Athemeus

Thats the whole point of the rotation(s).
You want to start the cast with the debuff still up, because the castime is calculated at the beginning of the cast, and finish the cast with the debuff already faded, because the mana cost is calculated at the end of the cast.
So you get a 1,5 seconds Arcane Blast for the (base) mana cost of a 2,5 seconds Arcane Blast.

 

BrTarolg

Err.. what I meant was - the Arcane blast debuff lasts 8 seconds.

What you aim for is to have the low casting time so that the debuff runs out before the end of the cast.

However in this case if you cast your third arcane blast (8 second debuff) and then you cast 2 fireballs (2 seconds of debuff left) then you cast another arcane blast, which takes 1.5 seconds, you still got 0.5s of the debuff left, which means you get the high mana cost instead of the low one.
I was wondering whether you compensate this with lag, or just waiting so that the 1.5s cast ends after the debuff is over.

 

◊ manly

Actually, in raid, thanks to my craptacular machine sub-10fps and generally average lag, I never get clipping issues with 3x AB 2x fireball rotations, using 2pieces of passive spell haste. However, when I am not in raid, i get almost 100% clipping rate doing that same rotation.

I think the proper way to make that rotation work would be to stack maybe 1 extra spell haste piece or wait in 2.2 for 15.7 haste values and go for 3x AB 2x fireball, scorch.

Log on with different model:
1- Create a character of the desired model. Log on/off.
2- At character selection screen, select your actual character; mouseover the new, desired model character, and hold down left click; hit enter and release left click at the same time.
bug Arcane Potency only applies to the first Arcane Missile bolt.

 

Athemeus

Oh sorry, I missed that. I sometimes do 3AB,2FB rotations when threat is an issue and I've never had problems with the overlap. That said, I don't really stopcast very tight on this occasions, since I'm already "going slow".

 

BrTarolg

Ah I see.. thats wierd >.<

I wonder if I load my computer (3.5ghz) up on more RAM then that might let me do the scorch without spellhaste :P

 

mickeyknox

This armory link really got me thinking. Am I wrong in thinking that without EMP but with IMP fireball, crit (with spellpowr + ignite) is more valuable? I remember irontygress or someone posted a graph comparing 10/48/3, 40/21, and 33/28, and stat requirements? to make each better than the other.

Although I understand you would still need a large amount of crit, would it still be to the extent that the graph claimed? With the R.E.D. and spellpower + ignite, the benefit to crits is quite large. +lightning capacitor benefit? 2pc t5? how much is the capacitor worth anyway? compared to nexus-horn, serpent braid, crusade, sextant?

If a crit is partially resisted, does the ignite tick for the partially resisted value, or the full value?

Rotation would be 3x AB, 2x FB, 1x scorch, with this talent spec, or a similar talent spec.
http://www.wowhead.com/?talent=of0yc0czLGzbZxxMzfcs0h

I posted this, pretty much, on the WoW mage forums and I got 75+ views in a couple hours but not one post.

 

Newman

I have had some success with a 47/11/3 build with an ABx3, Fireballx2 rotation... what I did was put 2 points in improved fireball and 3 into impact before moving on in the tree. I still grabbed the AM talents (and slow) since I wanted to fall back onto a more typical ABx3 AM scorch rotation should the fireball thing not work out. Anyway, I've gotten pretty good at the /stopcasting business and it turns out that ABx3 Fireballx2 is entirely sustainable with the AB debuff falling off mid-cast at the beginning of each rotation.

I don't know if someone can comment on the theorycrafted effectiveness of this rotation... but just from a casual observation I found that ABx3, AM, Scorch served me a lot better. My guess was that I had to sacrifice too much dmg and crit to make it up to the hit cap% (required me to swap in some stuff like moroes neck, prince cape, etc), and I wasn't benefitting much from any proc gear I had. Also our mages are all arcane now so no one's stacking scorches, and of course that's a big blow.

