Glasswizard

Right now (meaning sunwell on live, I don't play beta) I would take SoL over IHC any day for the simple reason that CoH is so overpowered, that every holypriest should be healing the raid with CoH. The tanks are best left to other healers like druids and paladins (and hardly take any damage anyway anymore).

SoL mixes well with CoH. 5 chances to crit on CoH means you will have alot of surge of light procs to toss around free flash heals between CoH which is more or less my play style right now.

 

• Richelieu

I agree that SoL is superior in the current content if you are casting a lot of CoH. With only 1 point in SoL I have the proc up most of the time in all Sunwell fights. As a primary raid-healer with only 17% crit buffed I chose to spend no points in HC or IHC, and spent those saved points on DP and other goodies. I have not gone OOM on any fight through and including KJ, my HPS output is excellent, I'm not missing HC at all. Some build paths that will make little sense at level 80 (or even at level 71) can make a lot of sense for the moment.

 

Thorough

Ok we are stuck at the whole "Renew can't crit, you are out of your mind thing." I have seen it with the Blizzard combat text, no mods on at all, on patch day. When I saw it on patch day, it proc'ed inspiration. I had no glyphs at the time and was CoH spec. The first renew ticked for 736, the second ticked for 1052 on a mage friend. It was just the mage and I in the group. We were not actively fighting anything, I was simply healing her after a duel. Our buffs were fort, shadow prot, and AI. We were at Stonebreaker Hold in Terokkar Forest. No one had the towers at the time. I am not sure what else I can tell you about the second occurrence. (The first was caught with SCT, which has been discounted as a source of information.)

Renew is this wonderful spell where I can sit and tell you exactly how much it will tick for depending on each buff I have. It ticks for the exact same amount each and every time. I can tell you when tanks get buffs and debuffs that effect it. Every priest I have ever met can do this easily.

If someone would like to post another "Renew can't crit," please include an explanation for why it will randomly tick for 150% more for the entirety of the spell and proc inspiration without being a crit. Efforts are being made to capture this in a suitable manner by myself and others. Also note that I am not the only one it is occurring for. If you would like further information about every single occurrence I have noticed, feel free to send me a private message.

Moving on.

We are still left with several questions. If the same crit percentages granted to flash heal are considered for renew, is it still verging on worthless? If renew is not critting, then what could be causing rare 150% increases in the ticks? Can this then be used to increase the viability renew in a raiding situation? Could it be that the rarity of renew crits is related to a partial (or accidental) implementation of a future WotLK dynamic?

 

Abygail

I've reviewed all the WWS reports for our guild since the patch and there have been no instances of a Renew "critting" but I have seen some outrageous numbers for Renew ticks. I had to look at one of our other priests because I don't renew a lot, but on KJ the other night her average Renew tick was 1277 and her max was 3165. I also had some insanely high Renew ticks as well but I'm very sure this is a bug. I'll try to go back and analyze the attempt with the crazy numbers because it didn't happen all the time. Regardless, the WWS report does not show this amazingly high Renew tick as a crit.

I'm seeing some other strange things like Surge of Light proccing for other priests in the raid although I'm the only one with the talent.

Ability : Surge of Light
[browse log file]

Buffs Count
Abygayle 195
Morathoris 36
Lourdess 7
Athanor 7


The Twins room also appears to create a very crazy bug where healing can be increased by about 30-40%. I'm guessing that as others have mentioned it has to do with a mod, or it was simply just a fluke like some of the other strange things going on with buffs right now. At any rate, to answer your question....IF we could count on Renews critting at the same frequency as Flash or Binding, then yes, it would be much more viable in terms of HPS and would be AMAZING in terms of HPM.

 

Tainter

Without wanting to be childish (and possibly looking stupid once WotLK comes out):
A spell doesn't crit until I've seen it crit. (And every other player who uses the spell too.)

It wasn't mentioned by Blizzard, it's not in the patch notes, it's not on any other forum or website, I've seen no screen shots, it hasn't happened to me, nobody can reproduce it. If it really happened, and I'm still not convinced that it did, then it was a freak bug which needs reporting.

