Elitist Jerks

Elitist Jerks (http://elitistjerks.com/forums.php)
-   Druids (http://elitistjerks.com/f73/)
-   -   The Restoration Itemization Thread for 3.3 (http://elitistjerks.com/f73/t88836-restoration_itemization_thread_3_3_a/)

Norfair 01/16/10 11:55 AM

The Restoration Itemization Thread for 3.3
 
This topic is a fresh start from http://elitistjerks.com/f73/t37578-r...n_itemization/ and http://elitistjerks.com/f73/t37441-r...tion_trinkets/. Its focus lies on 3.3 and it will not be updated for any Cataclysm changes (this will be done in a new thread when the time comes). This first post will discuss the different stats and current gearing. The second post consists of a write-up by GTtheBard about the trinkets in the game as they are often valued differently.

This topic is used only for gearing issues. For healing strategies, glyphs, and other druid questions see http://elitistjerks.com/f73/t88239-r...al_discussion/. Keep that topic free from itemization questions as well as they belong to this thread. For short and simple questions see http://elitistjerks.com/f73/t36776-d...imple_answers/.

Failing to post in the correct thread will lead to an infraction!







Throughput stats


Spellpower
Spellpower is our most important stat. It increases the healing of each spell differently. The talents that affect the coefficent for each spell work as follows:
  • Gift of Nature and Genesis stack additively. They increase HoTs by 15% (1.15 coefficient) instead of 10% and 5% seperately (1.155 coefficient). They also stack additively with improved Rejuvenation (making the coefficient for this spell 1.3).
  • Empowered Rejuvenation affects both the final Lifebloom tick as the initial Regrowth heal. Genesis does not affect these.
  • Master Shapeshifter, Tree of Life aura and Gift of Nature with Genesis are all multiplicative. This means: spellpower * 1.15 (GoN + Genesis) * 1.06 (ToL aura) * 1.04 (Master Shapeshifter).
  • Wild Growth's extra bonus healing is divided equally over the ticks, in contrast to the base healing which heals less each tick.
In the table below the number in the "no talents" columns are how they scale without any talents. The full talents column is the coefficient if you have Master Shapeshifter, Gift of Nature, Improved Rejuvenation, Empowered Touch, Empowered Rejuvenation, Genesis and Tree of Life active. The numbers in brackets are the coefficients per tick. The Nature's Splendor talent adds ticks with these coefficients, thus basically increasing the total coefficient for each spell it affects.

SPELL______________NO TALENTS____FULL TALENTS___
Lifebloom HoT0.666 [0.095]1.013 [0.145]
Lifebloom Bloom 0.5160.751
Rejuvenation1.88 [0.376] 3.233 [0.647]
Regrowth Initial0.5370.781
Regrowth HoT1.316 [0.188]2.002 [0.286]
Nourish w/o HoT0.6711.056
Nourish with HoT0.8061.267
Healing Touch1.6112.439
Wild Growthn/a1.225 [0.175]


Haste rating
Haste rating is the second most important stat for resto druids. Since 3.3 we need a lot more haste in order to get to a 1 second global cooldown, which next to spellpower is the only way to increase our throughput with HoTs (not counting 4T9).

32.79 haste rating increases your haste by 1%. Paladin aura and Moonkin aura do not stack; both give 3% additional haste. Wrath of Air totem provides an additional 5% haste. Here is a list of the amount of haste rating you need with 5/5 GotEM in order to obtain a 1 second global cooldown:

BUFFS______________HASTE PERC___HASTE RATING___
No buffs36.36 %1193
Aura only32.39 %1063
WoA only29.87 %980
WoA + aura only26.09 %856
CF only32.39 %1063
CF + aura only28.54 % 936
CF + WoA only26.09 % 856
CF + WoA + aura22.42 %735

Thanks to uliko, itecho84 and Zoltair's post for finding the mechanics. Thanks to cuddlekin for finding the numbers for 3.3. See his post for a more elaborate list of haste numbers (includes partial GotEM talents).


Critical strike rating
Critical strike rating is a very poor stat for resto druid, unless you are using 4T9.

45.91 critical strike rating increases your chance your spells critically hit by 1%.







Mana Regeneration stats
Mana regeneration stats are not very important for a restoration druid as each type of gear already contains a good amount of intellect, spirit and/or mp5. Spirit also gives spellpower through the improved ToL talent. Because of this, the spirit items are often better than their mp5 equivalents of the same itemlevel.

