Jump to content


Photo

Resto PvE Compendium and General Discussion


  • This topic is locked This topic is locked
420 replies to this topic

#1 Hamlet

Hamlet

    Mike Tyson

  • • Guide Author
  • 11,540 posts

Posted 13 January 2010 - 07:29 AM

This post is a brief compendium of the currently widely accepted information about PvE raiding as a Restoration Druid. It should serve as the starting point for further discussions on this thread about Resto healing in the current expansion.

This thread is for discussion of general Resto healing topics: talents, glyphs, spell usage and techniques, boss-specific healing strategies, etc. Specific discussion of itemization and gear selection is left for a separate thread here: http://elitistjerks....n_thread_3_3_a/ . Please read the Forum Rules before posting. In addition, don't ask questions that are clearly answered in this post; they will be infracted.

=Talents and Glyphs=

The Spec


The following talents will be included in any Resto spec for raid use: Genesis, Moonglow, Nature's Majesty, Nature's Splendor, Improved Mark of the Wild, Nature's Focus, Natural Shapeshifter, Intensity, Omen of Clarity, Master Shapeshifter, Improved Rejuvenation, Nature's Swiftness, Gift of Nature, Nature's Bounty, Swiftmend, Empowered Rejuvenation, Revitalize, Tree of Life, Improved Tree of Life, Gift of the Earthmother, Wild Growth.

The degree of choice in you spec comes when you allocate the last few points. The candidates are:
  • Nature's Grace: Somewhat improves your Regrowth, and is a prerequisite to Celestial Focus.
  • Celestial Focus: Provides a small amount of haste to all of your spells, and lowers the haste cap for Rejuvenation (see below). You have to spend 1 extra filler point in the Balance tree to get this.
  • Empowered Touch: Frequently taken for its effect on Nourish. Good to take if you plan on using Nourish very much, for example, if you ever have to heal tanks.
  • Subtlety: Usually taken simply to get to the next tier, this can occasionally prevent a death.
  • Naturalist: Used for very specific Glyph of Healing Touch-based builds.
  • Tranquil Spirit: the mana savings from this will only be significant if you're chaincasting Nourish frequently. Will be used in a heavy tank healing build.
  • Living Seed: This is a nice improvement to Regrowth and Nourish. If you're not delving into the Balance tree for Celestial Focus, you'll probably pick this up.
  • Living Spirit: A good place to put any extra points.
  • Improved Barkskin: Another place you might put extra points.
  • Natural Perfection: A small throughput gain for any spells that can crit. Usually only used with the 4T9 set bonus.

If you're just getting started or looking for a ready-made spec option, this is a versatile one that's good to start with:


The main variation will be to take some of the points out of Balance (typically done by well-geared players who can reach the haste cap without Celestial Focus):



Talent details


A few more mathematical facts I just want to keep here for the people who are interested:
  • Genesis: Affects the HoT portions only of Lifebloom and Regrowth. Affects Tranquility. Stacks additively with Gift of Nature, Improved Rejuvenation, and Glyph of Healing Touch.
  • Moonglow, Tree of Life, and Tranquil Spirit stack additively. Mana cost is rounded down after multiplying. Applies after -type effects. Moonglow does not affect Lifebloom. None of these talents affects the Lifebloom mana refund.
  • Nature's Grace: Does not proc from Rejuvenation crits with the 4T9 bonus, or the bloom of Lifebloom.
  • Celestial Focus and Gift of the Earthmother stack multiplicatively with each other and with all other haste bonuses. The Lifebloom GCD reduction multiplies the GCD by 0.9, rather than giving 10% additional haste.
  • Master Shapeshifter and Tree of Life Aura stack multiplicatively. Tree of Life Aura does not stack with the Paladin talent Improved Devotion Aura.
  • Improved Mark of the Wild and Living Spirit: These stack multiplicatively with each other and with Kings.
  • Omen of Clarity: Has a 3.5/60 = 5.83% chance to proc on spellcast. Does not proc from and is not consumed by shapeshifting. Gift of the Wild, if cast on 25 people, has a very high chance of proccing this.
  • Wild Growth: The base heal decays with each tick, but all extra healing from spellpower is equal across the ticks.
  • Empowered Touch: adds 0.2 and 0.4 to the base spellpower coefficients of Nourish and Healing Touch.
  • Empowered Rejuvenation: multiplies all coefficients of all other healing spells by 1.2 (note how these two work slightly differently).

Glyphs


Major Glyphs: Glyphs are largely situation-specific. The two most commonly used are and . A number of others have potential use at various fights:

  • : This significantly changes how Rejuvenation works, so the choice of whether to use it is important. For standard Rejuv-blanketing around the raid, this is typically not helpful as it reduces the number of targets you can have Rejuvenation on at once and requires you to refresh more often. But where you need more focused healing on fewer targets, the Glyph makes your HoT significantly stronger.
  • : Primarily a tank healing tool.
  • : A respectable mana return (on the order of 40 MP5 if used on cooldown). Often a good filler when you don't have three important healing Glyphs to use.
  • : Reliably strengthens both the direct heal and HoT on the tank, which is valuable. Otherwise, generally a small benefit, since you don't repeatedly cast Regrowth on one person very often.

A few Resto Glyphs are currently rarely used:
  • : Rarely provides a significant amount of healing. Most fights simply do not have people sitting at low HP for any continuous length of time.
  • : Frees up some GCD's if you roll Lifeblooms, but we rarely build rotations around Lifebloom anyway.
  • : This is primarily a leveling Glyph. But using this and the Naturalist talent does give us one new tool--a heal with cast time significantly under 1 second. But since the spell is locked out by the GCD anyway, this of limited benefit. Occasionally you might want the unusually rapid heal to respond to some boss ability.
  • : An option to keep in mind for fights where special mechanics make it easy for Rebirthed person to die immediately. Our Rebirth is an important contribution to the raid, and you don't want to waste it.

For more in-depth discussion of Glyphs, see the post immediately below this one.

Minor Glyphs:
can save a bit of mana if you have to rebuff people during a fight. Other than that, no Glyphs affect our combat mechanics. can prevent you from being embarrassed with no reagents, and saves an inventory slot to boot.

=Gear=

http://elitistjerks....n_thread_3_3_a/ is the place to go for detailed discussion of items and gearing. This one has the basic background.

Stats


In roughly descending order of importance.

Spellpower:
Improves all of our heals. For more detail on spellpower coefficients, you can see [below post] or look into TreeCalcs. For the purpose of improving throughput (healing done per unit time), spellpower will always be a top stat.

Haste rating:
32.79 haste rating gives 1% spell haste. This reduces the cast time and GCD of all of our spells, to a minimum of 1 second GCD. An important value is the amount of haste needed to reach this cap, because beyond that value, haste has no effect on any of our instant-cast spells or on Nourish. With all raid buffs and 5/5 Gift of the Earthmother, the cap is 735 haste with 3/3 Celestial Focus, and 856 haste without Celestial Focus. See TreeCalcs for the haste cap with other combinations of buffs.