However, I did find this build pretty useful on fights with adds... I'd throw on my usual 6% hit gear, and DPS the boss with an AM rotation, and then DPS lvl 71 or 72 adds with fireballs (I'm talking about vashj basically). Hopefully we'll get vashj this Sunday...

Just my two cents.

 

mickeyknox

Yeah, I guess I was thinking a little ways ahead, in t6 gear, where hit isnt't really an issue nearly as much as it is in t5.

Axira, do you have any WWS links or anything with Aknutal? I'm very curious.

 

Kavan

I've ran some numbers (LP optimization) in mana starved situation with JOW present and MSD as it is currently on PTR. Wanding comes ahead of AM3.

Encounter details: forced AOE to induce mana starved situation, all raid debuffs including JOW/shadow priest/shaman, lightning capacitor

Final wand characteristic: 169 dps, -123 mps
Final AM3 characteristic: 482 dps, -63 mps

 

BrTarolg

Newman did you try doing ABx3 Fireballx2 Scorch?

That (on top of giving you the scorch debuff) might improve the effectiveness of the build significantly.
I might give it a go if I get the computer to handle it one day haha

 

galzohar

3XAB-2XFB-scorch is practically impossible due to ping variance. If ping was constant or non-existant, you could easily use this rotation properly gaining a 1.5s AB with the mana cost of 2.5s. However the fact you use /stopcasting to overcome most of your ping and the fact the ping isn't fixed, forces you to play as if your ping is always the smallest possible value if you don't want to pre-click the /stopcasting macro (which is a huge loss of dps if you do for reasons I'm not going into - see the blizzard fix to /stopcasting thread for more information). This means that whenever your ping isn't that small value you lose the time equal to the difference between your current ping and the ping you're assuming. For every single person I've talked to as well as myself this time always adds up to more than 0.5s total over non-AB casts but never more than 2s obviously. This means that 3Xab-2Xfb works while 3Xab-2Xfb-scorch will not.

To make 3Xab-2Xfb-sorch work you will need an amount of spell haste to reduce the total cast time of 2 fireballs by very little. How little depends on your ping's variance (how late is your AB?). In my experience I miss it by ~0.1-0.3s so reducing fireball's cast time by 0.05-0.15s will get the job done. A little more if you want to be allowed to slightly miss the timing on your /stopcasting (by more than 0.2s which is the missing factor per spell I used to get 0.8 total time lost over the casts or missing the time when AB should start casting by 0.3s).

 

Newman

I haven't tried the 3xAB, 2xFB, 1xScorch rotation yet, but I can definitely see the benefit of pulling that one off. But I think pulling this off would be highly latency-dependent. For example, I think others talk about 3xAB, 2xFB in a raid environment, but I've tested it myself and I cannot pull this off if i have 5/5 imp fireball. The debuff doesn't wear off by the time I finish casting the first AB again. I think you'd have to be losing greater than 0.5 seconds over 3 spells, which implies either a lousy connection and/or lack of proper stopcasting. I could be spoiled, since I play at 20-40ms, but I still see a significant dps increase by stopcasting even with a low latency (I've seen 13ms). I found that I could only pull this rotation off by going 2/5 imp fireball. I imagine that you also get 6% less damage tax on those fireballs too...

Anyway I'll try out the 3ab 2fb 1sc rotation tomorrow at raid.

 

PSGarak

An interesting side effect of the whole "mage tax" business: while improved fireball does increase the dps of your fireball spam, it lowers the dpm. A build will less improved fireball *might* have better efficiency and longevity, although it is worth considering if the faster throughput would allow for enough regen time to make up for that. So: would 0/5 improved fireball actually be a better build, in terms of more damage per mana bar even though it's slower in doing it?

 

Kavan

Patch is coming close and I had to decide what to do in regards to the meta gem. As far as I can tell we didn't reach any conclusion after they changed channeled spells to consume focus buff.

