• constantius

Oo5SR time on Sapphiron can conceivably be as high as 30 or 40%, depending on how you manage your time. I can easily get full ticks every 15-20 seconds if I want to; the damage is extremely predictable and can be left alone for long periods of time with nothing more than a PoM. It's not a good fight to use as a test case.

Sarth+3 is. If you get more than 10% OO5SR time on Sarth+3 from the moment Tenebron lands to the moment Vesperon dies, you're unusual. It's incredibly intensive healing; chain-casting on every GCD to *hopefully* keep the tanks and raid alive. This is, of course, referring to 2-healer Sarth+3.10. Almost all of your OO5SR time comes from IHC procs, and to be honest, I haven't been getting those lately anyway. I just use it to catch up on healing I was falling behind on anyway, chain it into a PoM, then dodge yet another Churn.

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

 

Hegen

That is because a) you are specced disc instead of deep holy and b) that highly depends on luck, the amount of frost resistance and c) the other healer.

A deep holy priest healing Saphiron will be well advised to cast PoM on cooldown, which is 7s for a deep holy priest. That alone will dictate not really having a lot of time oo5sr. So far I have yet to do that fight in 10-man, but the aura more or less automatically makes PoM supremely effective, which leaves deep holy priests just 7s-1,5s-5s=0,5s per 7s interval of time oo5sr - even if he/she is not using anything else.

For disc priests, this may be a different story.

P.S.: Sure you *could* get more time oo5sr at Saphiron. But why would you want to if using PoM is most effective way to heal there?

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

 

◊ Cadfael

Disc at Sapphiron is basically PoM on CD, and depending wheter you just heal what you like or you have an assignment ocassionally put up a shield + penance combo on them (most often only penance would suffice but you start to do shields for the nice hasted penance + followup spell anyway). On collapses behind ice block you can do some additional PoH or HN, depending wheter you are still moving and how far away people are.

But top priority should be PoM, as it is incredibly efficient here.

I think I never did any Sapphiron with anyone in the raid having any Frost resistency beyond some aura or totem. It's just not required. Healing sapphiron is not hard at all except perhaps you start to iron man it with only 1 / 3 healers or something like that.


The only fight where mana is consistently a problem seems to be Malygos and the only fight where you are constantly GCD locked are the Sarth +2/+3 with low amount of healers. In the other fights I generally don't even use the shadowfiend or a potion.

 

The Not So Evil

You are all saying what estimated time you spend OFSR in your posts, but I am wondering what addon(s) you are using to get this information. Right now I'm limited to FuRegen, and I am not entirely sure if it still functions correctly. On Sarth3D.25 I was everything between 82% inFSR to 93% inFSR according to FuRegen.

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

 

Morakk

Mentioned a bit back. [e] Just realized that this may not actually track your time in 5sr, but rather just show you whether or not you are in it....

 

Havoc12

a) This is my PvP spec mate. I am always deep holy spec for raiding.

b) I dont agree with you. PoM on CD most of the time but not all the time worked better for me, as I had to use a lot more than PoM to keep people alive. ooFSR time can be very high due to the flight phases, especially for a priest with lightwell (and knows how to use it ofc)

c) The other healer was a pally mostly healing the tank. The fight took 8 mins and a lot of people were just taking the blizzards instead of moving out of them.

d) That was my first time in Naxx10 period. No frost resit on anyone (we got achie). I had 1900 sp and 800/270 unbuffed regen at the time.

e) Your calculations are horrible. If you cast PoM then ooFSR beins exactly 5 seconds after you cast it not after the GCD ends. If you cast nothing but PoM you get 2 seconds of ooFSR regen every 7.

 

bbartlog

Some good points here. I'll definitely concede the point about Shadowfiend - breaking OO5SR to use it, rather than thinking ahead a few seconds and casting it after whatever the last cast was, would be a significant mistake.

As for misunderstanding the way the ratios were expressed, sorry. It seemed like you favored spirit, so I had trouble wrapping my head around the idea that you were promoting 1500 Int and 1000 spirit. The 2K/2K number is still not achievable and you need fairly specific gear to hit 1500 Int / 1000 spirit (but this has more to do with Blizzard's itemization than with those overall numbers being too high).

