Well, if the object is to maximize dps for an AoE situation then being forced to wait for the server to tell your client that the channel has completed is in direct opposition to that goal.

Also never had to /stopcast Blizzard before because Blizzard had horrible dps since it was still using the old AoE calculations, now that it's been changed and in coordination with all the new haste effects it's dps is getting to the point where it is at least somewhat competitive with AE spam. Add IV into the mix and it should actually be at the point where a Deep Frost mage will do more dps with an IVed Blizzard then they can get from using AE.

hate to bump this but i havn't gotten a response. at 1430+dmg 35% crit, hit capped and 17 haste (slightly over 1%) i calculate that fireballx2/fireblast is more dps than fireball spam, but by 3% which is ~50 dps.

As all channeling spells, there is no "you can't cast that yet" server-side command. Even pre-2.3 you could break an AM/Blizz by casting another spell without /stopcasting

Isn't the '10 spell penetration' a reference to the first part of Arcane Subtlety's description, ie "Reduces your target's resistance to all your spells by 10" - which as far as I can see does show up on the character sheet as 10 Spell Penetration?

True but its very hard to tell when it is going to end when you have a bunch of haste active since the bar just stops and since the bar stopping is from a spell complete command from the server that means effectively you either are guessing where to start or waiting for that command from the server and either way you are losing dps.

Couldn't you circumvent the casting bar problems with Blizzard much like the way shadow priests do with Mind Flay?

/cast [nochanneling] Blizzard

You can spam that and it wouldn't recast until you aren't channeling anymore.

Well that is strange. I was initially going to trash you for thinking this, but testing shows me that without popping haste consumables, i'm getting it as a 3-5% increase (more haste-less haste gear), more with imp fire blast. Someone test this more but doing napkin testing my fire blast is doing ~1100-1200 non crit, with fireball 1900-2000 (4 peice t6), non scorched/buffed, which means that ignoring haste gear weaving it in is beneficial.

Hrmph, EJ masterminds go at it!

Edit: Also, when/did the description of permafrost change? It seems to imply that speccing into it reduces movement speed AND attack speed per point, wowhead has no additional info, and thottbot seems to be hinting at it also affecting attack speed

I'm currently weaving it myself but there is one thing I'm quite unsure about and that is how it is affected by the new cast-mechanics because to me castmechanics seem to be better when spamming one spell and not swapping spells (seems like there is some kind of interpolation/extrapolation at work there) which atleast FEELS like whenever i weave inn a fireblast, atleast with a 200+ ms ping, that I'm getting a bigger pullback/lag than what would've happened pre-2.3. Might just be my head playing games with me of course.

I've definitely noticed that as well. Especially noticable with Ice Lance, before 2.3 you could be lucky to get it to hit the same time as frostbolt when comboing it on a frozen target, now it almost always lands before frostbolt. Same thing with fireblast.

Issue is that [nochanneling] is determined client-side so you are still waiting for the server to tell your client that the channel has completed before it will start a new channel.

Current casting model is that you want to send the request for a new cast/channel as the casting bar hits the red bar in quartz so that the request reaches the server at the moment the server becomes free for the next cast. Using [nochanneling] means that you won't send the request till you get the spell completed from the server which means you are adding latency up and down to your dps.

A priest using [nochanneling] with mindflay is actually destroying their dps if they are chaining mindflays. Channeled spells will be optimal using the same methods as cast time spells, hit the request for 'button up" when the bar hits the red and then spam the button a few times on the offchance you hit the request too early.

Lets say you have enough haste to bring Blizzard down to a 5.9 second channel. Blizzards cap is 28950 which means the channel should do 4907 dps with that haste. However using [nochanneling] or waiting for the animation to complete adds up+down stream latency onto the channel. If you have 150ms latency the dps drops to 4669.

Considering my current AoE plan for 2.3.2 is Blizzard (rank 1)-> Flamestrike -> IV + Blizzard (max rank) -> Flamestrike -> Blizzard that 240 dps loss is something I would prefer not to have to deal with.

My calculations do not include the 5% bonus for tier 6 since I don't have currently have it. My spreadsheet is actually at my home and I'm at my research lab so I can't plug in that bonus.

The thing I concluded from a non-4piecet6 mage is that it's not the level of +dmg that makes fireball spam better it's the ammount of spell haste you have because the fireballx2 fireblast rotation is always going to be 7.5 seconds because of the cooldown on fireblast. Reading everyones posts about fireball spam so generally made me want to calculate at what this level of gearing was that made fireball spam optimal. I figured 1430+dmg raid buffed would be a pretty good number to think that I wasn't geared poorly. *edit* I also changed my overall +dmg to 2000 and it still didn't lean things to fireball spam. *edit*

Hopefully when I get access to the spreadsheet I can plug in the extra 5% and compare it to a non-haste mage and see how these two coorelate. The difference will swing towards fireball spam obviously, and it will obviously be less than the 5% difference in dps, so you could probably conclude that even if fireblast was slightly better, the range you gain may be more beneficial overall. Also, as you add haste the fireball spam will definately pass the fireblast rotation.

Thanks for the replies!

