Kortar

The reason I choose random WWS parses is that the site doesn't allow processing the data en masse in the ways I'd like to demonstrate. The fact that I can make an assertion and pick a random WWS that demonstrates the assertion strongly indicates that there's a good chance if you did a wider survey of WWS parses, you'd find the assertion was generally true.

Inarguably, it's far from conclusive proof, but in the absence of the WWS designers redesigning their site, it's about all I can offer people who don't understand the principle I'm trying to present.

Check your own WWS if you're willing to take a hard look at the way you heal. In all likelihood, you're overhealing a lot more when you use Greater Heal than when you use Flash Heal, and that ends up bridging the difference between efficiency/throughput of the two spells. The reason this occurs is that you're "racing" against other heals caster by other healers - and those other heals are more responsive than Greater Heal.

To convert mp5 <-> Spellpower, the easiest way is to observe that 1 point of Spellpower is approximately equal to 0.4 mp5 via Spirit @ 80% FSR w/ 3/3 Meditation, BoK and SoR. So let's run down the talents:

Blessed Resilience - Since this is about PvE uses, I think we can safely assume no one is considering taking this talent.

Test of Faith - This is a tough one to model. Since CoH is an auto-targetting heal, it's very likely to benefit from this more often than most spells. However, I don't have very good data on how often this will be useful in a raid setting - we really need a comprehensive set of WWS that we can re-parse to reveal the utility of Test of Faith. My gut instinct tells me that it won't actually occur all that much. The only time people drop very low on life is when they get unexpectedly hit very, very hard - in which case there's a good likely you'll be using that SoL Flash Heal you're almost certain to have sitting around (and thus lose half the benefit). In its original form, this talent may well have been good but I just can't see it winning out over the rest of this list in its current form.

Healing Prayers - This will generate approximately 20 mp5/point if you're casting Prayer of Mending every 15 seconds. This benefit is largely static, and due to the existance of Glyph'd Holy Nova and Inner Focus, we can generally disregard its impact on Prayer of Healing. Itemization can convert this to 50 Spellpower/point.

Spiritual Guidance - At 1000 Spirit, this generates 50 spellpower (itemization = 20 mp5).

MA - We'll reap about 4 mp5/point from our Prayer of Mending (once every 15s). To make up the other 16 mp5/point to reach HP/SG level, we'll need to cast a Circle of Healing every 5 seconds or more (over the duration of the fight).

Spirit of Redemption - This provides an additional 50 Spirit @ 1000 Spirit - very close to the +50 Spellpower/+20 mp5 from the talents above if you disregard the up-to-12.5 spellpower you're getting (in which case, it's a bit better).

Spiritual Healing - This provides about +85 spellpower/point @ 2500 spellpower.

Divine Providence - This provides about +85 spellpower/point @ 2500 spellpower.

Serendipity/EH/HC/IHC - None of these apply to CoH. I'll handwave these away by pointing out that numbers have already been presented for them, and they're all more powerful than the SG/HP/SoR-level talents above.

Surge of Light - At 20% spell critical, this provides a bonus Flash Heal about 47% of CoH casts. This translates into 293 mana per cast if you're using those Flash Heals. So even if you're only casting 1 CoH every 5 seconds and only casting 1 out of every 6 free Flash Heals, you're still getting a greater benefit/point than virtually any other talent on this list.

Circle of Healing - All the above math is somewhat meaningless unless you spec this talent.

Spiritual Guardian - One model for this is the "cast it and go get a drink" model where you get 10s worth of 0% FSR. With our 1k stats, this is about 34 mp5/point. Theoretically you could also use it for its intended purpose.

Lightwell - I can't think of any model for this that wouldn't make it look stupidly good compared to every other talent Priests get, since they all involve glossing over the technical difficulties of using the spell. Either you like it or you don't - that's a discussion for another time.

Due to the way the tree is structured, we have to take 5 points in the three tier 5 talents (HP, SG, SoR). These are also some of the only talents we'd really consider sacrificing for MA. Let's also assert that Lightwell is still junk and skip it. We can skip Test of Faith.