As for the other questions:
What you essentially ask is: How much increase in power does Renew need to be more desirable. Even if it could crit and trigger Inspiration it would still only amount to a 10% power increase at 20% crit and wouldn't have any secondary effects. For Holy maybe. For Disc no.

If you can't join them?
Beat them.

 

faesomething

I would agree that any sign of renew criting is probably a bug. I've seen many cases of odd healing behavior in 3.0.2.

This WWS shows a run in BT from a little while ago where all the healers were healing about 3x normal (example my Flash heal was 6131 non-crit) for the first few fights, as soon as someone died, their heals went back to normal.

Wow Web Stats

The only thing I can figure is there was a buff stacking bug.

 

Dekkar

I wish I had screenshotted at the time I saw it, but I wasn't quick enough to react or clairvoyant enough to have a combat log running for what should seem like a routine raid. I noticed though that I had a few outliers of very large healing numbers, to the tune of 20487 on a Mending crit (the subsequent Aegis proc being ridiculous as well). These happened at seemingly random times, and never with Greater Heal (I didn't use Renew, I was 56/5/0 disc). It did however happen with Flash Heal, but not Penance as far as I saw. I'm going to keep track of combat logs from now on and see if I can reproduce it with evidence.

 

Havoc12

Are you having mana problems? If so then SoL can make a difference, but if you don't you are worse off. An instant and a cast flash heal take the same amount of time, only a crit from a cast flash heal not only gives you more HPS it also gives you 2x ~1sec flash heals. Mana is a non issue on all level 70 content so SoL has no purpose in life except to reduce your HPS. 1 point on it, can be spent, but really you are better off spending it on lightwell.

In what way is it superior? It reduces your HPS in return for higher HPM. Do you actually need that higher HPM? The content is so nerfed that its a tall order. If you need to cast single target heals of any sort you are better off with IHC. Otherwise if you want to spam CoH test of faith and possibly DP if you can get it is the way to go. Lightwell even in its current state is miles better than SoL.

Getting lots of procs does not make a talent good. Its what the procs do that matters.

 

• Richelieu

Havoc , please keep in mind I am only talking about the transtion phase between current content and WotlK, in a raid-healer role. BTW I greatly respect your theorycrafting abilities and take your posts extremely seriously. You have helped me improve my gearing/gemming over the past months.

Am I better off with IHC in this phase? No. Just getting to any IHC costs 4 talent points. I have only 17% crit buffed as I mentioned.

With the talent points freed from not taking HC, my build takes Lightwell (which my raid actually uses) and 2 points DP (max available in a level 70 14/47 build). I like that much better than a very small chance of getting a hasted GH.

In this transient situation, raid-healing Sunwell and casting CoH a lot, a mere 1 point in SoL basically gives me a free no-mana instant FH whenever I really need it. And my PoM is cheap because I didn't spend points on HC.

Do I actually need that higher HPM from 1 point in SoL? Yes on KJ and M'uru, even nerfed. My guild has not been farming early Sunwell for months, our gear makes that 1 talent point very valuable.

Please keep in mind your theorycrafting has been oriented toward level 80, and I was merely talking about what works for a level 70 Sunwell priest raid-healer.

 

Havoc12

I am not able to specifically discuss any fight past brutallus. If your spec works out for you, then that is that. If you are willing to experiment then there are other possibilities.

Aside from that. If you have mana issues and you mostly cast CoH, SoL and then deep holy can help, but perhaps its worth considering mental agility instead, since you really can't get far enough into the holy tree. Otherwise test of faith could be a better investment than SoL.

As a rule of thumb, if 40-50% of your casts (which might come up to a lot less than 50% your effective healing) are single target heals, HC is a good investment. If its less than that, then HC is no longer efficient.

===================

With 15% crit chance and 2/2 SoL you are looking at a 0.5*(1-0.85^5) = 0.278147344 or 27.8% chance of procing SoL (for 370-470 mana depending on whether you have serendipity and the FH glyph) from a CoH cast. If you use every proc that is 18.7% - 23.8% in the reduction of CoH cost at a cost of 0-14% overall CoH HPS (depending on the circumstances), but I personally found it impossible to use even 1 every 5 procs. Anything past 50% of proc usage is completely unreasonably when you are spamming CoH IMO. So SoL is a 9-12% reduction in CoH cost depending on serendipity and glyph under the best of conditions. In BT my take on 2/2 SoL is that it represents a 4% reduction in the cost of CoH in BT, at a cost of like 1-2% HPS.

