Thanks for the numbers Vontre.
May I ask you for another frost calculation with 75% WE survivability, a 2nd frost mage and the usage of iceblock 2 times instead of running away (may be moddeld as no cast time loss and 1,5 gcd loss for IB, or simply 2 more frostbolts total than in your calc).

Yeah, you're right Rounced. I'm doing Zul'aman now, I'll do a few more sessions on Dr. Boom after with 10 48 3. Hoping it won't fail.

Also, there were no debuffs or boom <20% when doing fire.

First test with 1193 fire dmg 34% crit 65 hit and AToI

Vhadnow - WWS

I've no idea why it's reporting so low dps, I messed up reapplying scorch once like 2 fireballs before it should be but it didnt drop off or anything.

Test 2, changed AToI with Sextant.

Vhadnow - WWS

Test 3, same as test 2

Vhadnow - WWS

Test 4 with 10 0 51 frost spec, 1185 frost dmg 26% crit Sextant and Icon.

Vhadnow - WWS

Something must be wrong with my parsing/uploading or something...

Test 5 frost

Wow Web Stats

Went back and tried full Arcane, got much closer numbers to the previous specs. 1299 Arcane dmg Mage Armor 25% crit and only haste is the Trash one.

Vhadnow - WWS

Tried my haste gear and Molten Armor, changes it to 1238 spell dmg and ,35 seconds off AM with shoulders/belt/offhand/ring as haste gear.

Vhadnow - WWS

I must be doing something wrong with the log or something... I'll do some more tests tomorrow, getting too tired.

I'll be frapsing something next week to document the best time i ever had in raiding, this age of am spam patch2.2-2.3. I think i might have some of the best gear for it, so i hope i get a zillion focus procs out of spite and to entertain : /

With the time-honored ABx3, AM, Scorch rotation more in vogue than AM spam, combined with this so-called change that will negate Stopcasting, I'm curious. Let's assume exactly 0 delay infused by latency, because the almighty Blizz have solved that problem. What is the exact amount of haste rating we can wear, and make the cycle clear the next AB with the buff 2-stacked but NOT finish casting while it's still up? Ideally I'd like a 0.1 or at least a 0.075sec margin between "AB buff runs out" and "you gain AB buff" to make sure. Slightly complex math, I'll come back with my results and see if it correlates with some of the greater thinkers.

Edit: I suddenly dawn upon the realisation that 2xAB,1xAM won't in fact cause next AB to land on unbuffed without test-realm latency and thus adjust 2AB 1AM to 3AB,AM,Sc instead.

Edit2: End casting AB: time 0, AM duration, t=5sec, add scorch, t=7.5, add 3xbuff AB (1.505) t=8.005. Does this mean that if we have exactly 0 latency we fall short of casting the first AB before the buff is over by 5 thousands of a second? o_O Or am I calculating the cast time reduction of AB totally wrong? I was hoping to make a mistake somewhere in the "haste only changes the base AB time, thus is not additive with it's own buff, but multiplicative" kind of mistake, but if I'm correct with 1 haste it'll be buh-bye rotation...

Very good simulations Vontre.

But did you account for the TLC and Ashtongue in those simulations?

Because as we know, they would add alot of extra dmg to AM spam.


I assume you used calculations based on the optimal fire gear vs the optimal arcane gear (Haste gear included)?

Actually those numbers are pretty similar to what I would expect if you weren't maximising the new casting system.

Does Quartz work anymore?

If it does, without using any macros, start spamming the casting button *before* you get to the red and see if that doesn't improve your dps a bit.

Also mind running another AM spam parse now that the test server has settled down again so we could have a little more of a baseline for comparison?

and you had 1700 dps for AM spam and only 1200 for Deep Fire?

Something must've been wrong with the initial reports and one of my latest AM reports too, since I got something stupid like 1600 dps with AM on one try. I think it's because I fail with WWS.

Edit: Looking at them again I can't see what is wrong with the 1700 dps report, the first 16xx dps one is very short in duration so something is up. I'm confused, going to bed it might make sense tomorrow.

Sadly, Confirmed.

dont be sad it works with evo now

Yes thats right.
Well I really don't know, but this small fact makes me think about it... as far as I heard the developement of 2.3 started when 2.2 was still in developement as well. Might it be, they just forgot to implement the change of 6sec -> 10sec into 2.3 because it was never in? Because of the Cooldown, people using MSD still could really need the 10sec to make up their minds which spell to use the focus on (lets say get to a safer spot and cast evo), there is no need nerfing this as well.
And that makes me hope, that maybe the cooldown issn't 45sec on live servers...

Ah c'mon, this is a PTR, there is still approximately 0.01% chance that this change will not go through. I plan on keeping my spirit evocation staff/wand even if they won't be needed anymore, just in case it gets reverted back at some point in time. Plus, TLC is good for AOE.