At this point we've got enough for 3/5 MA. But we've thrown away one point to get there (I like to put that one point in Martyrdom to show my contempt for the 15th point in the Discipline tree , and we're now at the point where the only talents we can lose that aren't directly point-for-point better than MA are the ones that exclusively buff our direct heals (and do it a lot better than MA buffs CoH).

Considering that we've only gotten 75% of the normal benefit from MA (4 points for 3 point value), that means we'd need to be regularly engaging in fights where we're casting CoH on average every 3.6s or so. But with our 1k stat base, we'll be out of mana casting a CoH every 3.6s well before the 3 minute mark (not even counting that a certain portion of the value of MA comes from that Prayer of Mending we're flinging every 15s as well).

If you've followed me thusfar, I'd suggest that 14/57/0 is the 'cookie cutter' for Holy. The Lightwell question mark remains and the 5 points from third tier can really be placed however you want. For the remainder, I'd suggest that this is the most talent point-efficient way to spec presuming you're going to use a mixture of spells.

 

Bjork

Kortar, what beta experience do you have? Because not specing Test of Faith and calling it a cookie cutter-build seems so wrong that I don't have words for it. Have we seen the same encounters?

 

Shylena

I do not know how it will play out in WoTLK, but I use a technique/strategy in raids to maximize the value of CoH priests.

Take for example the ZA Hexlord encounter, which I typically do with 2 CoH healers. I put one in each of the two groups, and make sure the group members are aware of the strategy. As for shadow resist the BT fire resist cloak and neck, along with the shadow protection buff is plenty and does not gimp you.

When we get to the phase involving the AOE damage, each of the priests puts PoM on someone in their group as it starts, and then uses PoH and CoH to easily heal thru the damage bursts.

This strategy works very well on Bloodboil and Na'jentus in BT as well. The groups are centered around a CoH priest, and using PoM, CoH and PoH, altogether just overpower the damage very well.

I can get max PoM bounces on the BB and Hexlord encounters every time.

 

Sinndir

You are suggesting that you call our 'cookie-cutter' with Test of Faith?

Each time I spec, I start with the mandatory 14 points into Disc, then head to the Holy tree and end up for this...

Talent Calculator - World of Warcraft

This gives 2 free points which could be used in healing prayers. Also it is possible to remove one point from Desperate Prayer and move it to Holy Reach (I've always been a Night Elf so no idea if I'll actually want DP).

I just don't see Test of Faith being a must have, I think it is a unique talent and depending how well a few things pan out for us once Wrath goes live, I could se a spec something as this.

I think us grabbing Mental Agility is a thing of the past, at least 5/5 that's for sure.

 

bbartlog

A general comment that has little to do with specific talents and more to do with scaling of different stats at 80:
It seems to me that one effect of the changes is that priests (and probably healers in general) will have to spend more time casting to achieve the same overall healing (measured in tank life bars), and that some talents have been added to allow them to do this. I don't know whether this is a deliberate design choice by Blizzard, but its seems fairly likely.

A rough outline of the assumptions and reasoning:
- overall stat points will double from 70 to 80
- mana pools and life bars, for characters that are fairly geared (say halfway through 10-man Naxx or w/e) will be big enough that the base mana / base life is not too significant - so mana pools and life bars will also be just about 2x what we have in TBC
- the base healing of priest spells doesn't double (unless you count PW:S but I think this has been deliberately buffed vis-a-vis the other spells). Greater Heal and Renew, for example, are nowhere near double their numerical effect at 70 (more like 40% increased).
- on the other hand, the cost of the healing spells, also based on a stat (base mana) that does not double, is such that more can be cast before a healer goes OOM

The net effect for a hypothetical priest similar to my own Kara-level toon (9K mana, +1500 heals, scaled to 80) is that
- before talents are considered, I will be able to cast about 25% more heals (numerically) before I go OOM
- however, measured in tank life bars, the healing I can do before I go out of mana will be almost the same, because
- my heals will hit for a correspondingly smaller fraction of the tank's total health
- if a fight requires the same overall level of healing (measured in tank life bars) as a similar TBC fight, I will have to spend about 25% more time actually casting

