Cataclysm Arcane Mage Compendium - Page 10

Zersto · 07/14/11, 12:56 AM

I realize that Arcane's cooldown rotation is dynamic, changing with fights, but I'm having some trouble determining the best use of Arcane Power in relation to mana. Is it typically better to sit on the AP cooldown a little longer and use a full burn rotation under AP, or should I be trying to use on cooldown even if it means using a conserve rotation under AP?

What I am having trouble determining is: if using it on cooldown and using conserve rotations means getting another AP use during the fight (over sitting on the cooldown), does it outweigh using it during burn rotations only?

dar3652 · 07/14/11, 1:59 AM

Originally Posted by Snickernew

Please realize that with 4t12 it might be more beneficial to use AP on cd (every 90 seconds), so the net mp5 gain could be slightly higher than you have calculated. Again this is only a speculation on the existing strategy of using AP + mana gem every 2 minutes. Using AP every 90 seconds will not allow us to do a full burn cycle though as evocation will not be available for at least 30 seconds after the AP is finished. Nevertheless, I would be interested in the results of this strategy with the new hypothetical 4t12 bonus.

Also, the haste vs mastery discussion boils down to whether you have SoW:
Haste > mastery with FoW
Mastery > Haste without FoW (you still need at least 1200 haste rating, as only past a certain point depending on your gear will your mastery give more dps per point than haste).

what exactly happens @ 1200 haste? I've noticed that at around 1100 or 1150 haste, mastery starts to get better than haste in simcraft, but was wondering if this number had a reasoning (such as some haste thing i'm missing as I thought AB becomes 1 sec under lust @ 1768haste)

Loevty · 07/14/11, 2:23 AM

Originally Posted by Zersto

I realize that Arcane's cooldown rotation is dynamic, changing with fights, but I'm having some trouble determining the best use of Arcane Power in relation to mana. Is it typically better to sit on the AP cooldown a little longer and use a full burn rotation under AP, or should I be trying to use on cooldown even if it means using a conserve rotation under AP?

What I am having trouble determining is: if using it on cooldown and using conserve rotations means getting another AP use during the fight (over sitting on the cooldown), does it outweigh using it during burn rotations only?

Unless I am missing something, it could be said that the vast majority of Arcane's burst and overall damage is from the 'burn' phase, and unless there is a a resource retruning mechanic(Sinestra, Alysrazor etc)I don't see where the activation of Arcane Power would be beneficial on cooldown instead of the usually Arcane Power + Improved Mana Gem + (On use effects) on the two minute cooldown.

If the report's of the AP actually working as implied by the T12 4P set bonus, the above would change however, I still haven't seen any evidence of the T12 4P interacting with Arcane Power as this post suggests.

Silverwind · 07/14/11, 2:51 PM

Here's an updated Focus Magic chart, because some people had requested it.

It was done with SimC using 10k Iterations. Accuracy on the DPS values isn't 100% because of the way SimC works, there can be an error of about +-50 dps on the Target DPS gain.

Profile	DPS w/o FM	DPS with FM	Target DPS gain	Feedback uptime	RDPS gain
Mage_Fire_T12H	31703	32826	1123	98%	1906
Warlock_Destruction_T12H	33832	34686	854	100%	1653
Warlock_Demonology_T12H	35722	36617	895	87%	1590
Mage_Arcane_T12H	36003	36802	799	98%	1582
Druid_Balance_T12H	32365	33185	820	91%	1547
Shaman_Elemental_T12H	33031	33661	630	100%	1429
Warlock_Affliction_T12H	36084	36840	756	80%	1395
Mage_Frost_T12H	37117	37695	578	100%	1377
Priest_Shadow_T12H	33846	34759	913	45%	1272
Rogue_Assassination_T12H	34557	34907	350	93%	1093
Shaman_Enhancement_T12H	35167	35427	260	95%	1043
Priest_Disc_Smite_T12H	14649	14862	213	99%	1004
Rogue_Combat_T12H	34867	35194	327	84%	998
Rogue_Subtlety_T12H	35474	35576	102	77%	717
Death_Knight_Unholy_2h_T12H	35348	35613	265	52%	680
Death_Knight_Unholy_1h_T12H	30458	30731	273	50%	673
Death_Knight_Frost_1h_T12H	38100	38171	71	33%	334
Paladin_Retribution_T12H	33443	33479	36	36%	323
Death_Knight_Frost_2h_T12H	33753	33802	49	25%	249
Hunter_MM_T12H_Arcane	34632	34672	40	0%	40
Hunter_SV_T12H	32263	32302	39	0%	39