I've tried to model the MSD mechanic as well as I could (I probably missed some subtle details with dynamic rotations) and for deep arcane spec both gems now seem to be about even (including JoW). There are some changes in optimal rotations however. For MSD AM more or less replaces AB spam and at most ABx2 rotation used for the second cycle. Initially I overestimated MSD effect on AM, which made me think MSD would win (I used average MSD haste instead of average between focused and non-focused cast time, difference 26% haste compared to 15% haste). Main problem is that MSD has almost no effect on AB rotations. Since all AB casts are < 3 sec most it can do is reduce it to 1.5 sec (for ABx3+AM+Sc this is average 13 sec compared to 12.7 with MSD).

In conclusion, given that JoW uptime is usually unreliable and the problems with focus on AB rotations (like focus procs on last AM wave, no time to change rotation, which means Scorch eats it and in most cases results in spell not ready because of /stopcasting) I've decided to remain with RED. Once I get to the point where I'll be thinking about dropping 2/5 T5 bonus for T6 I'll probably come back and reevaluate, unless they give us an alternative caster meta gem in the meanwhile.

 

galzohar

Read my post. It's not about your ping, it's about your ping's variance. Of course the changes in your ping can never be greater than your max ping, which is why you're pulling it off with somwhere between 20-40ms ping differences. It's not poorly executed /stopcasting, it's impossible to execute it well enough with a certain level of varying ping. Go to a computer with 400ms ping and try it yourself (but make sure the ping varies enough too because if it's just a fixed 400ms ping you'll pull it off - however afaik this is never the case).
I think someone already mentioned that using a G15 keyboard he couldn't lose less than 70ms between casts macroing his scorches to cast every 1.57s - trying to do it any faster would cause some of the casts to be clicked too early - probably due to ping changing between casts. If ping was fixed, you could just macro it to cast every 1.5s.

According to my spreadsheet, if you have just the minimum time lost between casts to screw up the 3XAB-2Xfireball-scorch rotation then you will actually have to wait very little before casting the AB after casting the fireball when you do a 3XAB-2Xfireball rotation but it would still be a very effective spec if you compare it to any other spec given same amount of time lost between casts. And if you lose just a little more time between casts than that it gets even "better" as you don't have to wait at all before the AB and just cast as fast as you can.

Note that I've been getting slightly more theorycrafted dps as 40/18/3 than 33/28 by using those rotations (with 2/5 t5 obviously otherwise I don't see why you're thinking about using AB at all), and if you spam some AB 40/18/3 gets even better.

I like the idea of getting a little haste to make adding scorch possible, but wonder if it would really up your dps compared to using equivalent non-haste gear. Calculating a comparison shouldn't be that difficult if you're actually in a position to make such choises.

 

BrTarolg

Assuming you're not going to even bother scorching, - i.e the time you lose is greater than 0.5s but not more than 1s, then yes not going imp fireball will significantly (well ok 10% coeff) increase the damage of your fireball.

Edit: ok this is slightly OT but i was playing around with quartz and ive realised for some reason quartz sometimes shows a huge lag (like for me, that would be 300ms+) which if i stopcast even well into the stopcasting area, will call an interrupt. Thus ive trained myself to stopcast much later for those casts (which is annoying). Has anyone else had this?
Also, does anyone know how the "embed" function for quartz works?

 

Xei

I wouldn't be aiming for 1100+ DPS in T4/T5 gear. I have close to the best set of T5 gear you can get without Vashj/Kael loot and I can just hit 1100 DPS on some fights.

I know WWS calculates DPS differently so its a bit misleading, but on our last Leo kill I clocked 1110 DPS on Recount (not sure how it calculates DPS) with a SPriest and Resto Shaman - flasked, oil, food and chain-chug Super Mana's (43/18/0).

"Being a leader is not a position of power. It is a position of service." ~ Barestomper

WTB Oceanic Horde PvE Players

 

mickeyknox

When you say that 40/18/3 is showing better than 33/28, are you assuming someone else is stacking scorch? Is the 6% fire crit, 1% damage, and 6% fire damage not worth more than the 100ish spell damage, 1% crit (arcane potency) and 3% intel?