The point about OO5SR time remains, however. You achieved 80% OO5SR time on Sapph? So what? The question here is '*if* the fight stresses your mana, so that you *actually care about OO5SR time*, then what can we deduce about the percentage you're likely to be seeing?'. You mention a technically difficult fight you had against the Four Horsemen as well, but again, since the difficulties you encountered didn't involve running low on mana (by the sound of it) it doesn't matter in the context of this argument.

From TBC to WoTLK (lvl 70 to lvl 80), you can see roughly the following ratios:

Base mana / spell costs: 1.47:1, so 47% greater
Mana pool: almost 2:1 (stats) so ~100% greater
Passive regen: about 70% higher (60% regen coefficient nerf * sqrt(2) for int stat * 2 for spirit stat)

So if you create a situation where you cast as many spells as possible in five minutes (limited by available mana), the number of casts will simply be higher at lvl 80 than they would have been at lvl 70, and by extension the time available for full regen ticks will be lower. I'll expand on this in another post to look at specific percentages. Further, this also means that as your gear improves (whether it's mana pool or passive regen) your percentage of time at full regen will decrease, *assuming* you are operating at the limit of your mana pool - because you will be able to cast more and more spells in the timeframe being considered.

As for your point 4), it's well understood that both spirit and int are more valuable the more you have of the other stat. The claim about int providing half the passive regen of spirit assumes equal amounts of both stats. In your example using 1500 int and only 1000 spirit, a +300 spirit proc would be preferred... but it's totally irrelevant, because the card proc can only apply to the higher stat and so you would never be able to get a +300 spirit proc in that situation. The choice of procs is only available to those who maintain int and spirit at roughly the same level. So you have to look at the value for someone with equal amounts of the stats (the marginal case) if you're talking about choosing between them.

 

Havoc12

I think you have looked at this the wrong way. Here is the value of 1int versus 1 spi at 1000 spi and varying intellect at 80%FSR

formulas
0.033620879*SQRT(I)*(1-0.7*%FSR) per point of spirit

0.033620879*S*(SQRT(I+1)-SQRT(I))*(1-0.7*%FSR) + 0.33 per point of int.

Notice how the value of intellect is decreasing while the value of spirit is increasing. Repeat the calculation varying spirit and you will see the other side of the argument.

 

Ranc

Your maths for the regen element seems to have an error crept into the int line, 20mp5 may have felt wrong but shouldnt that 330 int line be
With an extra 330 int, you gain 5 * .005775 * 1100 * (sqrt(1430) - sqrt(1100)) = 148 mp5.

20mp5 may have seemed lower than you guessed but seems it could be 6mp5 at 1100 levels!

Thanks for the maths post tbh, sites are for refining them and taking out errors. Im not trying to argue against you (we seem to agree) or spirit whores just wanted fairest stats posted

 

Hegen

Regarding b) and c), if people are taking blizzards, then having PoM on cooldown is even better. Of course PoM alone doesn't cut it, especially not if you are healing the raid mostly by yourself and there's no frost resist. Using the air phases to get more than 20% OO5SR would mean that your raid is mostly topped up at that point. Is that what you do?

Regarding e), that's correct, one shouldn't post in a work break. That aside, your tone could stand some improvement. Mistakes happen, as you well know.

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

 

Hegen

I no longer trust FuRegen. I use the WWS HPS time as quick check to see if I am interested to really find out. If I then really want to know, I use the combatlog to get a number, though it's lot of work.

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

 

Shadowed

It'll do time in/out of the five second rule some point later today, but it's not going to track every single stat like RegenFu did regardless.

 

• Playered

Something that has not been noticed yet from the PTR from the official patch notes:
"Shield Wall, Barkskin, Guardian Spirit, and Divine Protection are now off the Global Cooldown."

 

siegfried

Not Pain Suppression?

 

Lanthon

This. My OO5SR regen on most fights is less than 5%, because if healing is easy, I'm keeping dots up and holy firing. In Naxx 10 the other night, I broke 1k dps on Noth in my healing gear (no +hit). The paladin who judged light had more healing than I did. On Patchwerk, my dps time was 93%, but we missed the achievement by one second. Next time, I'm calling for a shield wall during heroism and nuking! On Sarth+drakes (10 man), I don't dare waste a global cooldown or mana on anything unnecessary, as it's the only fight where I have mana issues under normal circumstances.