Edit: Updated with 10k Iterations using smooth_rng and deterministic_roll options, seems to be more accurate now.
Edit2: Added uptime percentage and RDPS columns. Rogues showing 0% uptime seems to be a SimC bug.
Edit3: A bug in SimC with procs not triggering FM was fixed. I updated the chart with new uptime values. Rogues are now included in the uptime column, Shadow Priests went from 36% to 55% uptime because of Improved Devouring Plague and Shadowy Apparitions.
Edit4: Shadow Priests Apparitions fixed, new uptime of Shadows is 45%.

fateswarm · 07/14/11, 3:22 PM

^ It would be interesting to see the HPS benefit when casted on a healer, now that they got a buf on their crit mechanic for cases a group might need higher HPS rather than DPS.

sierran · 07/14/11, 5:05 PM

Originally Posted by Ataxus

This is very good news! I'm glad they changed it.

If the 4pc Arcane bonus is now working as expected, does this fix the gear progression and now make T12 an upgrade over T11.5? It seems like the numbers might still be too close to be an upgrade, and I'm seeing confusing numbers over how much mana T12 will actually save.

But for sure, now 4pc T12.5 will be BiS.

Ataxus · 07/14/11, 6:12 PM

Updated AB mana costs from wyrmbane's post:

Without Shard of Woe

AB Stacks	Base Cost	Cost w/ AP	Cost w/ AP and 4pc	4pc Savings
0	870	957	783	174
1	2175	2262	1958	304
2	3480	3567	3132	435
3	4785	4872	4307	565
4	6090	6177	5481	696

With Shard of Woe

AB Stacks	Base Cost	Cost w/ AP	Cost w/ AP and 4pc	4pc Savings
0	665	731	599	132
1	1662	1729	1496	233
2	2660	2726	2394	332
3	3657	3724	3291	433
4	4655	4721	4190	531

Snickernew · 07/14/11, 9:47 PM

The 1200 value of haste was referring to my personal haste rating where mastery becomes more valuable than haste for dps. Also, I was merely pointing out that with SoW haste is much more valuable than without it, but the actual amount of haste needed to get to the break point b/w haste and mastery is different for every mage (but should be close to 1200).

If someone would be able to simulate the 4t12 benefit for a 6 min fight and post here - that would be great.

Psychedelics · 07/15/11, 2:46 AM

Originally Posted by Snickernew

The 1200 value of haste was referring to my personal haste rating where mastery becomes more valuable than haste for dps. Also, I was merely pointing out that with SoW haste is much more valuable than without it, but the actual amount of haste needed to get to the break point b/w haste and mastery is different for every mage (but should be close to 1200).

If someone would be able to simulate the 4t12 benefit for a 6 min fight and post here - that would be great.

Now, that magic 1200 haste rating number you speak of... is that for magi with T11 4-piece?

I would imagine that the "magic number" for a mage without the T11 4-piece bonus would be much different, if so.

Morthoul · 07/15/11, 11:12 AM

Originally Posted by Silverwind

Profile	dps w/o FM	dps with FM	diff
Priest_Shadow_T12H	33846	34759	913

It's worth noting that Mind Flay counts as a dot and its crits won't proc FM back on the mage, so although Shadow Priests benefit well from FM, the total benefit to raid DPS may be less. Back in WotLK there was math to the effect that the recipient of FM doesn't matter to the mage, because there's near-100% uptime regardless, so we tend to ignore it. But shadow priests at least are a special case. And whatever the uptime is, another issue is that the Shadow Priest "opener" waits a while before casting even one Mind Blast, so you can go through the initial Arcane burn without any FM proc at all.

Silverwind · 07/15/11, 4:05 PM

Originally Posted by Morthoul

It's worth noting that Mind Flay counts as a dot and its crits won't proc FM back on the mage, so although Shadow Priests benefit well from FM, the total benefit to raid DPS may be less.

Good point. I have added RDPS and uptime colums to my original post. Seems like Shadow Priests are only providing 36% uptime for the Mage, which pushes them far down in the RDPS the list. There is one flaw in the chart which I'm aware of: Rogues can proc FM as far as I know, but SimC doesn't trigger the feedback buff for them. I will have to research this further. I'm also not sure if hunters can proc it.

The conclusion seems to be, that if there is no Fire Mage or Destruction Lock in the raid (and assuming everyone is on the same gear level), Arcane Mages may as well circlejerk the FM.

Pyryte · 07/17/11, 11:14 AM

Do we have any more data on live to verify how the T12 4pc is actually calculating the mana cost?

fok · 07/18/11, 6:17 AM

Don't forget simcraft selects gear from whole content, and gloves from HEROIC Ragnaros seems like...
You will full geared at this time, but without really good gloves, so, how about makes t12 with gloves and without, for example, head? Heroic helm is from Ragnaros too, BUT, good 391 head can drop from Rhyolit, so you will be with 391 head AND 391 gloves, which can be obtained from Majordomo who is more-more-more easier than Ragnaros.