Some additional implications of this are that
- a considerably higher relative level of gearing will be required before healers want to stack spell haste in preference to +healing (I can outline the reasoning if someone doesn't see why this is)
- healers who want to get out of the 5-second rule will need to pay even more attention to bunching up their spells. In a regime where my greater heal only heals 1/4 of my tank's life (but, we assume, damage is still not overwhelmingly more than my mana pool can handle), it will make sense to stop-cast until the tank is low on life and then chain-cast four greater heals in a row or something
- burst healing will be gimped, assuming we measure it in terms of how many life bars we can heal in ten seconds

And a final thought: if Blizzard actually succeeds in 'making mana matter again' (so far, from what I've heard, they have not succeeded in doing this - quite the reverse), this will make Darkmoon Card: Blue Dragon even better than it is now. 25% more spellcasts to proc, plus mana-efficiency talents to compensate for the lost time OO5SR, will make it a must-have for spirit priests. Until it gets nerfed :-)

Apologies if this has been covered in more detail elsewhere. I searched for posts on priest scaling but didn't find anything along these lines.

 

gia

A problem I see with your reasoning is that you're not just filling life bars, you're healing the actual damage that gets put out. So what you want to compare is the amount of damage coming in relative to the amount of healing you can put out. People having more HP is actually helping you heal more reactively by giving you a larger buffer.

With no more downranking and players having relatively more health the approach to healing can be very different. I am absolutely not chaincasting in nax 10 except in situations where I think the tank might be in danger even when at full health and that is pretty much just for patchwerk.

Perhaps 25 man instances will get challenging to the point where we are chaincasting again but it's not the case yet.

 

Kortar

Desperate Prayer has always been a primarily PvP ability. Tack on a mana cost and the number of situations in PvE where its insta-cast nature trumps the value of Binding Heal seem pretty limited.

With +1500 healing, a BC-era Greater Heal does 4593 healing. With +2000 spellpower (entry level WotLK raiding), a Greater Heal does 9122 - or 198% of our BC Greater Heal. Also bear in mind that +1500 would have been extremely high for actual entry-level BC raiding - +1200 would have been more realistic for people who didn't have access to badge gear.

The phenomenon you're observing with base costs is a reflection of the fact that spellpower (unlike critical, haste, etc.) doesn't downgrade as you level so higher level gear always has to have more of it. In order to keep people's heals from going off the charts, they need to make a smaller percentage of the actual heal come from the base.

 

Ranc

Kinda liking the dual spec swap in raid to come I can have my cake and eat it.
If disc didnt quite have the Holy numbers seems we can now choose on a fight by fight basis yay.
Awesome change for us as healers, definitely gives us the edge as a healer class.

Now whos gonna tell our SP's their dual spec is responsible for 25 man raid DS...and may as well IDS

 

Ranc

Technically you wouldnt have had to loose all 25 mp5 from DS as you could always scroll for some of the extra spirit instead, not that im arguing against DS as the raid overall would all lose out which weights DS somewhat higher than just your personal loss of up to 25 mp5. But its a moot point for 25 man as that getting someone else to spec DS is an awful lot easier with dual speccing

 

Kortar

Fel Intelligence is 64 or (Improved) 70 Spirit, Scroll of Spirit is 64 Spirit. So Divine Spirit adds 16 Spirit over what you could get otherwise. For a Discipline Priest, Divine Spirit (but not Improved, which is superseded by Warlocks/Shaman) is probably a necessary talent - adding +10/+16 Spirit to the 7 or so people in your raid who care if they have a non-zero Spirit probably works out to a raid-wide total of around 30 mp5 or so. Which is pretty good for a single talent point.

 

Celillenna

For those doing math, at what point is it ok to give up spirit for spell crit on a disc priest? Would you take a 1:1 spirit -> spell crit trade on an item?