The problem I have with SoL is that you can cast 4 CoH in a row on a group of ppl to get them up to safe levels and there is a crit in all 4 of them. I don't want to use a free FH on one of them, because I just lose time, CoH would have healed that damage anyway, so I just wasted 3 procs. If I have 4-5 targets to heal I will heal them with CoH first and then look for someone to fire that free flash heal on.

SoL does proc a lot, but I just find the return of the procs very small. Also I find that even though I use FH a lot more now, I still get more effective healing with gheal, even when I raid heal, unless for some reason we have too many healers.

 

Glasswizard

Totally agree. If I wouldn't have mana problems our tanks would just pull trash even faster so yes, hpm is still an issue for me. And on KJ I still have enough mana problems to use a mana potion and ask for an innervate. Of course I'm totally dominating healing meters, so you could argue, that I'm overhealing or healing stuff that someone else should be healing but that's just my style. Overhealing is maybe not as elegant as relying on the right heal at the right time with enough HPS, but it get's the job done for me.



Well, that maybe the reason for the difference in specs. Post nerf damage from Felmyst to KJ is more or less mostly raiddamage. Tanks hardly take any significant damage anymore, one paladin (who can't heal AE anyway, we usally take 2 healspecced paladins) per tank is more than enough to compensate for that. Everything else is (still significant) raid damage. So as a holy priest, you use CoH most of the time anyway, because in those situations it's by far the most effective spell. And I agree, Mental Agility would also be a viable option.

 

François

I wonder if there has been a definitive answer on SoL with multi target holy spells ?
Is it 0.5*(1-(1-C)^#t) or 1-(1-C/2)^#t ?

Edit : I don't ask for probability formula explanation, but rather on the implementation detail by Blizzard. Imagine a "mana refund talent" working on CoH : "Each time your healing spell critically hit, you are refund 10% of the cost", would it have a chance to refund 20% if two targets are critically hit, or would Blizzard implement an "afterward check" that at least one target is critically hit, to make sure there can be only a 10% refund. Now translate it to the Surge of light proc, and depending on the implementation of the talent proc, you have the first formula (afterward check), or the second (independent proc chance on targets).

 

Havoc12

Actually I am not sure. I think you are right. its 50% for each crit so the chance of procing SoL is 0.5C. and the formula should be 1-(1-C/2)^#t that would make CoH with 15% crit have a 32.3% chance of procing SoL

 

The Not So Evil

Went to PTR and tested talents that may improve PW: Shield. Borrowed Time now works as expected.

Talents I found to improve PW: Shield:
- Twin Disciplines (TD)
- Imp. PW: S (IP)
- Focused Power (FP)
- Borrowed Time (BT)

Formula seems to be:
(1265*(1+IP*0.05) + SpellPower*(1.5/3.5 + BT*0.08)*1.88)*(1+IP*0.05)*(1+FP*0.02)

You may notice a lack of Twin Disciplines. The reason for this is that I still haven't figured out how it fits in.

0 SPP, 0/5 TD: 1265
0 SPP, 1/5 TD: 1277
0 SPP, 2/5 TD: 1290
0 SPP, 3/5 TD: 1302
0 SPP, 4/5 TD: 1315
0 SPP, 5/5 TD: 1328
1085 SPP, 0/5 TD: 2140
1085 SPP, 5/5 TD: 2147

No idea how 5/5 TD can only give a 7 absorb increase with 1085 SPP, but 63 absorb increase with 0 SPP.

Rawr - Coder of HolyPriest (Healer) and ShadowPriest (DPS) Modules.
Get Your Rawr 2.2.20!

 

Anaram

Just speculating here, but did the mage cast focus magic on you by any chance?