Spirit:
Spirit provides 0.15 spellpower per point with Improved Tree of Life. With Blessing of Kings and Improved Mark of the Wild, it provides 0.168 spellpower (0.194 with Living Spirit). This makes it a weak throughput benefit.

Spirit and Int will both increase your natural regen, which scales like (0.016725 * sqrt(Int) * Spi) MP5. With 3/3 Intensity, you get half this amount while casting.

Critical strike rating:
45.9 crit rating gives 1% to crit. Crit heals do 1.5 times the healing of non-crit heals (1.545 with a Revitalizing meta). Nourish, Healing Touch, Swiftmend, the Regrowth direct heal, the Lifebloom bloom, and Rejuvenation ticks with the 4T9 set bonus can crit. Crit rating provides significantly less throughput increase per point than haste rating.

Intellect:
Intellect provides 0.006% to crit per point. With Blessing of Kings and Improved Mark of the Wild, it provides 0.00673% (148.5 points per 1% crit, roughly 1/3 of a crit rating). This makes it an even weaker throughput gain than Spirit and crit.

Similarly, with these talents, one point of Intellect increases your maximum mana by 16.8, your regen from Replenishment by 0.168 MP5. It also increases your natural passive regen, as mentioned above. Int will generally provide slightly less regen than Spirit, even if you have Revitalize.

MP5:
Adds no throughput. Adds a similar amount of regen to Spirit and Intellect.

Stat priority and Basic Gear Selection


Our two primary throughput stats are spellpower, and haste rating until the cap. Crit, Int, and Spirit are all secondary.
In practice, mana problems are solved by using trinkets that provide substantial amount of mana, , Innervating yourself, , potions, or Mana Tide Totem. You won't go out of your way to gear for Spirit, Int, or MP5 for mana reasons.

Caster epics have Intellect, Stamina, spellpower, and 2 out of the following: crit rating, hit rating, haste rating, Spirit, or MP5. Keep in mind a few rules of thumb, which are enough to get a quick estimate of the value of any piece:
  • Higher-ilvl gear is stronger. Spellpower usually increases very regularly with ilvl, so we have a consistent gain.
  • More sockets are better (since they allow us to stack more spellpower, primarily).
  • Haste is significantly better than all other options. Haste/Spirit is ideal, but you can consider using haste/crit, haste/MP5, or even haste/hit before any non-haste items, until you're at the cap.
  • Set bonuses and unusual procs (like , , or ) provide additional ways to increase your total healing output where you can't gain any more spellpower or haste. That tends to make all of these things good choices.

For exact numerical comparisons, plug your setup into Treecalcs.

Other Topics


Gems: Gem in all sockets, or / to make the haste cap. If you're using an Insightful meta, use one wherever you can get the best socket bonus. In your meta socket (only use hats with meta sockets), use either or , depending on whether mana is regularly a concern for you.

Set Bonuses: The 4T8, 4T9, and 4T10 bonuses all provide significant buffs to Rejuvenation. Use whichever is the best one you have access to. All other set bonuses are comparatively inconsequential.

Idols: for 30 Emblems of Frost is the best. Until that's available to you, buy for 25 Emblems of Triumph (which is nearly as good), otherwise for 25 Emblems of Valor. You can also consider using Awakening to alleviate serious mana problems (remember you can swap to it mid-fight).

Trinkets: A large number of trinkets have spellpower, and you'll want to get two of these. The only likely exceptions are or , again if you have mana problems (either should provide around as much regen as ). The highest-spellpower trinkets are and . There's a good line of healer trinkets with passive spellpower and a mana proc, of which is now an easily-accessible example. The best trinket of this variety is . More on trinkets at http://elitistjerks.com/f73/t88836-restoration_itemization_thread_3_3_a/#post1526740 .

Consumables: Use and a 46 spellpower food (such as a Feast). You should also keep handy.

Enchants
Excluding profession bonuses, you should use:
  • Head: 30 spellpower/10 MP5 (Wyrmrest Accord revered)
  • Shoulders: 24 spellpower/8 MP5 (Sons of Hodir exalted)
    In the head and shoulder slots, you might favor the crit version if you have the 4T9 set bonus.
  • Back - 23 haste
  • Chest - 10 stats. 8 stats is cheaper option at a tiny stats loss.
  • Wrists - 30 spellpower
  • Gloves - 28 spellpower
  • Leggings - 50 spellpower/20 spirit. 50 spellpower/30 stamina is another option if you want more HP.
  • Boots - Run speed/15 stam. Run speed/9 stam is a cheaper option until you need the HP.
  • Weapon - 63 spellpower to a 1H, or 81 spellpower to a staff.

Professions


Tailoring lets you trade some haste rating for a spellpower proc. Lightweave Embroidery gives 295 spellpower for 15 seconds, 35% proc on spellcast, 60 second cooldown. It can proc off HoT ticks, making the average proc time instantaneous. 72 spellpower is a decent estimate of the average. Amusingly, the enchant also gives 1 Spirit for some reason.
If you're at or near the haste cap, shedding 23 haste can be a convenience for your gear/gem setup. Otherwise, you'd probably prefer the constant spellpower from other professions.
A minor convenience of Tailoring is that for the cost of one , you can swap your leg enchant between 20 Spirit and 30 Stamina for fights where a little extra HP is helpful.

Beyond that, Alchemy, Blacksmithing, Enchanting, Inscription, Jewelcrafting, and Leatherworking all provide a roughly equal gain of 46-48 spellpower. Blacksmithing and Jewelcrafting give their bonuses with gems, which allows you to take the bonus in the form of haste instead of spellpower if you're below the cap.

Alchemy: At any time, either or Mixology (with your ) will give you 47 spellpower.
There are number of small conveniences: , , double-duration Flasks, and the ability to make your own .

Blacksmithing: An extra socket each in your wrists and gloves, each with a , gives 46 spellpower, or 40 haste rating.

Enchanting: 23 spellpower to each ring gives 46 spellpower.

Inscription: 70 spellpower/15 crit to shoulders in place of the Sons of Hodir enchant gives 46 spellpower.

Jewelcrafting: 3 in place of 3 gives 48 spellpower.

Leatherworking: 76 spellpower to bracers in place of the usual 30 spellpower gives 46 spellpower.

Engineering and Skinning are weaker, both giving less benefit than the 46 spellpower.

Engineering lets you replaces 23 haste on your cloak with 27 spellpower. Since we have no particular use for Hyperspeed Accelerators, this is a weak profession.

Skinning gives 40 crit rating.

Mining and Herbalism provide minor survivability benefits, but no healing benefit.

=Casting Mechanics=

It's important to not waste time between casts. This is an easy way to lose quite a lot of healing over the course of a fight without realizing it. This section contains a bit of information that you should know about spell targeting, timing, haste, queueing, and the GCD.

Targeting


By default, a heal targets your target, if it exists and is friendly. Otherwise, if you have Auto Self Cast activated (Interface->Game->Combat), it will target yourself. If not, you will get the dreaded "blue hand" asking you to choose your target.