 

Ranc

I totally agree DS is indeed a great single point but my point was you can build a better Holy and Disc build (not much but a point is a point) without it - its not a filler point either.
The key as constantius said is its a matter of getting another priest to provide it - Dual spec should now solve that problem and so take the point out of our most optimised heal builds.

Personally I'll build a Holy and Disc dual specs for 25 man without DS, after I work on our SP's.....

5-10 man is diferent of course for that DS and even IDS (5 man no shammie/lock) are much more likely to be neccessary points.

 

Crepusculu

Shadow Priests will be likely pushing 54-55 points minimum in shadow to pick up all the PvE goodies, so don't expect your SPs to gladly pick up divine spirit to save your disciple build 2% mana on instant spells.

The whole spec switching in raids is an interesting thought. Will buffs be cleared if you swap specs, like when you respec? I suspect blizzard will ensure people aren't switching simply to buff up.

Addons
http://www.wowinterface.com/list.php?skinnerid=35267

 

• constantius

Only caveat I have is that it's 2-3 warlocks, 3-4 priests, 2-4 caster druids = 7-11. You could have quite a bit more than 7 people caring about having the buff, so in that situation, may have to justify it.

I'm looking seriously at having my dual specs be "deep Holy" (i.e. 14/57/0) and "balanced Holy" (21/50/0) for the start of Wrath. After I get enough gear to not make MA/DS useful / as useful, I'll be able to drop one of them into a pure deep Disc build for tank healing or PvP applications.

I think there will be too many times in early WotLK where I'm going to want both CoH and DS to skip having my second build set up as that. Especially when you consider heroics, where I may be going without BoK/BoW/GotW/AI, I'm going to want every possible regen talent max'd out. I may even in that situation consider building a 26/45/0 spec, for 5/5 MS and CoH+ a few goodies. Sometimes just having a lot of mana is helpful (i.e. on most trash pulls), as compared to having heavy regen.

Very cool that they're actually releasing this: now we can start arguing over what the optimal choice for the specs are. Once we get gear, I suspect it will be 51+/18+/0 and 14/57/0, but time will tell.

[e] Glyphs and Keybinds being tied to specs is *awesome*, btw. That's going to prevent the biggest hassle, which is setting up your UI again after respecing. Every time I go 41+ Disc for PvP, I have to drag all my bars around, get rid of my keybinds for CoH/etc, and put in the pvp-centric ones. No more!!

Anyone who cannot cope with mathematics is not fully human. At best he is a tolerable subhuman who has learned to wear shoes, bathe, and not make messes in the house. - R.A. Heinlein

 

Awina

When valuing MA against DP, for example, how are people accounting for overhealing?

One of the things that sticks out about building a spreadsheet where you actually model a fight (spell mix, etc) to value the talents and stats is that one has to make an assumption about overhealing on each spell.

For a CoH / PoM / BH mix of spells, if one assumed an average of:

CoH = 4/5 (or 5/6) targets; 20% overheal

PoM = 3.5/5; 20% overheal

BH = 2/2; 35% overheal

Even if you only penalize DP for the average overheal (and not it's portion of it, which would be higher), it will bring down the value of DP by 20% or more (depending on targets hit assumptions) relative to MA.

MA gets 100% benefit anytime you cast an instant spell; DP arguably only gets 80% benefit per target hit, and less than that if you factor in targets missed.

Any of the talents that benefit Spellpower/Healing need to be valued in this context. The HPS component is not considered here (ie: more people saved with the higher HPS?), and so that needs to be considered separately as a subjective component.

 

Tainter

But then overhealing reduces the effective hitpoints per mana spent. So, yes you definitely save the mana with MA, but the more you overheal the closer the talented and untalented HpM get to each other.

Imagine a spell that heals for exactly 1 Health for every 1 Mana spent (untalented).
At 0% overheal MA offers an additionial 0.11 HpM.
At 50% overheal that number comes down to 0.055 HpM.
In the extreme case of 100% overheal MA provides no benefit.

While the relative increase in HpM stays constant throughout the absolute benefit decreases with increasing overheal.

If you can't join them?
Beat them.