The mana regeneration formula on level 80 is: ManaRegen(SPI, INT) = (0.001+SPI*0.003345*(INT^0.5))*5


Intellect
1 intellect provides the following:
  • 1 intellect increases mana pool with 15 mana. On a typical 8 minute fight with Kings and IMotW this can be converted to 0.175 mp5
  • Assuming 1400 intellect and 1400 spirit, 1 intellect increases your main mana regeneration with 0.351 mp5 out of the five second rule or 0.176 mp5 in the five second rule (with Kings, IMotW and Intensity).
  • Since Replenish works off the size of your mana pool, 16.83 (Kings and IMotW) extra mana means 0.210 mp5 more through Replenishment.
  • 1 intellect also increases the chance you critically hit with spells by 0.00676% (this is with Kings and IMotW).
This sums up to 0.561 mp5 (0.572 with the meta gem) and a little bit of crit chance. However, this number can change a lot based on your intellect, spirit, Replenishment uptime, duration of the encounter and time spent in the five second rule.


Spirit
1 spirit provides the following (assuming Living Spirit, Intensity, Kings and Improved Mark of the Wild):
  • Assuming 1400 intellect and 1400 spirit, 1 spirit increases your main mana regeneration with 0.807 mp5 out of the five second rule or 0.404 mp5 in the five second rule.
  • If you have improved Tree of Life, 1 spirit also increases your spellpower with 0.194.
For a more elaborate table of values, see uliko's post.







Gearing
Druid gearing in a nutshell:
  • Aim for as much spellpower as possible
  • Try to get haste cap by using haste pieces instead of crit pieces
  • Spirit-based items are often better than mp5-based items
  • If you need mana, socket for bonus and/or use Insightful meta. Also swapping trinkets is a flexible way to get more mana. Other ways to gear for mana is bad.

Items
Some consensus that has been reached so far:
  • 4T9: Still good, but its power is based on your crit chance. To get more crit, you usually have to give up haste, which we need more of since patch 3.3.
  • Trauma: initial logs show proc can do up to 10% of total effective healing.

The tier 10 4-set bonus
The tier 10 4-set bonus has a chance to proc a new Rejuvenation on each Rejuvenation tick. Thanks to Shelendil, Xasthur, ttyl, Drane and sheaebay for testing this set bonus, we now know the following:
  • It can proc off a procced Rejuvenation
  • A 'normal' Rejuvenation will not overwrite a procced Rejuvenation
  • It will pick the target with lowest health that does not already have a Rejuvenation of the caster on him/her
  • The range is 40 yards from the caster, not the player who has the Rejuvenation
  • It can only proc on ticks, not on application
  • It can be Swiftmended
  • It shows on Grid
  • It is affected by all talents and glyphs that affect Rejuvenation.

Enchants
Gems

GTtheBard 01/16/10 2:20 PM

Trinket discussion always seems to be a point of interest when raiding, and more often than not specific trinket discussion is lost in the endless amounts of pages around here. Unlike the previous topic, this is focusing on competitive end-game Trinkets only, split into Throughput and Regen.

Throughput is self explanatory; these trinkets give druids the most spellpower or healing output compared to other trinkets. Ideally, you'll be using these trinkets when Regen is not an issue.

I'm lumping all trinkets with any sort of regen effect into the Regen pile, even if they have a lot of passive spellpower. Mechanics of all these trinkets will be discussed after the list.

Throughput:
[Althor's Abacus] - Gunship Battle (25)
[Illustration of the Dragon Soul] - Sartharion (25)
[Eye of the Broodmother] - Razorscale (10)

Regen:
[Solace of the Fallen] - Lord Jaraxxus (25)
[Spark of Hope] - Kologarn (10)
[Sliver of Pure Ice] - Lord Marrowgar (10)
[Purified Lunar Dust] - 60 Emblem of Frost
--[Show of Faith] - Yogg-Saron (25 Hard)
--[Sif's Remembrance] - Thorim (10 Hard)
[Ephemeral Snowflake] - Halls of Reflection (5 Heroic - Marwyn)

Other:
[Titan-Forged Rune of Audacity] - Wintergrasp Vendor (25 Badges)
[Titan-Forged Rune of Alacrity] - Wintergrasp Vendor (25 Badges)
[Battlemaster's Ruination] - PVP Vendors (62000 Honor and 1650 Arena Rating)
[Battlemaster's Vivacity] - PVP Vendors (62000 Honor and 1650 Arena Rating)

Notes on Trinkets:

[Althor's Abacus]
-Smart heal
-Can crit
-Does not scale with spellpower
-Procs off HOT ticks
-45 second internal cooldown
-Normal and Heroic versions do not work together

Quote:

Originally Posted by Carnathagia
2: I tested this in the arena with a group of 4. 3 were taking FoK and poison damage, 1 (Saht) was being autoattacked. 4 Rejuvenations were maintained on each damaged target.
World of Logs - Real Time Raid Analysis
Thanks to Jayde for her help.