You can control targeting priority using macros. For example, if you want to heal your target's target if your target is hostile:
#showtooltip
/use [@target, help][@targettarget, help] Rejuvenation

Most healers use mouseover macros of some variety (either with Clique (see below), or with keyboard binds, or both). The basic form of a mouseover macro is this:
/use [@mouseover, help] Rejuvenation

You can any combination of click-casting and keybinds to heal effectively, so long as you don't use the "blue hand."

Mechanics


The theorycrafting reflected in this article and in the associated spreadsheet uses the following understanding of spellcasting rules:
GCD: whenever you attempt any GCD-incurring action (including any spellcast), a client-side GCD begins to run immediately. If the action fails, the client cancels the GCD when it finds out about the failure from the server. If the action succeeds, the GCD runs to completion. The client will not transmit any action to the server while a GCD is running.
Queueing: when a spellcast ends on the server, if the server has recently (within 200-300 milliseconds) received a cast command from the client, it will begin that cast immediately.

Timing


First, recall that the client processes events when you release the key. Keep that in mind for learning your timing.
After a Regrowth, Healing Touch, or Rebirth: You want to press the next spell such that it arrives at the server in the "queueing window" as your spell is ending. Start spamming your next spell when your spell hits the red part of the Quartz cast bar, and stop once the new cast begins.
After an instant: No queueing here, and the ability to cast the next spell is determined by the client-side GCD. Watch your Quartz GCD spark and press your spell right as it ends. Pressing it early is a waste and will return a "spell not ready" error in your client.
After a Nourish or a Glyphed Healing Touch: Identical to an instant cast. Press your spell as the GCD ends.

We spend the great majority of our time casting instants and rarely get to use the spell-queueing system. So you want to get very comfortable with the rhythm of your GCD spark. If done perfectly, each GCD begins as the previous one ends, with no visible gap.
You should know when to queue a cast-time spell as well, to maximize the value of your Regrowth.

=Our spells=

This section is short, since discussion of various healing techniques and spell usage will be largely be the focus of the discussion thread.

Rejuvenation: The workhorse of our healing. The ability to have this ticking on up to 17 people in a raid at once is what really sets us apart from other heals. Together with Wild Growth, allows the most healing done per unit time of any of our spells. You will use this heavily at most boss fights. Get used to instinctively putting it up on tanks, people targeted by boss abilities, people with damaging debuffs on them, anyone prone to damage based on fight mechanics (often melee), and anyone else if you have time left.

Wild Growth: Strong multi-target heal. The 7s duration means it's used more like a reactive heal than a true HoT. It will automatically target the 5/6 lowest-HP people within range (not necessarily including the target). This combined with Rejuvenation accomplishes the majority of our raid healing. In heavy damage situations you'll use Wild Growth on cooldown. Make sure to cast it anytime an AoE effect hits some people in the raid. Wild Growth is unusual in that in can be targeted on a hostile unit and will still apply to the lowest-HP raid members near that unit.

Regrowth: Medium direct heal with a long HoT. This is a good way to get a single direct heal on someone in the raid who needs it. It shines particularly when you need to cast a few in row on various people and can take advantage of Nature's Grace. Chaincasting this on one person tends to be weak since it wastes the HoT--use Nourish for that. Regrowth also provides an added HoT to keep on tanks.

Nourish: A larger direct heal that attains its maximum strength when the target has HoT's on them. This is our only way to continually pump healing into one target. Use it when you're healing a tank, or anytime you've already applied HoT's to a target and still need to heal them further.

Lifebloom: A tricky spell to use. It's most often simply a way to drop a bit of extra HoT power on someone after they already have Rejuvenation. It's a nice spell to drop on tanks in spare GCD's, or on raid members who have some kind of DoT who are likely to make use of the healing. For advanced use of this spell, you can restack it right before it expires, only letting it bloom on the third stack. There's rarely a reason to roll it once it's stacked to 3, as it quickly becomes very expensive due to the mana refund mechanic (it is a very cheap spell otherwise). It can be worth keeping in mind that Lifebloom interacts very nicely with Omen of Clarity--a Clearcast Lifebloom will still refund mana, giving you the spell and the mana for free.

Swiftmend: A strong instant heal on a short cooldown. One of our best spells. Always be vigilant for people at low HP on whom you might use this. It's great for helping stabilize a tank anytime you see them sit low for more than a GCD, or making sure any raid member is safe while your HoT's do their work. You can swiftmend another Druid's HoT's; if you're using the Glyph this doesn't interfere with them at all. Swiftmend will always work off of the lowest-duration Rejuvenation or Regrowth on the target. A Swiftmend off of Rejuvenation is stronger. Swiftmend uses your spellpower and talents at the time you cast it, but will be smaller if uses a downranked Rejuv.

Healing Touch: Very limited use. Combined with Nature's Swiftness, it provides an emergency instant heal, but this is barely stronger than Swiftmend. You'll usually use it with Swiftmend when you need two consecutive instant heals on someone, or when you need an instant heal on someone without a Swiftmendable HoT (example, someone who was just Rebirthed). Macro for this:
/showtooltip Nature's Swiftness
/stopcasting
/use Nature's Swiftness
/use [@mouseover] Healing Touch
(You can replace the @mouseover with whatever target you like). Note that this will cast both spells at once if you're still, but you'll need to press the macro twice if you're moving.
The only other use of this spell with and Naturalist to provide a heal with a very short cast time. Due to the 1 second GCD minimum, however, and the fact that Nourish has a 1 second cast time in typical current gear, this is rarely of much benefit. It's a gimmicky setup for when you need as fast a heal as possible to respond to some boss ability (and for some reason Swiftmend isn't a viable option).

Tranquility: This spell is unfortunately rather weak for the length of its cooldown. The issue is that channeling it for 8 seconds means you're not putting up your HoT's for those 8 seconds, so after the channel ends, you have no HoT's up and your problems may well have gotten worse. It's a desperate measure, used for trying to stave off a wipe for a few more seconds while finishing off a boss.

Rebirth: Our most unique contribution to the raid. The only important issue is to avoid wasting it. First, make sure to coordinate with other Druids in your raid using macros or Vent so two of you don't cast on the same target. Second, people love to accept the resurrection as soon as it appears and die to something immediately. It can be good to warn them if it's a bad time to accept. Here's a macro that casts Revive instead of Rebirth if you're out of combat, and also alerts your raid if Rebirth is used:
#showtooltip
/use [nocombat] Revive
/stopmacro [nocombat]
/use Rebirth
/ra Rebirth on %t

Innervate: Unglyphed, this returns 2.25 * 3496 = 7866 mana to the target.

Abolish Poison/Remove Curse: Nothing fancy to say about these. Remember that Remove Curse won't allow you cast it on a target without a cleansable debuff, but Abolish Poison will.

Tree of Life Form: This macro will instantaneously unshift and reshift you into form, breaking all roots and snares on you (taking one GCD):
/use !Tree of Life

Barkskin: Remember that this doesn't use the GCD, so you can cast it almost anytime without disrupting your healing. It should be on an easily-accessible bind, and you should make it second nature to hit this instantly when you foresee a threatening amount of damage coming.