 

BrTarolg

Don't forget the 3% hit also from elem precision. Also I think the build looks like this
Talent Calculator - World of Warcraft
(Assuming someone else scorches) so you also get the benefit of MoE and full points in arcane potency.

Though i really have no clue about the validity of one over the other, except that AB is usable with full MM.

 

◊ manly

There are soo many bad assumptions in the previous posts I didn't feel like posting in the past days, but this is getting pretty absurd now.

1- 40/18/3 is Talent Calculator - World of Warcraft . There is mostly no point in master of elements in a deep arcane build because you already have your mana regen/extra int from your arcane tree, and besides, your fire crit rate is going to be low anyway, in addition to have you not cast a lot of fire spells. In other words, its arguably 3 wasted talent points.

2- 40/18/3 has several merits in its own rights as long as you have top-end gear (ie: t6-level gear) which have lots of extraneous hit ratings. The build is what I believe is best suited for end-game raiding, assuming a non-perfect top-of-line gear. The build is good because it has every AB-spam DPSing talents, as well as extremely solid fire damage (ie: fireball 245% crits). 43/18/0 and 48/13/0 might have the same 245% fireball damage, but the builds don't use fireball as part of their rotation. Throwing some wild numbers in the air, I would say that if you play with an elemental shaman, and have approx 8-9% hit rating from your gear (ie: with no hit pieces swapped in, just your regular 'best gear'), then that's probably hands down the best build you can take.

Basically what many many players seem to forget is that under a real boss scenario, you spend at least 20% of your time in your high DPS rotation (ie: 1.5s AB spam). You can check any of the parses I linked, I rarely ever even do my 3x AB 2x fireball rotation, or 3x AB, AM, scorch, because I do almost 40-50% of the fight doing 1.5s AB.

3- 'someone else doing scorches'. I'm sorry, but that will have such a minimal impact on DPS you probably won't even notice. If you're trying to explain why you can't seem to get above 1100 DPS with t5 gear and that I am telling you you could push 1300, and you don't understand how, well, let me tell you. The fights in hyjal/BT are typically much shorter than SSC/TK (and/or less mana intensive - as-in no aoe). This means the relative worth of bloodlust is bigger, in the sense that one bloodlust will last you for 30% of the fight, if not more. The only way you can get really high numbers is by having almost mandatorily an elemental shaman / shadow priest, and a bloodlust.

Here's what I would recommend to anyone that asks:
SSC/TK - 10/48/3 or (if 2pc t5) Talent Calculator - World of Warcraft (3x AB, AM, scorch) (crusade/TLC)
if you have 8-9% hit from gear, raid with elemental shaman/shadow priest, you might consider seriously Talent Calculator - World of Warcraft (3x AB, 2x fireball or 3x AB, 2x scorch, fireball)
if you have the top-notch best-of-best gear, then you might seriously consider Talent Calculator - World of Warcraft

Log on with different model:
1- Create a character of the desired model. Log on/off.
2- At character selection screen, select your actual character; mouseover the new, desired model character, and hold down left click; hit enter and release left click at the same time.
bug Arcane Potency only applies to the first Arcane Missile bolt.

 

mickeyknox

Question for you, Manly: You have 9.2% hit rating from gear, Does the 5 points in arcane focus do ANYTHING once you pass the 16%? Why not get imp mana shield or wand spec (yes, i know.. its horrible) but, if it does nothing at all why not buff something that will do a tiny bit of something?

 

Unleash

There are many problems with this spec and why it is not ideal for raiding. Its not horrible by any means, but easily is outperformed by other specs.