 

Shylena

And when you add in that CoH will now enjoy raid wide application and Intelligent Targeting, I see +healing being a better choice in that you will be healing the lowest targets in a set area.

 

Havoc12

Chain heal which is a shamans MAIN raid heal is 2.5 seconds no one is saying its too slow. Priests like to spam flash heal because its easier and because they can with the new regen. You can heal in the exact same way with gheal (I do it every raid), only you don't need shadow priests, innervates, mana totems and chain potting. Gheal is not too slow or too big.

Flash heal has nearly half the HPM of downrank gheal. You never enough overheal to turn things around, especially if you are good at using gheal. When you couple it with the ability to squeeze more ooFSR time from gheal the difference becomes extremely large.

No shadow resist gear. I heal both groups. The sequence uses 2x PoH. 2x PoM, 4-6x CoH and an occasional gheal/fheal. No one drops below 4k HP (they start at 8k), everyone is full 3 seconds after spirit bolts end.

bolts are about 550 DPS on 10ppl, which is 5500 DPS over 10 seconds. CoH/PoM and PoH can supply >4500 HPS over this period if used correctly.

CoH is not a stabilising heal because if a CoH/PoM can't outheal damage neither can chain heal, unless of course grouping is a limitation. You can have chain heal doing the topping up OR you can have *a second* priest with CoH instead of a shaman topping the damage up. This applies to ANY damage and ANY situation where everyone taking damage are in the same group.

 

Anedris

This is usually true. Keep in mind though that there are encounters where 2.5 seconds is too slow - Twins. The reason chain heal excels in that encounter is because the shaman can simply cast it continuously and even if the initial heal overheals 100% the jumps will still find targets. Twins is of course by no means the norm - it's an encounter that was specifically designed to be challenging to heal in an era when healers had nearly unlimited mana - but there are instances where speed is important. (I wouldn't dispute that many many priests use flash heal way too much when downranked gheal can do the same job for less mana... of course, downranking will shortly be a thing of the past which changes the calculus again...)

 

Sinndir

Nid, while I do like the glyph switching and possibly the Keybind switching the most... Do you think our glyphs are going to switch much between the two specs?

No matter the spec, I'll be keeping the Levitate, Shadow Protection, and Fortitude minor glyphs.

For Holy, I will have the Circle of Healing, Flash Heal, and either PoH or Mass Dispel.

For Disc, losing the CoH one for sure but I would definitely be keeping Flash Heal, PoH/Mass Dispel and probably changing the CoH one for the PW: Shield glyph.

I would like to hear everyone's proposed glyph setups.

 

The Not So Evil

Lately I have been working on Rawr.HolyPriest as an evolution of my Spreadsheet.

By loading Constantius' priest (Nidaba) into Rawr, using a 14/47 Spec, the model rates the talents in this order (pr talent point):

This is using a spell ratio of 1xRenew : 1xProM : 5xGreater Heal.
This is fully raidbuffed, 100% Serendipity Returns, 50% uptime of Replenishment, 6 minute fight, mana potion, 80% of full effect of Shadowfiend, 95% time spent in FSR.

Nothing really surprising was there.


Doing the same for a discipline 51/10 build:

Spell ratio of 1xPW: S : 3xPenance : 2xProM : 4x Greater Heal : 1x Flash Heal
Rest is same as Holy, 90% effect on Rapture.

 

Ranc

Well they havent stopped our swap to an alt for spirit previously so im hoping they dont bother with dual spec.
Obviously I was talking about there 2nd spec, Id take it if needed rather than making an SP in a main spec.

 

Hegen

Spirit bolts are - at the start of the fight - 450 shadow per person, but twice per second! Some of these will be partial resists (with shadow protection buff) at 338 shadow damage, and much more rarely a 50% resist at 225 shadow damage.