=UI=

I'm not going to say too much about UI; it's largely a personal issue. The only major point specific to Resto Druids is that we're the only class that can move almost continuously while casting our normal spells. You want to have a control setup that allows you to be proficient at moving and casting independently--good Druids get very comfortable doing this. But it's a matter of practice more than anything else.

You need some kind of raid frames. Grid, Vuhdo, Healbot, and Shadowed are all in use currently. You can find all of them at typical addon sites. Choosing between them is up to you. Whichever you use, you want to have it set up to show you at least the following:
  • Each raid member's HP (including pets)
  • Which of your HoT's are on which targets
  • Their durations
  • Which raid members are in range
  • Which targets are Swiftmendable
  • Debuffs you can remove (Curses/Poisons)
  • Customizable debuffs for important boss abilities
  • Which players are being targeted by mobs
  • Which players have incoming heals from other healers
  • Notifications for when a player is resurrected, offline, or has the Spirit of Redemption buff (all important to make sure you use Rebirths wisely).

Clique: If your raid frames don't inherently support click-casting (Grid, for example), this is a simple, popular mod for setting it up.

Quartz: as a primary caster, you should have a proper cast bar. This is an excellent one. One Resto-specific tip is that since we use instants so much, you want to put the GCD spark somewhere prominent. I keep mine right next to Grid.

=TreeCalcs=

TreeCalcs is my Resto theorycraft spreadsheet, attached to this post. It will show you exactly how much various stats/talents/glyphs affect your current setup, and lets you experiment with different spell combinations. Compared to a DPS class, you're not going to be leaning as heavily on a spreadsheet to make gear choices, but it's useful to have a concrete, accurate numerical sense of how much each stat affects your various spells.

1)Input your stats and setup on the front page. In the three columns that go with each stat or buff, the first (left-most) is the one you edit, either by inputting a number or selecting an option from a dropdown menu. For spellpower/crit/haste/spirit/int, input the unbuffed value that appears on your character sheet or the Armory in your current gear. For MP5, add the total bonus MP5 from your gear only and type that in.
2) The main output is the green box at the bottom--the total raw healing output with your selected setup and spell rotation.
3) For more detailed information, look at the second and third columns mentioned above for each individual stat (the ones labelled "HPS" and "MP5"). These show the HPS/MP5 benefits derived from those particular stats/talents. For stats, it shows the benefit of having 1 more of that stat. For talents and other bonuses, it shows a) if you have no points in the talent, the value you would gain from one point, B) if you have points, the value you currently gain from those points (i.e. the amount you would lose by dropping them).
4) Since healing, unlike DPS, isn't quite as focused on the singular output from one rotation of spells, you often want to see how individual spells are affected by your stats. The second page lists all spells, the HPS/MP5 of each, and the amount which that HPS/HPM of each would increase from added points of spellpower/haste/crit. The top box on this page will also contain a bit more info on spellcasting cycles once I come up with them (ignore the leftover Moonkin stuff for the moment).

I've uploaded a second copy of the sheet without the tables. This will work in OpenOffice (and will run much faster generally), and everything will work besides the scaling factors. Might not do this every time, but will try to occasionally until I find a better solution.

Attached Files



#2 Rijndael

Rijndael

    Don Flamenco

  • Members
  • 300 posts

Posted 13 January 2010 - 09:34 PM

This is a somewhat expanded take on the glyphs a main spec restoration druid might consider using. I will consider the following glyphs: healing touch, innervate, lifebloom, nourish, rapid rejuvenation, rebirth, regrowth, rejuvenation, swiftmend, and wild growth. There is no best set of glyphs for all situations -- glyphs are meant to see situational use depending on the mechanics of the fight.

Special thanks to Arentios for writing the initial glyph guide and starting the discussion -- much of this information is from his post here.

Healing Touch.

With the Empowered Touch talent now affecting both Nourish and Healing Touch, on top of the multiple Nourish-boosting talents, Nourish really does have much better average case performance than glyphed Healing Touch (gHT). The advantages of gHT is that it can cast very quickly (0.7 second cast time with 0.5 second under heroism is not difficult to achieve), and has better worst case performance than Nourish (e.g. it heals for more than a non-crit Nourish on a target that's hot-less). The reason has to do with a slightly odd way spellpower coefficients interact with the glyph of healing touch (the precise formula can be found here).

Speed and worst case performance are situationally useful -- many druids use glyphed healing touch for healing penetrating cold on heroic Anub 25, for instance. Furthermore, gHT is fast enough that it can be used to heal tanks and heavy damage recipients reactively, minimizing overheal. Unfortunately, getting glyphed healing touch generally means taking 5 points in naturalist, which means you end up with this basic spec.

You must then choose whether you want to give up CF (lowers haste), nature's grace/living seed (lowers the value of crit), subtlety (raises your aggro which is important on some fights like Anub), or living spirit (lowers mana regen and throughput somewhat).

You also lose nature swiftness + healing touch combo.

The consensus among resto druids seems to be that healing touch is a situational glyph, used only in situations where direct heal sniping is important enough to justify the sacrifices the glyph entails.

Innervate.

Grants the druid 20% of the mana innervate grants whenever the druid uses innervate on any target. Currently resto druids are limited by GCD and throughput, not by mana. Thus, the consensus in the resto druid community has been to concentrate on glyphs that improve throughput not glyphs that improve longevity. Possible mana-stressing fights in Icecrown may change this consensus.

Lifebloom.

Adds 1 second duration to lifebloom. The value of this glyph depends heavily on the "rotation" used, or more precisely on the proportion of spells used in a given fight. Many druids healing raid damage on fights like Twin Valkyrs use Rejuvenation, and Wild Growth exclusively with an occasional Swiftmend. In such situations this glyph has no use.

However many fights are less well suited to a pro-active hot blanketing raid healing strategy druids use, some fights like Festergut alternate between heavy raid damage and heavy tank damage phases, and yet other fights feature heavy tank damage which make druid hot rolling on tanks an important contributor to tank stability, even if this involves losing some raid healing throughput on the part of the druid. In such situations, the lifebloom glyph is useful in proportion to the amount of lifeblooms cast.

In particular, if the majority of GCDs are spent on lifebloom, then the glyph adds about 10% to throughput. If the majority of the GCDs are spent on spells other than lifebloom, with a lifebloom stack or two maintained then the glyph adds 10% more lifeblooms cast per unit time, which translates to creating extra GCDs equal to 10% of the amount of GCDs used on lifebloom. If we assume for simplicity that a druid uses an 20 second rotation with 3 wild growth casts, 1 lifebloom stack, and rejuvs for the rest of the GCDs, then at haste capped 1 second GCD, the druid casts lifebloom a little over 2 times per rotation, which means the GCD gain from the glyph is 10% of (2+epsilon)/20 which is a little in excess of 1% throughput.

The glyph doesn't save mana per se, since any time gained by casting less lifeblooms will be used to cast other spells.

Since in the vast majority of fights resto druids don't spend most of their GCDs on lifebloom, the HPS gain is generally too small for wide use of this glyph.

Nourish.