I've noticed that quartz does not track the cast time of hasted blizzards very well. Quartz always correctly tracked AM with MSD procs, and bloodlust/troll zerking, but it seems stuck on the 8 sec cast time for blizzard. For instance when I had MSD and I cast blizzard when it procced, I got the double time blizzard, but the cast bar would stop halfway through.

BTW go spellsteal the haste buff in ZA off a flamecaster and cast blizzard or AM, it's great fun.

With spell haste the damage ticks from blizzard speed up.

Or pyroblast spam. I think it brings it to a 1.5sec cast actually, every other single target spell is s under the GCD for sure.

Arcane Missle can be spammed around the GCD, and if you have the threat reduction you'll actually be able to do some fun damage with it. Don't forget you can kite as well (with 200% increased movement speed) when you inevitably pull aggro.

Seen many discussion about this and how Arcane is dead.
I don't really believe its dead untill you have 4-Peice T6.
I am trying to show evidence, I am sure you guys have seen better but this is the best I have seen.
I recently became arcane from being fire for a long time so still trying to get used to rotation, and get used to casting AM when Clearcast procs so there should be alot of improvements.

The parse is not suppose to compare other guys just overall damage done by me.

Obviously have bunch of upgrades still comming:

Want hex trinket and Blessing deck that right there will move my + arcane damage over 1400 raid buff.
4P T5 will lower damage by about 40 points due to 7% int bonus to damage loss but will gain in the +70 damage for 6 sec every crit which should happen offen, etc.

The World of Warcraft Armory

Wow Web Stats

We just hit Mt. Hyjal last night, and winterchill killed the first time we saw him.
I kind of got owned by lag and didn't even dps the boss which sucks.

Basic question is, am i correct about my decision to stay Arcane untill i am able to build 4P T6 or am i just doing something wrong and due to that not able to perform as well as arcane. Not the perfect player but hope to get some insight on what you guys think.

Thanks in advance.

In my (generic) opinion - You are incorrect about 4pc t6 but you're not failing to generate appropriate damage w/ arcane spec. You would probably deal more damage as deep fire.

This can easily be modeled. If i have a 30% crit rate with fireball (34% with fireblast) the probability of pulling a fireball crit is 3/10 and fireblast is 34/100:

3/10*(34/100) = 0.102 = 10.2% chance this will happen. So every 10 rotations you get this to happen or once every 75 seconds. so, ignite from a fireblast being 40% of total crit, if you crit for 2750 it'd be 1100 damage lost (40% of 2750) which equates to 14.66~dps. This isn't going 3 to 5% of your total dps, which still makes fireballx2 fireblast better until a haste threshold is hit. But yes I agree it does close the gap.

So, without 4 piece tier 6, and including this, I still conclude that fireballx2/fireblast is higher dps with 0 haste gear, and X ammount of damage. The cross over to fireball is dependent on spell haste not spell damage, which could be vaulable information to someone.

*edit* I've mentioned it a few times if someone would like to see my spreadsheet who does this type of analysis shoot me a PM *edit*

In regards to keeping Fire Blast in your normal rotation, it's relatively easy to prove that the higher your base +damage is, the gap between the much larger coefficient on Fireball vs the (comparatively) poor coefficient on Fire Blast will widen substantially as your gear improves. Even without the 4pc T6 bonus, 1 Fireball will ultimately do more than double the damage of 1 Fire Blast, particularly factoring in that haste rating is wasted on Fire Blast. At that point in time, you're better off with pure Fireball spam than trying to mix Fire Blast into your rotation.

Of course this situation is probably only going to be possible with a fair amount of MH/BT gear. For those still at T4/5 levels of spell damage, it may be worthwhile to Blast every time it's up.

Jarlyn, this was the reference i was talking about, i personally have almost the best in every spot, and doing the math it's still a 3-5% dps increase.

Btw ignoring incinerate, spell haste, and ignite stacking, fire blast can be modelled as half the casting time of fireball. Simply, look at fire blast damage, and look at fireball, if fire blast does more than half of fireball it's beneficial dps. I'm still not convinced though that the ignites stacks are bugged still, i know they were in 2.0 but i haven't seen evidence since then.

Ignite stacks are bugged if and *only* if your spells go off such that your Fireball and Fire Blast hit (and obviously, crit) at the exact same time. Screenshots for proof:

Control (no fire blast):


Tests 1&2 (casting fire blast while the fireball was still in the air):




Last one, timing my blast so it hits exactly the same time the fireball does:


Did this about 2 minutes ago.

I don't know if you read my early post, but according to my math, even with 2000+dmg the gap won't be bridged to make fireblall spam neglecting haste more dps. So, the point at which it is better to eliminate it for maximizing dps is better suited to a mage with some ammount of haste gear. It has nothing to do with how much +dmg you have, because with realalistic +dmg numbers assuming my equations are right and the other mage isn't lieing, , fireball spam is a slight dps (3-5%) downgrade when you have 0 haste. This 3%-5% drops ~1% when taking into account 1/10 fireblast will crit with a fireball and you'll lose the ignite damage. This is specifically for a non 5-piece tier 6 mage as well.


The gearing to make fireball spam better is revolved around haste gear.