The spec replaces AM in your rotation for fb/scorch. Is this bad? Depends, it seems to be every mage is getting a shadow priest in raids anymore if at all possible. Elemental shaman is probably not as common to count on for every fight you encounter. So first off, your gonna have to stack a bit more hit rating then normal(without a shaman), losing +crit on your ab spamming. You will also lose the ability to use AM, which on occasion proccing focus, clearcast x2, and another focus results in a nice 2000+dps for 15-17 seconds. I can go into a few AM proc fests in fights as short as akama.

Mana purposes this spec would benefit someone a bit better, but our gears and having shadow priests, using mana pots, gems, jow, etc... I can spam ABs nearly an entire fight. The times that I stop spamming ABs are when my clearcast/focus procs are going crazy and ill spam AM till my procs quit coming.

Illidan p1 = ab spam entire time unless am proccing (evocate here)
Illidan p2 = rotation abx3 am (i actually wait a second and refresh from here) - plus can get a bit of AB spamming done
illidan p3 = ab spam entire phase (evocate again here)

That is just an example how I dont see going into changing up my rotations into Fireballs/scorchs. The mana would be the only reason I could see to do this, but yet if thought out your mana shouldnt be a problem. I know not all fights are long enough to evocatex2 in, or have breaks inbetween like illidan, but generally every fight you can come up with a plan just like you can on illidan to max your time ab spamming, and have very little time in a rotation.

 

Roywyn

I've been using a 40/18/3 build for a while and really like so far. I switched to it when we were at Kael'thas, after having played with AB rotations for two weeks to get used to it.
I wanted to keep the versatility of fire - a strong 41 yard nuke with 70% non-interruption with a good mana-efficiency - and didn't have a TLC. Also, we always have mages who scorch in raids, so fire vulnerability was always up as well.

Even at T5 level, you will get quite a lot of spell hit on your best items for the slot.
Head Enchant, Ruby Drape of the Mysticant, Vestments of the Sea-Witch, Belt of Blasting, Spellstrike/T5 Leggings, Boots of Blasting, S2 Arena Dagger, Arcatraz Off-Hand, Tirisfal Wand of Ascendancy/Wand of the Forgotten Star.

Chances are that you will wear quite a few of those items. Not all of them are best, not all of them will always drop, but it's hard to avoid spell hit completely.

I plugged in my gear (with ~80 hit rating that are hard to avoid) into Vontre's spreadsheet, modified it a bit so that for every cycle it calculates it uses up all excess mana by using the rotation with one more AB (see his thread) for some part of the time, enabled Judgement of Wisdom, disabled Totem of Wrath, and used Ashtongue/TLC trinket.
I made up a spec with 3 frost (precision), 18 fire (with master of elements) and all arcane talents.

Even in that scenario, 3AB/2FB was a tiny bit more overall damage than 3AB/AM+SC.
Yes, AM is better DPS than FB for that spec, but it's less mana efficient, allowing for less additional AB usage.
I see that if your connection/system is too good and you stopcast too well, you have to be careful with your AB debuff running out. I recently got the spell haste ring from BT which doesn't make that any easier.


I'd definately recommend some 40/18/3 build/variant for people who were fire before and want to try arcane blast specs. You still have a pretty strong fireball (41 yards and 70% uninterruptible is nice to have in mobility/random AoE fights), your hit gear doesn't become useless, and you don't have to worry about getting a TLC if you don't have one.

I don't say that a 48/13 build is worse, but I feel that a 3AB/2FB isn't given enough credit.

 

Unequaled

that kills me.. our locks do 1200 dps without even having T5 gear, etc. My gear is slowly increasing but it seems that I have to agree, 1100dps sustained with craftable/t4/t5 is going to be a bit harder then I though.. especially after seeing locks break that number easily.

 

Unequaled

This is the first time I am seeing this spec written up as being 'competitive'. It's a spec that in theory sounds very interesting to me, I was wondering, what 'gear' do you need for this to be competitive? I would assume, at a min, 2t5. and as you stated, 8-9% +hit. I raid almost non stop with a spriest, but a ele shammy i rarely have. The spec sounds interesting for me to try once I get my 2t5