[12:06:46.555] Emraldè gains Rejuvenation from Emraldè
[12:06:51.512] Emraldè Echoes of Light Joslin +0 (O: 6521)
[12:07:37.344] Emraldè Echoes of Light Saht +6520
[12:08:23.864] Emraldè Echoes of Light Saht +5288 (O: 1659)
[12:09:07.845] Emraldè Echoes of Light Saht +6599
[12:09:53.053] Emraldè Echoes of Light Saht +6416
[12:10:38.487] Emraldè Echoes of Light Saht +6389 (O: 101)
[12:11:24.104] Emraldè Echoes of Light Saht +6518

The first proc occurs 5 seconds after the first application of rejuvenation, but procs on the lowest HP player since everyone is at full health for full overheal. Each subsequent proc happens on Saht, who always has the greatest health deficit. They occur at an interval of 46, 46, 44, 46, 45, and 46 seconds.

The proc is roughly equivalent to a critical, instant cast Nourish on the lowest health raid member in 40 yards, every 45 seconds. It could be a nice increase to raid survivability when there are 6 of these in your raid. Yes, it's ~2.5% total healing, but its valuable healing that can save someone's life. If the overhealing is high, it is because there was no healing for it to do when the proc happened.


[Solace of the Fallen]
-Stacking 144 mp5 (Heroic - 258 ilevel), 128 mp5 (Normal - 245 ilevel)
-Normal and Heroic mp5 effects stack


[Spark of Hope]
-Applies effect before talent reductions (Tree of Life, Moonglow, Tranquil Spirit, etc.)
-100 Spirit => 115 Spirit = 17 Spellpower (assuming 3/3 Improved Tree of Life)

Quote:

Originally Posted by GTtheBard
Assumed Talents: Improved Tree of Life (3/3), Moonglow (3/3)
Base Mana at 80 = 3496

Mana Reduction on Rejuvenation (18% Base Mana):
Cost = (3496*.18) = 629 Mana
(629 - 42) * (1 - (.09+.20)) = 416
Without Spark, Rejuvenation costs 446 Mana, a reduction of 30 Mana.

Mana Reduction on Regrowth (29% Base Mana):
Cost = (3496*.29) = 1014 Mana
(1014- 42) * (1 - (.09+.20)) = 689
Without Spark, Regrowth costs 719 Mana, a reduction of 30 Mana.

Mana Reduction on Wild Growth (23% Base Mana):
Cost = (3496*.23) = 804 Mana
(804 - 42) * (1 - .20) = 609
Without Spark, Wild Growth costs 643 Mana, a reduction of 34 Mana.

Mana Reduction on Lifebloom (28% Base Mana):
Cost = (3496*.28) = 979 Mana
(979 - 42) * (1 - .20) = 748
Without Spark, Lifebloom costs 782 Mana, a reduction of 34 Mana.

Mana Reduction on Nourish (18% Base Mana):
Cost = (3496*.18) = 629 Mana
(629 - 42) * (1 - .09) = 534
Without Spark, Mana Cost is 572 Mana, a reduction of 38 Mana.

Mana Reduction on Swiftmend (16% Base Mana):
Cost = (3496*.16) = 559 Mana
(629 - 42) = 517
Without Spark, Mana Cost is 559 Mana, a reduction of 42 Mana.

Assuming a five minute fight with a 5x1 rotation (250 Rejuvs, 50 Wild Growths), you will save 9400 Mana, making this trinket roughly 157 mp5 under such conditions, not taking into account the effect from Spirit.


[Sliver of Pure Ice]
-Normal Mode and Heroic Mode have 2 minute cooldowns
-Used on cooldown, Normal = 68 mp5, Heroic = 76 mp5


[Purified Lunar Dust]
-45 second ICD, 10% proc rate; for modeling, using a 55 second cooldown
-Can proc simultaneously with similar trinkets ([Je'Tze's Bell], [Show of Faith], etc.)
-Returns 912 mana every 55 seconds, roughly 76 mp5

Quote:

Originally Posted by Qiyo (Post 1535284)
Reviewing parses of my own logs (yes, small sample is small), I find it to proc once per minute on average. Assuming wowhead data is correct, the 45sec ICD combined with 10% chance to proc on cast makes this proc on average once every 55sec (45 sec ICD + 10 * 1sec GCD casts), which seems to match my experience.

This brings its mp5 value from 101mp5 down to 76mp5 on average.