I used to feel guilty about doing damage and not maximizing regen, but then I studied parses where top raiding priests spam greater heal on Patchwerk (25) with 90% overheal and I'm now glad I chose to play with more than 1 finger.

 

bbartlog

I decided to crunch some numbers to try and figure out what sort of time could be spent inside and outside the five-second rule, if we assume a priest uses all the mana he's got. (As an aside, is the 100% regen when someone is *inside* or *outside* the rule? I see both usages, it's almost always clear from context, but I usually think in terms of '*outside* the rule means 100% regen...').

Anyway, the results show a couple of things.
First of all, time outside the five second rule is highly dependent on how good you are at clumping spells. This I already knew.
Second of all, even with good regeneration and gear (i.e. a lot of spells getting cast) it should be possible to spend a fair bit of time outside the five second rule. This I was wrong about.
Third, the general effect I suggested in my last post, where better gear means less time at full (oo5sr) regen - assuming we are actually using all that mana - does exist and is big enough to matter.

Variables and formula I used:

CT = casttime of a 'typical' chunk of spells used by the priest. This could be just one spell, if no attempt at clumping is assumed. The casttime as I'm using it in the formula is not the time actually required to cast the spell, but is the time between first and last expenditure of mana in a set of spells being cast. For one spell therefore it would be zero; likewise if all our casts are cast-followed-by instant, it will also be zero.
CC = mana cost of a typical chunk of spells.
M = total size of manapool, for the entire fight, before considering oo5sr regen. So this will include all mana from other sources (replenishment, potions, mp5 not oo5sr, shadowfiend and base mana, etcetera).
O5M = additional mp5 we get when outside the 5-second rule.
O5T = total mana gained from oo5sr regen in the fight.
L = length of the fight, in seconds.
I = what we're trying to derive (intermediately), namely, the interval between clumps that we can sustain.
O5% = percent of time spent outside the 5-second-rule

I'm ignoring the effect of IHC clearcasts, Inner Focus, and the likely existence of natural breaks and phases in encounters.
Then:

O5% = (I - 5 - CT) / I (percentage of time spent oo5sr)
O5T = O5M * O5% * L/5 (total mana gained during that time)
(L / I) * CC = M + O5T (the constraint: we assume we spend all that mana plus whatever else we had)

Algebraically (I can post intermediate steps if you care):

I = (CC + O5M + O5M*CT/5) / (M/L + O5M/5)

Now we just need values to plug in. Reasonable numbers for O5M are in the 550-800 range, raid-buffed. For fight length (which will also affect total mana, M) we can just use 360 seconds. Total mana can range quite widely depending on how much support we have (things like Blessing of Wisdom, Replenishment, or mana totems) but 50K to 80K seems reasonable for a raid setting (and for the assumed fight length).

Priest casts spells one at a time (no clumping), average cost 800, 600 O5M, 60K mana:
I = (800 + 600 + 0) / (60K/360 + 120) = 4.87. Time oo5sr is zero.
This basically shows that if you just lazily cast a cheap spell at intervals, you may not see any time outside the five second rule.

Priest casts spells in bunches of threes, with average total cost 2400 and time between first and last cast of 2 seconds (same stats otherwise):
I = (2400 + 600 + 600 * 2/5) / (60K/360 + 120) = 11.29. Time outside of five second rule is 38%. Notice that even though the cost of the spells is the same as for the priest who casts one-at-a-time, this priest can cast nearly 30% more spells (!) due to more time oo5sr. Clump your spells when you can!

More geared priest, clumps of three:
I = (2400 + 800 + 800 * 2/5) / (80K/360 + 160) = 9.21. Time outside of five second rule is 24%. This demonstrates the effect I mentioned: as you get more geared, *if* you find encounters that are still challenging to your mana pool, you will spend less and less time oo5sr.