AND THEN, After killing Ragnaros you always can take gloves from his chest.

Upd: Also, great robe from Beth'tilac - haste/mastery, very easy to obtain, and great stats

Malcophant · 07/18/11, 1:05 PM

Originally Posted by fok

Don't forget simcraft selects gear from whole content, and gloves from HEROIC Ragnaros seems like...
You will full geared at this time, but without really good gloves, so, how about makes t12 with gloves and without, for example, head? Heroic helm is from Ragnaros too, BUT, good 391 head can drop from Rhyolit, so you will be with 391 head AND 391 gloves, which can be obtained from Majordomo who is more-more-more easier than Ragnaros.

AND THEN, After killing Ragnaros you always can take gloves from his chest.

Upd: Also, great robe from Beth'tilac - haste/mastery, very easy to obtain, and great stats

Keep in mind that the Rhyolith helm is a random enchant helm, similar to Conclave/Al'akir's drops from last tier, so there is no guarantee that you will get a good enchant.

Also do note that as of yet we have no proof that the 4pc has been changed. I've posted on the official forums seeing if anyone has the 4pc and can post screenshots, but until then I'd continue under the assumption that it hasn't changed.

dar3652 · 07/18/11, 1:57 PM

For confirmation / screenshots you can try getting a hold of the mage in Raiding Robots referred to on the last page with the 4 piece.

Dracones @ Shattered Hand - Game - World of Warcraft

KroLeXz · 07/18/11, 6:29 PM

maybe u can just extract information from here

World of Logs - Real Time Raid Analysis

Looks like it's confirmed - look at that buff he gains when popping arcane power.

Incognieto · 07/18/11, 9:11 PM

Originally Posted by KroLeXz

maybe u can just extract information from here

World of Logs - Real Time Raid Analysis

Looks like it's confirmed - look at that buff he gains when popping arcane power.

The only buff I can see is Arcane Power which shows the modifier [less][more], however this is showing for all mages regardless of T12. I cannot see any other buffs referring to Arcane Power and if there were, its uptime will match that of AP.

I'm not sure much info can be extracted from the logs but rather confirmation is required from someone with the 4p.

As far as I know WoL can't track mana cost.

tlosenjin · 07/19/11, 3:17 PM

Until simcraft is updated to reflect the 4pt12 bonus changes, I thought I'd add my own analysis of the bonus' damage value. There are basically two ways the bonus increases our damage, the first being its synergy with Arcane's mastery, and the second being the added "free" AB4 casts we have added to our burn phases.

Damage Gain from Synergy with Mana Adept

← Click Here

Damage Gain from "Free" AB4 Casts

← Click Here

The second part of the bonus is the AB4s that you can cast with the saved mana. The time spent casting these would otherwise be spent in a conserve phase, so the damage gain of the bonus is the damage of these "free" AB spells minus the damage that would have been done using a conserve rotation. This part of the bonus is pretty difficult to model without using approximations, so I'll outline the simplifications I use along the way.

First, the number of "free" AB casts can be calculated as the quotient of the difference of the mana left over after the AP burn phase with and without the 4pbonus and the mana cost of AB4 without AP activated:

$\frac{{M - m \cdot c + \gamma \cdot t(c - 1) - (M - m \cdot c \cdot b + \gamma \cdot t(c - 1))}}{{{\rm{AB4 Mana Cost}}}} = \frac{{m \cdot c(b - 1)}}{{{\rm{AB4 Mana Cost}}}}$

This is where the biggest approximation is made. In practice, this value would be rounded up so to ensure as little time spent channeling mana when at full mana as possible. However, at the same time each point of mana you finish your evocation with below the maximum is a damage loss, from both the reduced effect of mastery until ideal mana levels are reached, and the need to conserve harder to reach an ideal mana level. I'm not entirely sure how to model this damage loss in the conserve phase, so I made two different simplifications.

In the first approximation I rounded up the number of free casts, so that the damage gain would be,

$\left\lceil {\frac{{m \cdot c(b - 1)}}{{AB4ManaCost}}} \right\rceil \left( {\alpha \left( {1 + \frac{{\beta \left( {M - m \cdot c + \gamma \cdot t(c - 1)} \right)}}{M}} \right)} \right) - t\left\lceil {\frac{{m \cdot c(b - 1)}}{{AB4ManaCost}}} \right\rceil \left( {{\rm{Average DPS of Conserve Phase}}} \right)$

This approximation doesn't penalize the damage of the conserve phase at all and it always rounds up the number of "free" AB casts, and so it should be treated as a maximum approximation.