[Ephemeral Snowflake]
-0.25 second ICD (unconfirmed)
-Procs off HOT ticks
-Is not affected by Clearcasting or Overhealing
-Maximum of 220 mp5

KrinKer 01/16/10 4:54 PM

Ok quick question, as a pure Throughput trinket, would "regular" [Althor's Abacus] be better than [Illustration of the Dragon Soul]?

Mainly, is the spellpower loss from althor's abacus made up by it's proc compared to IotDS ??

Norfair 01/16/10 6:10 PM

Quote:

Originally Posted by KrinKer (Post 1526829)
Ok quick question, as a pure Throughput trinket, would "regular" [Althor's Abacus] be better than [Illustration of the Dragon Soul]?

Mainly, is the spellpower loss from althor's abacus made up by it's proc compared to IotDS ??

On most fights it does around 3% of my healing. So yes, it does beat the loss of 50 spellpower.

Hamlet 01/16/10 6:32 PM

It's 21 spellpower. And yeah, I haven't looked into the Abacus in detail, but 3% effective would be a lot better than 21 spellpower. And finally doing away with rampup time would be nice.

Carnathagia 01/16/10 8:05 PM

Quote:

Originally Posted by Arawethion (Post 1526891)
I haven't looked into the Abacus in detail, but 3% effective would be a lot better than 21 spellpower.

A minor note: The trinket does not benefit from spellpower and as such will scale negatively with gear.

((5550+6450)/2)*(1+0.15[Crit%]*0.5[CritBonus])=6423[AverageHeal]/45[Cooldown]=142.7 healing per second assuming no overhealing.

Playered 01/16/10 10:58 PM

Quote:

Originally Posted by Carnathagia (Post 1526945)
A minor note: The trinket does not benefit from spellpower and as such will scale negatively with gear.

(5550+6450/2)=6000[Average_Heal]/45[Cooldown]=133.3 healing per second assuming no overhealing.

Please try and pay attention to what is listed about the trinket because then you would notice that it can crit which means your formula is inaccurate and also because it is a smart heal it should have relatively low overhealing especially when we reach harder content where more healing is needed.

You seem to be trying to imply that the trinket is worse than it actually is which is misleading considering you could have thrown in some ballpark overhealing figures from the current content to go along with your HPS evaluation. It's somewhat odd how we value haste and SP on a "best case" scenario and yet seem unwilling to do so with procs like this.

Carnathagia 01/16/10 11:30 PM

Quote:

Originally Posted by Playered (Post 1527026)
You seem to be trying to imply that the trinket is worse than it actually is which is misleading considering you could have thrown in some ballpark overhealing figures from the current content to go along with your HPS evaluation. It's somewhat odd how we value haste and SP on a "best case" scenario and yet seem unwilling to do so with procs like this.

If I factor in overhealing, considering a best case scenario, then this trinket proc is worse than I described above.

Heroic Twins

With 10.7k effective healing per second, in an ideal aura type fight where I was in 40 yards of all range members, the trinket proc had 23.8% overheal. The point is, the healing can't be attributed as a percent of effective healing. In a fight where 4k-5k hps is typical, this will be closer to 3% healing done. With closer to 11k hps, this is only 1% healing done. It should be considered as healing per second instead of a percentage of healing done.

thunderbg 01/17/10 1:06 AM

For those making a wish list, after digging through and looking at the items available in normal ICC25/10 and ToGC10/25 minus Anub in 25, I have come up with the following lists:
(Note: The gemming was done assuming you are a Jewelcrafter, since that's what I am.

Termi (4pcT9 Resto with 864 haste, 856=cap): Profiler - Wowhead
Termi (4pcT10 Resto with 857 haste, 856=cap): Profiler - Wowhead

Termi (4pcT9 Resto(CF) with 774 haste, 735=cap): Profiler - Wowhead
Termi (4pcT10 Resto(CF) with 787 haste, 735=cap): Profiler - Wowhead

Another note with gemming, I had originally planned just a nightmare tear to activate meta, as regen's not needed at all, however putting the purifieds in the places I have them actually work out to be the best choice as far as I can tell.

Feel free to use these as a consideration for your own lists.
Edit:
I personally prefer to see a mix of lists others make, to get an idea of what they value, how they mix and match, etc.
I won't discuss this list anymore and ask that no one else do so as well. Use it for comparison if you want, otherwise ignore it.