Example:
21:26'38.069 Hexlord Malacrass Geistblitze hits Hegen for 338 Shadow. (112 Resisted)
21:26'38.401 Hegen Gebet der Besserung heals Hegen for 1959.
21:26'38.415 Hexlord Malacrass Geistblitze hits Hegen for 450 Shadow.
21:26'38.961 Hexlord Malacrass Geistblitze hits Hegen for 450 Shadow.
21:26'39.478 Hexlord Malacrass Geistblitze hits Hegen for 450 Shadow.
21:26'39.932 Hexlord Malacrass Geistblitze hits Hegen for 450 Shadow.
21:26'40.382 Hexlord Malacrass Geistblitze hits Hegen for 450 Shadow.
21:26'40.967 Hexlord Malacrass Geistblitze hits Hegen for 338 Shadow. (112 Resisted)
21:26'41.365 Hexlord Malacrass Geistblitze hits Hegen for 338 Shadow. (112 Resisted)

On average, that's more like 7500 dps he does on the raid without SR gear. At 4 stacks, which should be doable even for very low dps groups, it's on average 550 (roughly) per hit after partial resists, giving a total of 11000 dps on the raid.

Like this (at 4 debuff stacks):
21:31'23.571 Hexlord Malacrass Geistblitze hits Hegen for 630 Shadow.
21:31'24.159 Hexlord Malacrass Geistblitze hits Hegen for 630 Shadow.
21:31'24.569 Hexlord Malacrass Geistblitze hits Hegen for 473 Shadow. (158 Resisted)
21:31'25.165 Hexlord Malacrass Geistblitze hits Hegen for 473 Shadow. (158 Resisted)
21:31'25.582 Hexlord Malacrass Geistblitze hits Hegen for 473 Shadow. (158 Resisted)

So, at the start of the fight, spirit bolts do around 75000 damage total, at 4 stacks around 110000 damage total. If you can indeed pull 4500 hps over 10 seconds, you are missing 30000 damage with 0 stacks, and 65000 at 4 stacks. How do you top that up within the stated 3 seconds?

Now, as for using two PoMs during spirit bolts, this is not guaranteed to work. If you are a bit unlucky, the person you cast PoM on about 3 seconds before spirit bolts takes damage so you only have one PoM guaranteed. This much depends on who gets spirit tapped. With a spirit tap on hunter, mage, warlock, shaman and druid you don't have a safe target for PoM. That's a side issue, though.

The next problem is: even if you do have both groups at around 4000 HP after spirit bolts (which requires quite some luck and also some high-HP classes in the raid), you take the risk of getting people one-shot if Malacrass spirit taps a warlock or mage.

The scenario you describe doesn't match my experience with the hexlord at all (which is basically 90% of all IDs since launch), and neither does it match my WWS. My guess is that the last run you remember actually WAS with SR gear, since in that case your scenario works out.

Edit: in the above log, "Geistblitze" is German for Spirit Bolts, and "Gebet der Besserung" is PoM.

"Ferals are obscenely good at soaking balls." - Nidaba

 

dukes

Just fixing tables (I'm totally anal about this, and I find it's quicker to reformat the table and read it than try and read it as it is). Pro tip: default Notepad has a fixed-width character font, so if you lay tables out in it, then just add spaces to make it line up, they show up perfectly with code tags.

What exactly are you trying with the spell breakdown? Is this testing tank throughput? I would assume more flash heals / some CoH given an "average" fight and the way the spell costs and things change with 3.0. I'm also not sure on the 100% Serendipity returns.

 

The Not So Evil

This was meant as a more generic Tank healing breakdown yes. I will be doing Raid healing later as soon as I do some modifications. 100% returns on Serendipity is probably not always going to happen, but seeing as Holy may be able to do 8000 GHeals before crits in 3.0, and Tank hp pools not going up, there is a very large likelyhood of having a very high return on Serendipity.


Doing the same to AoE Healing, same Holy and Disc specs:

Spell ratios is: 1xProM : 4x CoH : 2x Flash Heal
50% serendipity returns.

Spell ratios is: 2xPW: S : 1xPenance : 1xProM : 2x Flash Heal
90% Rapture returns.


Note that some talents are not in my original Talent specs, but I included their expected worth anyway. like Surge of Light, Borrowed Time and Divine Providence.