I'll certainly keep my spirit weapons, they'll be useful whenever I'm not casting nor evocating... Okay, it doesn't happen that often, but pressing two hotkeys to switch weapons isn't a lot of effort for a few dozen mana during a transition.

I'd thought id chuck a random few "thoughts" into this.

Firstly, people have been getting unusually high AM DPS numbers. Well, try not to forget that you dont have any ramp up time, and also about your mana bar also. Chances are you will get a focus proc straight away, and then you will get your next focus proc 45 seconds later. Given you arnt bothering to use supermana pots/evocation on boom, an AM mage might run OOM just after the second focus proc if youre using molten armor, and burned pom-pyro at the start. Don't take my word for this, im just throwing out random ideas but that might artificially increase your "focus uptime" more than it should be
Secondly, you dont take into account molten fury, nor do you get to take into account molten fury + combustion synergy. Also you get a vast decreased DPS due to the fireball dot after you run OOM and stop DPSing.

Ok who knows - I could just be completely wrong ofc :P. But I do think it does show that AM spam wasnt actually that bad pre 2.2 anyway, it just wasnt optimal (super high mana usage, and slightly lower DPS than fire).

----

Anyways, even with this new change, I dont think its time to completely take arcane out of the picture.

Main point is, i'm sure you have seen several times that "with a good shadowpriest you never run out of mana" - and this will be even moreso for fire. Next thing is that earlier manly said in the thread "with a 40/21/0 you should actually beat fire due to the insane crit scaling" (note the new metagam aswell which might increase crit scaling even more). Maybe that was an overexxageration, but-
With the new "improved stopcast" that blizzard apparently gives us, we can assume zero latency. This can be important, because it opens up a whole new wad of rotations which havnt been possible before.
Things such as 40/21/0, ABx3 Fireballx2 scorch/fireblast rotations, assumping that they fixed the "AB rampup bug" - this is because previously nearly everyone didn't have the latency to do this, but now hopefully they do.
This spec also gives you the big advantage of having a mana dump (thats also assuming flat out AB spam = higher DPS than fireball spamming or the above rotation) - so just keep chugging the super mana pots and going all out with AB - or potentially replacing scorch with fireblast in the above rotation (does fireblast have higher DPS than scorch/fireball?)

Also if fireblast does indeed do a higher DPSC than fireball (it does for me at any rate, but my gear is pretty low to compare) then you can always do rotations such as fireballx2 fireblastx1 - seeing as we still know that fire never, ever runs out of mana with a shadowpriest. (fireblast also gives the options of doing some of your DPS whilst moving around)

Is it possible if anyone could simulate/test the above situations to see how the DPS for those comes out?

I'm not willing to give up on AM spamming QUITE yet. It just may be taken in a slightly different direction. Using mage armor, mana pots and fishing for clearcasting procs.

It's not looking good. But I'm not quite ready to nail this coffin shut yet.

My biggest gripe is this:
If you check the ZA loot, and start thinking forward, you get roughly those numbers:
12.6 hit for 1% more dps
15.7 haste for 1% more dps
22.1 crit for 1% more dps

If you think about it, it means that the cheapest route is to start stacking hit > haste > crit from your gear, ignoring here the limitations of the used rotation. That means if you can easily fill the hit cap (which can be quite trivial depending on spec and ZA / hyjal / BT gear). Your next best option will be stacking haste. The way I view it is that basically 'haste is the new hit' in the sense that its the cheapest route to increase your DPS once you've reached the hit cap.

This means in all likelyness, projecting forward, I do have plans so have my gear follow as close as possible the order above. This means I can expect quite a number of haste ratings from gear - much more than I currently have. And the ZA loot does suggest that haste ratings will be everywhere. What I believe is that we will probably end up with hit cap then gear mostly oriented towards haste, then leftover points tossed on crit. This implies spell rotations that try and avoid falling below GCD, as it would only imply totally unused itemisation points. As such, I always smirk - even now despise my 'low' amount of spell haste gear - whenever I see a rotation involving 1.5s ABs or scorches or fireblasts.

I am not discrediting 40/18/3, I always believed that that spec has the best scaling out of every AB rotation spec there it. But it sure as hell is a bitch to maintain (ie: reliance on COS, COE and scorch). But now that we can see lowered latency (which BTW I estimate to be roughly 20-50ms with the stopcasting fix), that would indeed points us towards a 3x AB 2x fireball, scorch rotation. Now what irks me is both of theses 2 things: 2pc t5 and the first AB as well as the last scorch being 'unable' to use the haste ratings.