Hamlet 01/17/10 1:46 AM

At the 277 level, the Spirit/haste shoulders are a better 4+1 offset piece than the badge chest. So I suppose for a sub-277 BIS list, yours is technically right. But given the Emblem economy, it might be best to buy the 4 non-Shoulder pieces and use the ICC shoulders. Stats are just as good as using the badge chest, and you save Emblems, and then don't you don't have to further waste Emblems changing tier pieces when you start having access to 277 loot.

Carebare 01/17/10 1:51 AM

Even if you assume that 1 haste = 1 spell power to the cap, using your Dragon's Eyes on haste is a poor choice. Runed =39 vs Quick =34, already a downgrade before we get into the fact that socketing yellow/red vs orange is a waste of 1 spell power every time you do it with the normal versions (23 spell power + 20 haste vs 24 spell power + 20 haste).

That aside, I'd like to avoid BiS lists generally. Discussing specific pieces is great. BiS are complicated, people will inevitably disagree over socketing. Some will choose cloth, some are in guilds that won't allow them to take cloth over a cloth-wearer, etc etc. Some will prefer to use a 735 set-up with T9 utilizing a 3/3 CF build allowing for some more crit. As you can see I've only mentioned a few examples, but they are more than enough to show that making a universal BiS is generally not going to happen.

Playered 01/17/10 7:58 AM

I'm still personally in favor of using [Sanguine Silk Robes] over [Shoulders of Frost-Tipped Thorns] because it does seem slightly better and from initial inquiry there wasn't going to be every cloth user wanting this item in my guild (mainly SPriests).

Item Comparison Tool - World of Warcraft

Assuming the socket bonus is +9 on the chest and +7 on the shoulders and you meet them both it ends up with a difference of:
[Sanguine Silk Robes] set: 24 haste, 3 sp (1 base, 2 from socket bonus), 10 spirit.
[Shoulders of Frost-Tipped Thorns] set: 24 crit, 10 haste.

While the difference is not that huge (14 haste, 3 sp, 10 spirit vs 24 crit) it is still a superior option in my mind especially as past the haste cap we can move a reckless gem into a runed somewhere resulting in that excess haste becoming SP while with the shoulders set you cannot do an equal exchange with the crit.

Hamlet 01/17/10 12:45 PM

I don't like assuming the socket bonuses will be met, since the Robes have a blue socket. But even still, just due to the haste vs. crit, the set with the Robes probably is ever so slightly better.

Playered 01/17/10 1:07 PM

Quote:

Originally Posted by Arawethion (Post 1527363)
I don't like assuming the socket bonuses will be met, since the Robes have a blue socket. But even still, just due to the haste vs. crit, the set with the Robes probably is ever so slightly better.

If you intend to use a reckless gem in it then you have the choice of using a purified gem for 10 spirit + socket bonus (9 SP) or ignoring the socket bonus and putting a runed there with a gain of +2 SP. Considering that 10 spirit gives you 1.94 SP it seems illogical to use a runed instead (or rather it doesn't really matter).

If you have enough haste that you don't need to use a reckless at all then obviously pure runed will win but I couldn't honestly say where would be the best choices at the moment to do so without a complete/confirmed drop list available to plan this.

Demagogue 01/18/10 3:29 PM

Quote:

Originally Posted by Carebare (Post 1527110)
Even if you assume that 1 haste = 1 spell power to the cap, using your Dragon's Eyes on haste is a poor choice. Runed =39 vs Quick =34, already a downgrade before we get into the fact that socketing yellow/red vs orange is a waste of 1 spell power every time you do it with the normal versions (23 spell power + 20 haste vs 24 spell power + 20 haste).


Carebare, I've always had a hard time wrapping my head around this part. I mean I get (mathematically) that 23SP+20Haste is less than 12SP/10Haste x2. But would you go to the point of essentially socketing straight reckless gems?

For example, you currently have 17 red/orange gems with CF. Putting Dragon's eye in 3 of your slots and leaving the current 3 Reckless you have to hit your cap. That brings us to 11 Using only Reckless gems that would get you 132SP/110Haste. As pure Runed Gems you would get 253SP.

Looking at Norfair's Haste chart at the top of the thread CF is worth 130 Haste.

No buffs 36.36 % 1193
CF only 32.39 % 1063

So in the end is it really coming down to comparing 121SP (253-132) [in your case] vs the 7 Talent points you could take out of your Balance tree. Or am I missing something?

***Note***
I realize I just put you 20 Haste rating under cap at the moment, but I am just using this as an example. Please assume your gear had that 20 haste somewhere else. (Weapon/T10 gloves/etc)


All times are GMT -4. The time now is 7:01 AM.

Forum Infrastructure by vBulletin 3.6.12 ©2000-2007, Jelsoft Enterprises Ltd.