Best tank healing glyph for resto druids, this turns Nourish into an HPS monster on any target with full hots. Most raiding guilds use paladins for healing tanks due to the power of Beacon of Light, with druid tank healing restricted to rolling hots for stability. Nevertheless, this glyph is powerful enough where some druids run with it for the strategic value of being able to shift to tank healing in the middle of a fight. Some fights in Icecrown such as Festergut or Putricide create tank pressure in some phases, which may give druids a reason to run this glyph for those fights. This glyph is also useful for new druids with poor gear healing 5 mans, since their hots alone will be insufficient for keeping up tanks, and for even well-geared druids in 10 man raids where any healer can be called upon to help with any healing role.

This glyph is practically mandatory for the Valithria Dreamwalker encounter.

Rapid Rejuvenation.

The new glyph of patch 3.3, this makes haste rating lower the time between rejuvenation ticks (without adding new ticks). This leaves healing per cast time of rejuvenation unchanged while increasing healing per second. This glyph is similar to the priest glyph of renew in this way. The value of this glyph depends on the damage pattern of the fight. Some fights, such as Twin Valkyrs, have raid wide aura damage which favors Rejuv/Wild Growth spamming. In such fights, what determines druid throughput is healing per cast time, which makes this glyph not particularly useful. Moreover, if the DPS from the aura is less than the HPS of Rejuvenation modified by the glyph then the glyph can even decrease effective healing throughput and increase overhealing (since the HPS gain from the glyph on one target isn't translating into effective healing due to lack of damage).

On the other hand, some fights feature damage that is more focused on tanks, and specific targets (for instance Rotface's infection, or Saurfang's mark). In such fights druids don't spam as much, and use more targetted healing instead, so healing per second becomes a more important characteristic. In such fights the glyph may be very valuable. Finally on some fights (Anub) you may want to restrict the amount of HPS you do. Such fights also aren't a good time to bring out this glyph. The glyph becomes very powerful in 5 man or 10 man content since it allows druids to blanket the entire group/raid with faster ticking Rejuvenations with little sacrifice of coverage.

This glyph increases the power of the Revitalize talent on the targets you do rejuvenate (although not on the raid as a whole). If you want to specifically take advantage of the dps gains of Revitalize by concentrating Rejuvenations on specific targets that gain a lot from Revitalize such as death knights, this glyph may be helpful.

Rebirth.

A situational glyph, it brings a measure of raid stability (since a dead target wil come back at full health) at the expense of throughput you give up by not slotting a throughput glyph. May be a handy glyph for the few fights where deaths are planned for and raid damage is spiky (Iron Council hard in Ulduar for instance).

Regrowth.

Adds 20% to any regrowth which was recast on a target already affected by Regrowth. Currently Regrowth rarely accounts for more than 10% of resto druid healing, which would put the gains from this glyph at 2% at best (and likely quite a bit less). With the new version of the Gift of the Earth Mother talent, and the increased emphasis on haste, Regrowth with its many supporting talents and a long hot may well become a mainstay raid healing spell for some druids. If Regrowth becomes competitive with Rejuv, this glyph would become mandatory for Regrowth users.

I should say that one disadvantage of a Regrowth-heavy healing strategy is lessening the value of Revitalize, which procs off Rejuvenation. Revitalize is a very powerful mana regeneration and dps boosting talent for the raid.

Rejuvenation.

Adds a tick valued at 50% of a normal Rejuv tick to a target which was at less than 50% HP at the time Rejuv ticked. The small tick appears with a small time lag, and (as of last testing) only if the normal tick does not bring the target over the 50% HP mark. Empirically, the gains from this glyph are very modest, rarely exceeding 2%. Ticking aura fights where many raid members are expected to reach low health are ideal times to use this glyph. This glyph interacts well with the glyph of Rapid Rejuvenation in fights with burst damage on a few telegraphed targets. Both glyphs add HPS to Rejuvenation ticking on a target of burst damage. Rapid Rejuvenation glyph makes the ticks faster (more likely to save the target of burst), and Rejuvenation glyph makes the tick bigger (bigger tick -- less probability of death).

Swiftmend.

Prevents Swiftmend from eating a hot when it's cast. This glyph is similar to the lifebloom glyph in the sense that it saves you GCDs since you have to recast less hots. The value of this glyph isn't straightforward to compute -- it essentially saves you whatever healing the remaining hot would have done had it not been eaten by Swiftmend. If hots are eaten with half their duration on average, and if Swiftmend heals for (approximately) the full healing value of a hot, then a rough approximation to the value of this glyph is about a third to a half of healing Swiftmend did for the encounter (assuming Swiftmend and the hots both have approximately similar overheal). This glyph is also very helpful for tank healing druids as it eases the GCD lock burden tank healing druids have (since they have to recast hots on the tank and simultaneously quickly react to burst events on the tank).

Wild Growth.

Wild Growth tends to account for 15-25% of healing done for many druids on most fights. Theoretically, this glyph adds 20% (by adding a 6th target) to this spell. In practice, the gains from the glyph are smaller, for two reasons. The first is that on some fights, the raid is spread out enough that Wild Growth simply does not hit the glyphed 6 target maximum. Furthermore, even if Wild Growth does hit 6 targets, frequently some of these targets are pets. While healing pets is generally a useful thing to do, these days pets are very resistant to random AoE damage, and often have self-healing options. Having the 6th target be a pet is certainly less useful than having the 6th target be a player. A more detailed discussion of the practical performance of this glyph, with data, can be found in the sequences of posts after this one.

Despite these issues, the glyph is a solid choice, and widely used. It is especially useful on situations where druids find themselves spamming Rejuvenations with Wild Growth on cooldown (the so called 5+1 rotation), or in cases where the strategy calls for the raid to clump up. This glyph increases the power of the Revitalize talent. Further, Wild Growth is sometimes used as a way to quickly power up Nourish by 20% on multiple targets. The glyph obviously helps this role of Wild Growth.

#3 Norfair

Norfair

    Piston Honda

  • Members
  • 246 posts

Posted 17 January 2010 - 08:30 PM

I have been reconsidering Glyph of Swiftmend. Although having it is immediately noticeable as you're HoTs won't be removed anymore, I have been thinking: how much does this glyph really help? I don't cast Swiftmend that often in order to justify having it, except for a fight like Faction Champions where I use it almost every cooldown.

Glyph of Lifebloom also saves you a gcd, but that's every 10 seconds on every fight where you roll Lifebloom (and frankly, in ICC10 I have started doing it again). It's probably even more than a 11% increase as you won't recast LB at the very last second but more like when it's 1-2 seconds remaining. So recasting it after 8 seconds instead of 7 is a 14% increase in gcds.

The only thing I'm wondering right now is if the Swiftmend glyph is better than Glyph of Rejuvenation, for encounters where Glyph of Rapid Rejuvenation is not helpful (or is even counterproductive).

