being at 3 AM stacks; and never used Evocate earlier than 50% mana.

So could you please help me to clear this question? I would like to play like TLTeo suggests, but what should I do in a situation, when I am between 50 and 70% mana, I have 2 AM procs and it is too early to finish burning. May I use AMs in this interval? I will try to hold the burn phase for 25-30 secs for sync.

And ty for the guide and your answers. A very good forum

[TLTeo]

The "watching the evo timer" part is key. You still want to sync that with PC/AP (specifically have it read about 15-20 seconds after), and depending on things like fight length, fight mechanics (i'm looking at you Blackhand bombs), you may need to change your burn phases accordingly. Obviously burning as long as you can is nice, but not if that comes at the cost of lower AP uptime, or losing one PC cast.

The idea of skipping AM is just one tool of being absolutely sure your cooldowns don't get de-sync'd.