Clumps of two, cast+instant, with some more expensive spells like PoH mixed in to make CC = 2200 but CT still zero, priest with intermediate gear:
I = (2200 + 700 + 0) / (70K/360 + 140) = 8.7, but time oo5sr is 43% due to the effective cast time of zero. Chain instants after spells with cast time and win!

I'm aware of course that priests don't have a 'rotation' and so these numbers will not exactly match any real fight. But:
- first of all, they do suggest how much variation can be achieved (both in oo5sr time and by extension total healing) by managing your spells;
- second of all, they show I was wrong when I proposed 10% as a typical percentage of time to be spent oo5sr;
- third of all, moving around spell clumps in this model will only affect the oo5sr percentage if they get to be significantly closer to each other than five seconds. If 'I' is supposed to be 10 seconds average but actually varies between 7 and 15 seconds, it will not affect the percentage. The critical variable is the size of the clumps.

Of course all of this is also relevant primarily for longer fights. I haven't plugged in numbers for something like 2-healing Patchwerk while trying for the 3-minute achievement, but obviously you're not going to see much time oo5sr there...

 

Hegen

Regarding these points:

1. To make this a bit more precise: they suggest how much variation can be achieved in a theoretical encounter that matches your model closely enough.

2. That is not true. I am not saying you were originally correct, though you may have been. The point is that in order to derive that from your model you need to prove that your model is statistically close enough to reality to allow for this conclusion. This you have not shown and thus you don't know whether you have been wrong or not.

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

 

Blightmuffin

I think this is accurate from a priest's perspective but it is my opinion that this is NOT good from the tank's perspective as I believe it mitigates their TPS accumulation, which is imperative to establish as early as possible. PW:S a tank prior to pull can cause mobs to peel off onto dps when it otherwise wouldn't.

I've got a 80 DK tank where I've been doing some 25 man raids. Sometimes when I group with PUG priests, some of them have this tactic of wanting to PW:S me prior to any pull. Correct me if I'm wrong, but I believe that doing this prevents me from accumulating additional threat when I do group pulls. I commonly group with 2500-3000+DPS characters, and I already have trouble holding aggro with that much dps getting dropped. So I've been asking them not to shield me and usually I don't need it anyways, but just curious what you guys think.

I have a 72 shadow priest that I want to gradually level and make him a raid healer and I plan to go with a holy build. When I have player healer as a shadown spec, I generally find PW:S is great on overzealous clothy dps to add some time to throw an appropriate heal but I would never do that to the tank. Am I wrong?

Opinions much appreciated!

 

◊ Cadfael

Let's just say it has been proven that it won't be any problem for the two Rage based tank classes (Warrior and Druid) who get a bit less rage and thus might need to wait a very minor time longer to push an ability.

Now I really wonder what negative effect it is perceived to have on a DK who doesn't even have any rage.


A PVE Discipline priest will always try to shield the tank as that gives him +20% spell haste (yes, percent) for the next 6 seconds, might proc' DA on crit for further mitigation, returns mana on shield breaks (after 3.0.8 anyways) which is a dead given on a tank and increases his crit chance by 4% on him. He *should* be shielding the tank and it can be done prior to any combat as that will not draw any threat at all.

Now as you are saying "I already have trouble holding aggro with that much dps getting dropped" the problem lies elsewhere.

EDIT: Please note that you quoted an old post. Currently on live for example, no PoM activity produces any threat at all. Neither applying nor the heal itself. It's not been established if that is a bug or intended behaviour due to the change that now properly attributes the PoM healing to the casting priest.

 

Morakk

Power Word: Shield reduces threat indirectly. A warrior or druid doesn't get a little bit of rage from the damage he otherwise would have taken. A paladin doesn't get a little bit of mana from the healing he would otherwise would receive.

For death knights, you only way damage prevention can cost you RP is through delaying Scent of Blood, but I don't think that is significant. You still parry if you have a PW:S on you, so I don't see how it costs you threat.

 

Veldefice

Sorry to interject during the theorycrafting but:

Thanks for the response ;D It's nice to see another Ner'zhulian (??) on here. I check this thread often but there are so many things that you learn just by playing the game.