#4 Arentios

Arentios

    Wisdom as dump stat

  •  Patrons
  • 6,796 posts

Posted 17 January 2010 - 09:54 PM

I've been experimenting with Glyph of Rejuvenation lately, and haven't gotten very stellar performance. On our Putricide this week it accounted for 1.2% of effective healing, on Heroic Twins it was 0.6%, and on Festergut it only proc'd a total of twice. World of Logs - Real Time Raid Analysis is the parse from that night. Those are the three fights currently available where I'd say you'd get best performance. There's still a lack of truly healing intensive fights, where the Glyph of Rejuvenation would probably scale better. Even on Putricide and Festergut we're rarely seeing people dip below 50%, and when they do they rarely stay there for more than one tic of Rejuvenation.

Of course, the Glyph of Rejuvenation is also healing on a target who presumably really needs healing, so the healing is likely more valuable than the numbers immediately suggest. However, the same can be said for the Glyph of Swiftmend. For glyphing for specific fights, there's some overlap between the two, but I can see swapping between the two on various bosses. Any fight where you're using a blanket Rejuvenation approach you aren't likely to be using Swiftmend as much as on a spot healing style encounter.

#5 Hamlet

Hamlet

    Mike Tyson

  • • Guide Author
  • 11,540 posts

Posted 17 January 2010 - 10:04 PM

To be honest, we could heal most fights with no Glyphs at all and not feel much noticeable ill effect. They just don't do anything significant (Wild Growth might be the only one that stands out somewhat). Rapid Rejuvenation is the only one that has a serious meaningful impact on the spell we use the majority of the time, but it's not even a change that's desirable at most fights. Nourish is also "good" but only in a situation we're almost never in.

I think that overall people just wind up using Wild Growth because it has a concrete measurable effect, Swiftmend out of pure convenience, and whatever third one they feel like. They all provide minor and sporadic benefits.

#6 Rijndael

Rijndael

    Don Flamenco

  • Members
  • 300 posts

Posted 17 January 2010 - 10:21 PM

This is a "theoretical" post, trying to address the following questions: "what is the value of a heal?" and "which heals are more valuable than others?"

Heals are there to prevent death. If healers could see into the future, they would always land a heal before a fatal blow lands, as long as damage is healable. We can't see into the future, but we can summarize our ignorance with "probability of death" (P(death)) -- how likely someone is to die given the damage pattern of a fight. We don't know this quantity exactly either, but we can say some things about it: the lower someone's hp is, the higher their probability of death, because more blows will kill the lower your hp is. So, P(death|hp1) > P(death|hp2) if hp1 < hp2.

For each "state" of the raid, defined by the current hp of each raid member, we can define the "instability" of the raid as the sum of the death probabilities of each member conditioned on their current hp. The job of healers is to decrease "raid instability" by increasing raid hp, and thus lowering the sum of death probabilities (ideally to as close to 0 as possible).

I propose the real value of a heal is the amount by which it decreases total raid instability. Some healers, like druids, work by lowering the death probabilities a little bit on lots of targets with hots. Others, like priests and shamans, often work by decreasing death probability a lot on few targets via burst heals.

So far this is all fairly obvious, so what's the point of writing this? Well generally healing someone for some amount at low hp is "worth more" than healing someone at high hp for that same amount. This is intuitively obvious, but the reason this is true is because you decrease the death probability more by healing someone at low hp.

If you accept this notion of "heal value," some things follow:

(a) Healing meters don't measure the true value of your heals. They can't because relevant probabilities are not possible to estimate accurately. An approximation to the true value is to weigh heals more and more the lower the current hp of the target they heal is (the exact weighing is encounter dependent and subject to debate). If I were to plot death probability vs current hp as a curve, I would expect the probability to start at values close to 0 and slowly rise as hp falls, and at some point sharply peak towards 1 as hp reaches 0.

(B) If you want to measure glyph or heal (or healer!) value, you can't just compare throughput numbers, you have to think about whether the glyph or heal (or healer!) is likely to heal someone at low or high hp. For this reason, the glyph of rejuvenation is probably worth quite a bit more than the healing meters would indicate, while resto druids are probably worth less than their meter positions indicate.

#7 Carebare

Carebare

    ::stare::

  • Moderators
  • 5,200 posts

Posted 17 January 2010 - 10:42 PM

I really do think some of it comes down to healing style. I find the mend glyph irreplaceable on 95% of content. I play on a Boston-based server from NJ with an average ping of 40ms (provided Mal'ganis isn't choking like it's prone to do). This means I will land my mend for full value a majority of the time, and the RJ staying on to tick after will probably heal also. I use mend very aggressively, so I would be doing a lot of reapplying of RJ if I removed that glyph. Some folks may favor the spell less, either due to style or simply ping in which case it's not nearly as valuable to them.

i miss raiding with carebare :< she makes me feel like i am not the only person that hates everyone
Aldriana: I am an asshole, it just so happens that some of my colleagues are even *bigger* assholes.
[R] [85:Neux:2]: i hear if you die on Good Friday they are going to make it where you can't get rezzed until easter sunday
Khazal: Yeah, I don't know about Magic Rainbow Unicorn Land, but here in Reality, Rhyolith is the worst encounter Blizzard has ever designed.


#8 Arentios

Arentios

    Wisdom as dump stat

  •  Patrons
  • 6,796 posts

Posted 17 January 2010 - 11:02 PM

The simplest example I can think of to demonstrate relative healing value (I've generally called it non-critical versus critical healing, but I like the idea of a probability of death) is tank healing versus raid healing.

Numerous cases have shown that outside of gimmicky fights like Anub'arak, or fights with negligible raid damage, a healer focusing solely on tank healing generally won't put out the effective HPS of a raid healer. Yet nobody in their right mind would suggest dropping the tank healers for more raid healers. Healing is incredibly dissimilar to DPS in that regard, where it is generally the more gimmicky fights where DPS has to be carefully broken down by type.

With the Glyph of Rejuv, I don't see it lowering P(death) very much. Looking at what we've seen from ICC, most damage that will kill someone from half health (or lower) will also kill them from half health + 0.5 Rejuv tics. Maybe if Sindragosa's aura is really brutal on heroic, or the Lich King has some nasty sustained AoE that will be different. In any given spike case I don't imagine seeing more than one glyph proc is likely before the target dies or is healed to full, unless you're also running Rapd Rejuvenation with a high amount of haste. I'm going to agree with Hamlet and Carebare that it's largely a moot or style based decision.

#9 GTtheBard

GTtheBard

    Von Kaiser

  • Members
  • 77 posts

Posted 17 January 2010 - 11:03 PM

One thing that frequently gets lost in the countless discussions on the Swiftmend glyph is whether you run with other Resto Druids. Removing another healer's Rejuv/Regrowth can be devastating if they are counting on that HOT to be there (either for their own Swiftmend, to buff Nourish, whatever).

As Hamlet said, all the glyphs provide minor bonuses; given the option to keep our HOTs on a target after a Swiftmend, why would we pass that up? Especially when the other options provide similarly minor choices and are certainly less likely to be "harmful" to a raid if they are NOT there.

At the end of the day though, Resto glyphs are a choice. There is nothing that says "This glyph is best in slot" as healing strategies change constantly to deal with the current situation. If you're the sole Resto Druid in a raid and have no problem removing your own hots from a target, then feel free to use something besides Swiftmend. But I do think it's safe to assume that most Resto Druids will use Swiftmend, simply because it's a nice option to have; if you run with another Resto Druid and don't want to use the Swiftmend glyph, at least let them know so they aren't surprised by a premature HOT removal.