I understand that in a nutshell 2pc t5 would be a greater DPS boost than 4pc t6 as far as that AB rotation is concerned, but it does still irk me about that one scorch. Even using a max spellhaste gearset (yes, even including the haste neck from ZA and the spellhaste pants), we can't do 3x AB 3x fireballs rotations.

Also, consider that 40/18/3 is a spec that would excel *if* your crit rate is really high, due to arcane spec fire crits (245%). The more we go towards haste, because it is cheaper itemwise, the less that spec will make sense as a whole. You could say that the spec just requires different gearing to truly be capitalised upon, and I could also play the devil's advocate and say that the spec as a whole makes a bad use of the ilvl formula and crit costs vs haste costs. I guess its all a matter of perspective.

I also agree with Manly to a certain extent. The thing with haste is that it subtracts cast time based on a percentage of the base spell cast speed. This of course means that AM will benefit more from haste than would Frostbolt or Firebolt. As time goes on I am very interested to see how much base/passive haste you can stack up coupled with Proc on Hit gear to see how AM will do.

I am predicting that we will see passive haste comming close to 300 with Proc on Hit haste to come very close to 500. IE:Ashtongue. If this is true then 3.5 second AM spam (~30% cast speed reduction) would look pretty sweet compared to 2.5 second Fireball spam (using the same ~30% cast speed reduction).

Long live AM spam

I havent seen any concrete dps parses yet for the new MSD with cooldown, but if we assume the worst and that it's completely useless (though I'm sure it will be competitive), has anyone tried/modelled/considered the option of using the Insightful Earthstorm meta with AM spam?

I get the impression that the mechanics changes to AM will allow that 2% mana proc to proc off each tick of the missiles, and possibly a sixth tick for the actual cast (based on what others have reported). Is this the case? I recall someone saying they used this meta for a more standard AB rotation Arcane build (as at least arcane spec gets damage out of intellect to some degree), but the MSD was so ridiculously good when they changed the calculation mechanics of AM to proc stuff off every missile that I don't think anyone bothered running calcs for any other meta!

The mage I regularly team with for heroics/kara etc (where optimal groups means she rarely gets a shadowpriest) uses Rank 8 AM spam with Rank 10 on clearcasting, with the MSD, and her damage has been phenomenal. With the nerf to the MSD, might it be interesting to swap to the Insightful Earthstorm and go for Rank 10 spam with a focus on passive spellhaste from the upcoming ZA loot?

I'm not sure AM is dead yet - both using the same gear except for one trinket (Icon swapped for TLC for Deep Fire) and both with Molten armor and I was showing Deep Fire with an average of 1113.3 over 3 mana bars and Deep Arcane with an average of 1122.5 over 6 bars.

Seems like Arcane damage is very dependent on clearcast procs - you get them and its competitive, don't get them and the damage is pretty pathetic.

To try and keep everything even I didn't use AP at all for the Arcane Runs and I didn't spec the Molten Fury for the Fire runs.

Mage armor definitely gives arcane some serious staying power with the changes to Arc Med - easily equivalent to 10/48/3 but the dps definitely takes a serious hit using it - although if you get the clearcasting procs Deep Arcane even in Mage armor is very competitive.

Not sure how any of that will translate over to real raiding but it does seem to imply that the nail is not quite in AM spam's coffin just yet

Manly, mind doing a similar test of 4T6 with both specs?

No molten for fire and don't use AP for arcane. Would really love to see how the dps numbers work out for other mages using similar testing conditions.

I'm on PvP PTR atm by Dr. Boom and at least there is no ganking going on atm - only issue is Dr. Boom dies very fast.

Some of our mages are fans of arcane spec and probably wont respec even if deep fire will be superior dps. So i have question: with MSD nerf, will it still be best meta for deep arcane, or new CSD will be better? And can someone test on ptr how exactly 3% critical damage increase of CSD works with crit multipliers.

I'll be saving mine for Innervates.

Does it now? On something like a focus proc you would be correct - casting AM you save 2.5 seconds, whereas casting fireball only saves you 1.5 second

However with passive haste assumping youre just constantly casting then it will benefit fireball and frostbolt and AM equally (unless god forbid you somehow get over 100% haste even with trinkets and the like and start hitting the GCD cap)

Evidence shows (above stuff) that basically AM spam *does* push out decent DPS, but it simply isnt as good as fireball. Unless somehow you can stack ashtongue + TLC + more "on hit" stuff, and have enough mana to do it on boot, its unlikely AM will ever catch up with fireball.
It also shows that AM was actually pretty decent pre 2.2, its just that nobody bothered to use it because fireball did slightly more without the nuisance of mana problems.

---

This might be slightly offtopic, but i would like to ask if someone can test passive haste + focus with evocation

Do you lose a tick of evocation if you have passive haste?
If you focus + evoc, do you lose 2 ticks, or does it tick twice as fast?