On Sartharion 3 drakes last night I went with my normal setup (4 piece T7) just because I was used to it and my mana was generally ok. Didn't have to pot but I used my fiend in between walls and on a drake. Still, with such a healing intensive fight, I was only oo5sr for about 15% of the entire fight (and most of it was after the drakes were dead).

For our 5 minute Malygos attempts, I tried the offset gloves and shoulders for more haste (haste rating is at 386 unbuffed) and there wasn't a huge difference in healing. It was nice to have the short cast time on flash heal and gheal, but I don't think it was really that noticible. I'll try it out more tonight during our raid.

Thanks for the feedback, I appreciate it. I think I'll keep the sets and work out a crit set and a haste set... or if I can, both stats together with what you're suggesting. I still like to stack spirit... so I'll see which pieces are worth it to drop some spirit.

 

Xaphania

It already annoys me enough when warrior/druid tanks tell me to stop shielding them because they can't build threat (I'm discipline). Now a death knight saying the same? There might be some special DK ability that I'm not aware of that depends on taking damage for threat generation, but I still really doubt that one PW:S is gonna make the difference between keeping a mob and losing it, if you are a decent tank, regardless of your class.

 

Morakk

I only noticed after I posted that you're in Lemmings.

The nicest crit+haste item is probably [Signet of Manifested Pain], mostly because it's 226 and there is no other 226 ring option. You might see a lot of competition though. [Pennant Cloak] is also 226, but there are 226 cloak options. Also competition again.

Everything else is 213:
[Cosmic Lights] is good since there are nicer options for many people (the sapph/maly necks etc.).
[Gothik's Cowl] isn't quite as well itemized as the other 213 helms imo and is dwarfed by the Malygos helm.
[Shroud of Luminosity] is like a jr. Pennant Cloak, and I don't think you'll see that much competition since all casters have the tailored cloak, right?
[Valorous Handwraps of Faith] could be easy or hard to get, no idea how many drops you've seen and how many Paladins/Warlocks you guys have. It is a nice piece though.

The best slots for this are the ring, neck, and the gloves somewhat behind I think.


5 minute Malygos seems like a fight well suited to haste+crit items. It's short enough, and for a lot of it the damage low/predictable enough, that you want to stack as much throughput as possible since you'll run the minimum of healers (my best attempts have gotten to p3 with 22+ alive with CoH/shaman x 2/paladin x 2, so I'm wondering if 4 is doable).

 

Kashir

There is absolutely no harm in shielding a DK tank, not even the almost negligible reduction caused by shielding a Warrior, Pally or Druid.

The only way we generate Runic Power are using Rune abilities, or on Parries / Dodges if you have Scent of Blood or Blessing of Sanctuary. Shields interfere with neither of these mechanics.

Any tank who complains about shields is probably an idiot. Any DK tank who complains about shields is definitely an idiot.

// Edit: Actually I just checked the talent, and Scent of Blood is on hit, not on Parry / Dodge. Maybe PWS interferes with that, but anyone taking that talent is an idiot anyway. It's a very poor talent for 3 points.

 

Lhyssa

I ran some Heroics as Disc with my guild's best geared warrior progression tank. Amazingly played, etc. We ran with our guild's highest DPS. As the warrior was overgeared, his rage gen was a little low, as he barely took any damage on trash. I don't think I shielded as much as someone who plays Disc full time (and the tank certainly didn't need the burst healing potential outside of a few bosses), but at the end of the run, he told me rage hadn't been a problem. The only time I saw it be a problem was when I accidentally shielded him at the start of a pull, and even then he kept threat.

Basically, with a warrior or druid in a Heroic, you don't always want to do it at very the start of the pull when aggro is the touchiest, but any time where you need the shield for the burst healing or to interrupt damage, threat/rage is NOT going to be a problem if the tank knows what they are doing. As for healing in a raid scenario, it's so unlikely that the shield will last more than a hit, perhaps two, on a boss. Even on trash I see them evaporate pretty quickly. A hit or two of prevented damage isn't going to mean much on something that hits hard enough to keep rage bars full on the other 15 seconds of hits.

I've healed a DK tank as Disc, and the shield made absolutely no different to his threat gen, and he loved it. So, the poster confuses me.