#10 Rijndael

Rijndael

    Don Flamenco

  • Members
  • 300 posts

Posted 17 January 2010 - 11:16 PM

With the Glyph of Rejuv, I don't see it lowering P(death) very much.


I just meant that people evaluate the Rejuv glyph as follows:

They look at the meter, see that it does 0.5-2% of healing, and say "wow this is not a lot of throughput, I ll go glyph something else." In reality we know something about this 0.5-2% -- it always healed someone at less than 50% hp. That means the value of this 0.5-2% (in terms of P(death|hp)) is a quite a bit higher than the value of your average hot tick heal (making up perhaps 90-95% of your meter). If meters had some weighing scheme for heals perhaps that 0.5-2% would display as 2-8%.

The glyph tick doesn't lower P(death) very much, but it lowers it more than an average tick for the same amount. This is all similar to how tanks value effective health -- it's all about improving survival probability, so even relatively minor amounts of block value/armor/max hp help in a "worst case situation."

edit: vaguely related amusing comment, I had a fight where the rejuv glyph did 10% effective healing the other day (ic 10 Stinky) :).

#11 Fallenangel

Fallenangel

    Don Flamenco

  • Members
  • 485 posts

Posted 18 January 2010 - 12:06 AM

I have been reconsidering Glyph of Swiftmend. Although having it is immediately noticeable as you're HoTs won't be removed anymore, I have been thinking: how much does this glyph really help? I don't cast Swiftmend that often in order to justify having it, except for a fight like Faction Champions where I use it almost every cooldown.

Glyph of Lifebloom also saves you a gcd, but that's every 10 seconds on every fight where you roll Lifebloom (and frankly, in ICC10 I have started doing it again). It's probably even more than a 11% increase as you won't recast LB at the very last second but more like when it's 1-2 seconds remaining. So recasting it after 8 seconds instead of 7 is a 14% increase in gcds.

The only thing I'm wondering right now is if the Swiftmend glyph is better than Glyph of Rejuvenation, for encounters where Glyph of Rapid Rejuvenation is not helpful (or is even counterproductive).


That's really not how it works - a glyph that saves you a GCD every 10 seconds would be amazing.
With the glyph, you have to cast LB 9 times every 90 seconds and without it 10 times. So it saves you only 1 GCD every 90 seconds - a bit more if you refresh ahead of time and a lot more if you roll on more than 1 tank.
The mana saving is about 40mp5 for each tank, so it's pretty on par with the innervate glyph if you roll on 1.

#12 Rijndael

Rijndael

    Don Flamenco

  • Members
  • 300 posts

Posted 18 January 2010 - 01:14 AM

The mana saving is about 40mp5 for each tank, so it's pretty on par with the innervate glyph if you roll on 1.


I don't think the lifebloom glyph saves you any mana given the way druids cast in practice -- you spend less mana on lifebloom but then you just use those freed GCDs to cast other spells, so at the end of the fight you don't end up with any more mana than before (unlike the innervate glyph which actually gives you something regardless of how you cast). The glyph just lets you put out more spells per unit time.

#13 Paininabox

Paininabox

    Piston Honda

  • Members
  • 137 posts

Posted 18 January 2010 - 05:16 AM

I have lately been considering the worth of a heal much along the same lines as you, Rijndael. My line of thinking stems from how unsatisfactory the healing meters tend to be when describing healing performance. I theorized that healing meters are partly accurate in that healing volume is important, but, much like you say, it is also important to heal those with lower %'s of health.

I attempted to fuse volume and the "who needs healing most" factors by arbitrarily picking the function of 1/x and finding its antiderivative over the interval [A,B], which comes out as M = ln(B/A) and then multiplying by the effective healing amount. The final form, then, is HP = ln(B/A) * H, where B is the after-heal health % value and A is the pre-heal health % value and H is the effective healing amount. The theoretical result is that heals delivered when the target would otherwise have died is worth much more than heals delivered on targets that have cushion room. It also penalizes raid healers, and in particular druid hots, because they happen in small chunks and therefore the natural log part of the equation would be small, and it rewards tank healers since tanks tend to yo-yo between low and high health %'s.

Since the choice of 1/x as the base derivation function is arbitrary, I expect that it might not be a good choice. As a theoretical way of taking into account the need to triage appropriately, I think it's reasonable to say that it's a step ahead of traditional effective healing meter totals.
Level 80 Restoration Druid Spreadsheet here.
(v1.41)

#14 Rijndael

Rijndael

    Don Flamenco

  • Members
  • 300 posts

Posted 18 January 2010 - 06:18 AM

Since the choice of 1/x as the base derivation function is arbitrary, I expect that it might not be a good choice. As a theoretical way of taking into account the need to triage appropriately, I think it's reasonable to say that it's a step ahead of traditional effective healing meter totals.


1/x has the right sort of shape -- the death probability curve starts off shallow and close to 0, and spikes towards higher values at some encounter dependent point. If we really wanted, we could collect raid death statistics with some addon and get an empirical plot of the death probability curves for various raid zones, and then use that.

#15 Lightflower

Lightflower

    Von Kaiser

  • Members
  • 97 posts

Posted 18 January 2010 - 06:24 AM

The counterpart to that line of thinking is then to factor in human reaction time and cast speed:

1. Smart heals are more likely to land on a person who will maximise the value of Paininabox' equation
1a. Smart heals with raid wide range are more likely to maximise critical healing than their more limited counterparts (cf. Abacus vs Chain heal bounces)
2. Instant heals integrate over the shortest possible timeframe again providing maximum benefit.
3. Blanket hots, while not being "smart", do fall under #2 and thus provide their packet of healing where it is most needed - even if it is a shotgun strategy.

Druids are like a person attempting to fill an array of tiny cups with a bucket of water. While it's efficient to pour water into each cup individually, if you don't care how much water you waste then it's just easier and quicker to get a bigger bucket and throw it over all the cups.

Ideally, healing meters would enable a person to set a threshold (or thresholds) for bracketing effective healing. Sub 30%, 30-70% and 70%+. Effective healing in each category could then be weighted according to the type of fight and the raid role of the recipient. This could lead to heal mapping where someone might be congratulated for dropping 8% of their EHPS onto the 3 highest dpsers in the sub 30% bracket, thus ensuring that the raid beat the enrage timer.

Alternatively, one may be criticised for having minimal heals in the sub 30% bracket but could make the case that an enormous contribution to the 70%+ bracket meant that fewer heals were necessary sub 30%.

#16 Deku

Deku

    Glass Joe

  • Members
  • 10 posts

Posted 18 January 2010 - 08:30 AM

Thought I'd link a WOL from our 1 shot of PP last night. (seeing as the old resto pve thread is closed I'm assuming this is the new one to post stuff like this in)

World of Logs - Real Time Raid Analysis

As you can see the benefits from Revitalize are quite amazing, considering this was only from 1 attempt and all we did was make sure it had a rejuve active at all times. Our Abo controller had so much energy compared to last week when we didnt spec for it.