The second approximation allows fractions of AB casts to occur. The benefit of this approximation is that it differentiates between the number of "free" AB casts between whole numbers. Values closer to their ceiling, or scenarios where you would end up with more mana left after evocation if you used the approximation above, yield higher results than values further away from their ceiling. The drawback to this approximation is that it places an equal value on the mana spent during a burn phase and during a conserve phase, and since the damage is modeled as the time integral of a continuous Damage per Time function, it assumes that partial casts can happen.

Here's the Damage as a function of time and its corresponding integral

$\frac{{d{D_i}(T)}}{{dT}} = \left( {\frac{\alpha }{t}\left( {1 + \frac{{\beta \left( {M - m \cdot c + \gamma \cdot t(c - 1)} \right) + T \cdot \gamma - \frac{{({\rm{AB4 Mana Cost}})}}{t}T}}{M}} \right)} \right)$

${D_i}(T) = \int\limits_{T = 0}^{T = t(\frac{{m \cdot c(b - 1)}}{{AB4ManaCost}})} {{D_i}(T)dT} = \frac{{\alpha \cdot T}}{t} + \frac{{T \cdot \alpha \cdot \beta }}{{M \cdot t}}\left( {\left( {M - m \cdot c + \gamma \cdot t(c - 1)} \right) + \frac{T}{2}\left( {\gamma - \frac{{({\rm{AB4 Mana Cost}})}}{t}} \right)} \right)$

Then the damage gains using this approximation would just be the difference between the above damage and the average conserve damage done in the same amount of time. This approximation should likely be treated as a minimum.

Also, since the dps of a conserve phase is dependent on your own character, and very difficult to approximate using only your character's stats, an average estimate is used for all these calculations.

RESULTS:

Here are the results using the above methods. I used the stats from the BIS Arcane Mage on simulationcraft.org. I'll include a set of estimates with and without Shard of Woe. I used an average conserve phase dps value of 18000. Each set of calculations also take into account that one burn phase will be under the effects of heroism/bloodlust.

← Click Here

TL;DR: Using the above methods the 4pt12 bonus appears to be in the range of a 125 -225ish dps increase, dependent heavily on if you can squeeze in an extra AP or not. The approximations aren't ideal, but with both maximum and minimum approximations the general value of the bonus can be seen. Of course, once simulationcraft is updated a more exact value can be determined!

fok · 07/19/11, 9:51 PM

Very great info, but omg, it seems like full non-set gear with more relative stats will be better than t12p4, isn't it?

kamogawa · 07/20/11, 9:03 AM

I obtained the Moonwell Chalice trinket yesterday and noticed that I was able to activate it simultaneously with the Jeweled Serpent JC trinket and it certainly seemed that both on use effects stacked. I will test it again in a couple of hours when US servers go live. Has anyone else noticed and/or duplicated this effect and can confirm?

Tyrian · 07/20/11, 9:24 AM

The Moonwell Chalice also stacks with Engineering Glove tinker.

Gloinn12 · 07/20/11, 10:25 AM

Can anyone actually confirm that the T12 set bonus changed? There is all this buzz about the 4-piece bonus being stealth buffed, but there is absolutely no concrete proof that it has been changed at all. All there is, is a guy that has the 4-piece and a completely different guy from the same guild saying the 4-piece changed.

Ataxus · 07/20/11, 7:51 PM

Originally Posted by Gloinn12

Can anyone actually confirm that the T12 set bonus changed? There is all this buzz about the 4-piece bonus being stealth buffed, but there is absolutely no concrete proof that it has been changed at all. All there is, is a guy that has the 4-piece and a completely different guy from the same guild saying the 4-piece changed.

Proof: http://i53.photobucket.com/albums/g7...811_191626.jpg

tlosenjin · 07/21/11, 3:22 AM

I just finished compiling simcraft with their changes to Arcane's 4pt12 bonus. Simming the BIS gear set on simulationcraft.org with tier shoulders and chest added to make up the 4p, and toggling the bonus off and on, I got a dps value of 194 dps. Using my own toon with non-heroic 4pt12 and no Shard of Woe the bonus had a dps value of 220 dps. So it seems like the bonus lies in the upper limits of my earlier estimations, and like most people were predicting it isn't great.

Sinless · 07/22/11, 2:41 PM

People who are asking about 4pT11 vs. 2pT12 the short answer is it depends.

The long answer is, typically (and underline that word), if you are replacing ilvl359 tier gear, it's worth it. If you are replacing heroic T11, it's typically (again) not worth it. You have to use Rawr or Simcraft and decide for yourself.