 

• Richelieu

That will clearly be true at level 80. For the next few weeks however, while we are in this strange situation of being level 70s in a 3.0.2 world raiding TBC content, it is worth noting that a relatively well-geared priest casting 40% direct heals can get the mana-regen effect of spending 4 points in HC/IHC just by wearing the Blue Dragon trinket. Here is my back-of the envelope calculation:

Assumed constants and variables definitions:

CRIT% = .17 about right for a buffed and moderately well-geared level 70 SW priest

DIFF = 120 = difference between 5SR and OO5SR regen, measured in mana per second. Again this differential is typical for a buffed and moderately well-geared 70 priest.

H% = .4 = percentage of cast spells that are GH, BH or FH (Havoc's threshold quoted above).

PROC% = % chance of proc per spell cast. This will be .02 for the trinket. For HC, let's compare 4 points in HC/IHC. So for HC PROC% = .35 * H% * CRIT% = .0238

FSR = .85 = overall percentage of time spent in FSR

GH = 825 = mana cost of Greater Heal with 3/3 Improved Healing. I'll assume the caster uses every HC proc to cast a GH to get maximum mana savings from HC.

T = number of seconds of OO5SR regen experienced after the proc that would not otherwise have been experienced. For the trinket this will be 15 * FSR = 12.75. For HC, I'll make a seat of the pants estimation that T is 4. Sometimes you can chain HC with an Inner Focus and get a 10 second OO5SR, sometimes IF is on cooldown and you must keep spamming and get no additional OO5SR even with the proc.

OK so for such a priest in an 85% FSR fight the average mana per cast gained from the trinket is then:

PROC% * DIFF * T = 30.6

The average mana gained and saved per cast from 4/6 HC/IHC would be:

(PROC% * DIFF * T) + (PROC% * GH) = 11.4 + 19.6 = 31.0

The opportunity cost of equipping the trinket is roughly 70 spellpower (e.g. not equipping Battlemaster's Courage), so by equipping the trinket you free up 4 talent points at a cost of about 17 SP each. That's not too bad, plus the trinket can be used situationally, as opposed to always having 4 points locked up in HC.

Of course, if you are primarily raid healing and casting less than 40% direct heals, the trinket makes even more sense. And if you are casting CoH heavily, I'd expect that Eye of Gruul is effectively as good as full HC/IHC. (I still don't have that trinket after 20+ attempts.)

I did the same calculations using some vaguely informed guesstimates of what the values would be for a level 80 Naxx priest, and it was clear that under any sane estimate of level 80 regen and crit HC will pull far ahead of the Blue Dragon by level 80. But for now the Blue Dragon still holds its own.

 

Promethia

Sure it's not 2247 instead of 2147? 2140*(1.05) = 2247, and then it would all make sense.

Also, you found focused power only increased the spell power bonus? I thought it increased healing, not just bonus healing.

It is the latter formula, which is the probability of getting at least 1 SoL proc in #t tries. The logic gets a little confusing but it's basically this:

1. C/2 (50% of C) is the probability of getting one SoL proc
2. 1 - C/2 is the probability of not getting a SoL proc
3. (1 - C/2)^#t is the probability of not getting a SoL proc, #t times in a row, i.e. the probability of failing to proc SoL #t times in a row
4. 1 - (1 - C/2)^#t is the probability of not failing to proc SoL #t times in a row, which is the probability of getting at least one SoL proc in #t tries.

Make sense? Double negatives are fun...

The first formula is half the probability of getting at least one crit in #t tries... which isn't what you want.

 

Thorough

No. She was deep fire at the time.

 

Incoherence

Yeah, I think I mentioned this a couple pages back: if any one priest in the raid has Surge of Light, anyone can get procs from it. Happened to me in a battleground where I was discipline and another priest was holy. Still haven't seen any renew crits, though.

edit: my previous post is here

 

Suhné

I've also experienced the contagious surge of light bug. It was weird when the only holy priest in the raid gave me SoL procs, but it became weirder when he left the raid and I kept getting those procs as full disc.

Another thing I wanted to point out is how procs like the one of Eye of Gruul (I think there are similar items in WotLK) work with penance. It seems any of the ticks from penance can proc Eye of Gruul, however something is bugged as the proc should have made my next penance free and I still spent the same amount of mana, that didn't happen if I used the proc on something different than penance.