Considering the slowing debuff costs 50 energy, gaining 504 just from Revitalize is quite significant.

#17 Carebare

Carebare

    ::stare::

  • Moderators
  • 5,200 posts

Posted 18 January 2010 - 08:40 AM

I think it really comes down the roles each healer fills. This varies by raid but for the purposes of our conversation (and based on my opinion/experience naturally) I see healing falling into categories:

1) Proactive Offensive Healing/Shielding - This is where your paladins and disc priests fall. Their job is pretty straightforward, provide steady HPS (or damage reduction by way of absorbs) to the tank(s) throughout the encounter. Once upon a time back when LB stacking was viable, we were here.
2) Reactive Offensive Healing - This is where your resto shamans and even holy priests would fall. When the damage hits they provide the burst to bring people out of the danger area. Prior to WG cooldown and even after with 4t8, we were here.

So where does this leave us now?

You could argue that our presence on the healing meters is a result of sniping and/or just spamming and catching effective healing. I would respectfully disagree (for the most part), but again I feel like this comes down to style. Most end-tier raiding druids are playing the versatile middle ground which I've always referred to as "Proactive Defensive Healing". It means that a strong resto druid is going to see the holes the other healers can't cover and fill them. The other healers aren't bad, this is a naturally occurring hole that our tools give us the opportunity to fill. To do this the idea is simple, yet harder to execute. You need to know the damage that is going to happen before it occurs. Your HoTs need to be in place to meet that damage as it happens. I feel like this is a major failing point for new druids.

Our healing prevents that sub 30% bracket if said druid has a definitive understanding of the fight mechanics. I won't disagree that some sniping is inevitable with this style. However, our job is to mitigate damage and keep people out of that danger area and doing that results in a fairly aggressive healing style when done correctly. This is the reason you will see some druids top meters regardless, even if the fight is not a Val'kyr Twins type scenario. They see the damage before it happens and have already provided the proactive defense to meet it. So, that when it happens, those same druids are pulling out tools like Swiftmend/Nourish as opposed to casting Rejuvenation/Regrowth (because those people that took damage had HoTs on them before it even hit).

It is less tangible in terms of "life-saving" measure compared to what pretty much every other class does. But to call it less valuable seems unfair to me. I will concede that it is not hard to be a mediocre or even "good" druid. Being an exceptional druid requires anticipation, awareness, and a very strong fundamental understanding of the proper tool for the job, both before and after the damage has occurred.

i miss raiding with carebare :< she makes me feel like i am not the only person that hates everyone
Aldriana: I am an asshole, it just so happens that some of my colleagues are even *bigger* assholes.
[R] [85:Neux:2]: i hear if you die on Good Friday they are going to make it where you can't get rezzed until easter sunday
Khazal: Yeah, I don't know about Magic Rainbow Unicorn Land, but here in Reality, Rhyolith is the worst encounter Blizzard has ever designed.


#18 Fallenangel

Fallenangel

    Don Flamenco

  • Members
  • 485 posts

Posted 18 January 2010 - 09:47 AM

To be honest, we could heal most fights with no Glyphs at all and not feel much noticeable ill effect. They just don't do anything significant (Wild Growth might be the only one that stands out somewhat). Rapid Rejuvenation is the only one that has a serious meaningful impact on the spell we use the majority of the time, but it's not even a change that's desirable at most fights. Nourish is also "good" but only in a situation we're almost never in.

I think that overall people just wind up using Wild Growth because it has a concrete measurable effect, Swiftmend out of pure convenience, and whatever third one they feel like. They all provide minor and sporadic benefits.


Agree with what you said but not with the conclusion. I see RRejuv glyph as a must have and it is desirable on most fights. The only reason you won't want it is maybe (and that is also arguable) is aura fights with one 1 resto druid on 25 man - allow me to ignore gimmicks like Anub P3 if you will. In other cases - where you have less targets than time - powering up our rejuv by 50% has a noticeable effect.
Also, LB rolling is still viable and has been viable for the entire of WotLK. It's still decent HPS, it's still a nice buffer and it will do 15-20% healing on fights where damage is focused on the tanks - Gunboat and Saurfang for instance.

#19 Erdluf

Erdluf

    Great Tiger

  • Members
  • 960 posts

Posted 18 January 2010 - 01:51 PM

Rij, Pain,

I think you're on the right track. However the formulation is a bit too simplistic.

Suppose I'm the only healer, and Bob who has 20k maximum health is taking aura damage at 1k/s throughout the fight.

I can keep Bob at nearly full health (never more than 3k damage) by rolling unglyphed Rj. Alternatively, I can cast Nourish every time Bob's health drops to 18k below maximum.

I really don't like metrics that would say the Nourish strategy is more "effective" than the Rejuv strategy. A 5k effective heal just before someone takes a big hit is at least as useful as a 5k heal right after a big hit.

Here is one algorithm I've used to identify "significant" heals from a combat log:

1) Identify a danger threshold (perhaps -10k for a raider, or -20k for a tank, depending on the instance). In some cases you might decide to use zero as your danger threshold.
2) Look at all damage and heals (including overheals) that hit the toon during a particular fight. This will give you a graph of health-deficit as a function of time.

The toon's "danger factor" is the "area under the curve" between the health deficit and the danger threshhold:

integral (over time) of max(health_deficit - danger_threshold, 0)

3) Remove a single healing cast from the log (perhaps a FL or an entire Rj, not just a single HoT tick). The heal was "significant" to the extent that removing the heal increased the toon's "danger factor."

Note that a very effective heal (say NS+HT, with no over heal) might not be all that significant The log might also contain a Swiftmend that landed 1 ms later, and was entirely over heal. Had the NS+HT not landed, the toon still would (probably) have been safe.

Clearly this could be automated. When I've done this type of analysis I've used a spreadsheet, or just estimated the values using a graph that shows incoming hits, heals, and current health deficits. WowLogParser, an offline parser, has such a graph.

#20 mhenrique85

mhenrique85

    Von Kaiser

  • Members
  • 29 posts

Posted 18 January 2010 - 03:42 PM

In my opinion, as a 10 man strict raider, Glyph of Rapid Rejuvenation is a must have glyph on 10 man raid environment, meanly at ICC.

The only fight that i thought it would be counterproductive was Professor Putricide, because of Malleable Ooze debuff, i tried a couple of tries to run without this glyph when we were progressing trough the fight, and it was there that i really felt the difference, then I reglyphed it, and the heal became a lot smoother. Malleable Ooze debuff can be easily avoidable and the difference between how fast you heal heavy AOE damage with this glyph is evident.

On 10 man raid environment, you have a cap of how many players you can heal with your Rejuvenation, even with your Rejuvenation ticking between 2 secs, with 12 secs duration, you can keep up Rejuvenation on most players in the raid. And you just have 2 healers in the raid, the necessity to heal faster is evident.

For 25 man raids, i really think that depends on your healing style and raid healers setup. With the glyph, you heal faster, less overheal. Without it, More Blank Rejuvenations in the raid.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users