 

Promethia

A few points, especially on the rapture returns from divine aegis:

First, I think there is a missing factor of 0.5 on the crit rate in one place. The total/raw HPS is (1 + 0.5*crit) times your base (non-crit) HPS. That is:





Second, the change in Mp5 from changing int is the derivative of the above Mp5 formula with respect to int. This does get a little complicated because both the crit rate and max mana vary with int. Moreover, as you change int, your raw/total HPS will change due to an change in your crit rate, but your base HPS should remain the same. So it's easier to stick with base HPS most of the time. However, one can use the product rule to get a simpler yet more precise result:





The above formula makes no assumptions about maximum mana or crit rate and should be valid for all values. It does assume all divine aegis shields are completely consumed by damage, which may be untrue to varying degrees. So the above is really a ceiling estimate, and in practice you may see lower values. Also note that "raw" HPS means HPS including overhealing, but it only includes healing on spells which can crit and proc a DA shield.

 

The Not So Evil

Well, now I feel stupid. I went back to PTR and tested 5/5 TD with 1085 SPP, and not so surprisingly, I got 2247. So basically, formula for PW: Shield on Current PTR is:

Base Absorb (BA) = 1265
Improved PW: Shield (IP)
Spell Power (SP)
Borrowed Time (BT)
Focused Power (FP)
Twin Disciplines (TD)

(BA*(1+IP*0.05) + SP*(1.5/3.5 + BT*0.08)*1.88)*(1+IP*0.05)*(1+FP*0.02)*(1+TD*0.01)

Focused Power increases the entire Absorb amount by 2/4%.

Rawr - Coder of HolyPriest (Healer) and ShadowPriest (DPS) Modules.
Get Your Rawr 2.2.20!

 

bbartlog

under any sane estimate of level 80 regen and crit HC will pull far ahead of the Blue Dragon by level 80

Why do you think so? My impression of the 70 to 80 mechanics is that while crit rating will increase a lot (double, more or less), the actual crit chance will remain the same at comparable gear levels. As for the regen, the Blue Dragon trinket scales with regen, so that's not a factor either. Blue Dragon seems like one of the few trinkets that scales pretty much indefinitely.
The rest of your analysis seems pretty good. I do think it's worth pointing out that the Blue Dragon trinket does not require any particular choice of spell (such as Greater Heal) in order to be maximally effective. On the other hand, the fact that it achieves its (quite good) mana boost by having a low chance for a rather large gain means that it's unreliable. If we imagine that we're modeling a 5-minute endurance fight and expect that we're going to cast 75 spells during that timeframe, some trinket that procs on 2% of casts has about one chance in five of not proccing at all over that number of casts. If you're one of eight healers in a 25-man raid that probably doesn't matter, but if your interest is 5-mans and you're healing Heroic BM or something that could be a problem.

 

• typobox

If you just take a quick look at T7, it's itemized with a lot more crit rating than TBC gear was. Crit rating is scaling in a way that you generally won't gain crit percentage by simply upgrading your old crit gear to new crit gear... the difference is that you'll likely be wearing more pieces with crit than in the past. Also, with the shift in preference towards Intellect, you'll likely be getting more crit from there.

 

• constantius

I'm unusual in that I ran 16% crit through the back half of Sunwell. Most priests didn't bother going that high due to diminishing returns on Inspiration.

In a mixture of Sunwell, T7.10 and T7.25 gear on the Beta, I have 24% crit, raid buffed. Now, part of this is due to the fact that I now get Moonkin / Ele shaman (+5% crit), but it's still a dramatic climb. We're getting items with crit on them now. It's a big difference.

Look at all the T6 and Sunwell pieces. The only piece itemized with crit that I can think of off the top of my head is [Pantaloons of Calming Strife]. Now, look at the Naxx pieces. You could put crit on 90% of your slots if you wanted to. The only exception I can think off of the top of my head is chest, because both T7.10 and T7.25 chests are stam/int/spi/spell/Mp5 for some reason.

[e] To below: that's what I was trying to get at. I expect to be running 30% crit in Ulduar. It's finally viable, and it helps our throughput as well, given that we're taking it for regen, but gaining so much 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