After breaking this down a bit, I'd have to disagree. Haste does affect all three measures, and is the only stat to do so. Saying that haste has no effect on HPM is in fact false, given that the "time dilation" effect changes the amount of effective mana spent per cast.
I do not understand what this means. How does the "time dilation" effect changes the amount of effective mana spent per cast? Do you mean you lose less to overhealing? Or do you mean you will lose mana faster therefore having less mana to gain from regeneration. In either case this is not correct.
Basically, the HPM equation for your basic heal comes out to this (if I did the math right):
Can you explain the parameters in each equation? I think a general formula explaining what each number is would be more helpfull in understanding your reasonining.
It looks to me that your reasoning is based on decreased mana regeneration during a cast and if that is the case it is incorrect to base HPM calculations on this. HPM is how much mana is required for a certain amount of healing and the mana you gain from regeneration is a function of the fight's length, not the casting time. Only a decrease in overheal can change HPM, but the effect of haste on this is difficult to determine and is most likely going to change non-linearly with increasing haste. It is hence best to ignore it for now.
It looks to me like you are trying to incorporate the increased mana consumption into HPM, but that is not a usefull number, the values of HPM, HPS and mana consumption are the 3 values that a healing priest is interested. The amount of mana regenerated through mp5 is dependent solely on the length of the fight and is not affected by your casting speed. The amount of time regenerated through the 5 second rule, is dependent on your casting pattern and though it may seem counterintuitive, faster casts for priest can improve your oofsr time in a similar way to +heal. I expect the difference will be negilible but we can calculate it if you are not convinced.
The only other way haste can change HPM is by allowing complete replacement of flash heal with gheal, but again this is a meaningless value, since every priest is different. I use flash heal very rarely. I only use flash heal, when I can't use an instant, binding heal or gheal, and that is a very rare occurence indeed. Some priests say that you need flash heal later on, but I have zero confidence in their words. I had people swearing blind that flash heal is better on leotheras and vasjh, but on leo I am nearly always the top healer by a 30% margin and never cast a single flash heal, while on vasjh after 3 days of attempts and including our successfull kill I was without fail double the next healer and above 500k total healing (549 in fact in the vasjh kill) with under 5% flash heal. Exactly the same thing for the first set of Hyjal and BT trash and that is with only 4 healers and a 15 man raid.
Originally Posted by Hegen
The downside here is that you need to heal back the stamina loss each time you switch from your spellsurge mace to the badge mace, costing you mana for the heal as well as a GCD (minus your haste) of OO5SR mana regeneration, which is a lot at around 1000 mp5 raidbuffed. Is this worth getting the spellsurge proc statistically 1,9% faster?
This effect is neglible I do not heal the deficit at all, I wait for it to be passively healed, though AoE or passive heal effects or through direct healing when I get hit. It is often passively healed by the shadow priest anyway who is no longer threat limited so he can use VE. The sta loss is nothing more than a nuisance and the dark blessing is indeed for me the perfect spellsurge weapon.
==> Re: Haste and exponential scaling.
Remember that x% haste does not mean x% reduced cast time. The reduction in cast time does not increase linearly with % haste, in fact the curve is similar to a parabola, it can be approximated as linear early on, but it takes increasing amounts of haste to achieve the same increment in cast time reduction with increasing haste. On the other hand throughput i.e. how many casts you can get per second scales absoludely linear with haste. Thus 1% haste means 1% more casts. 10% haste meaans 10% more casts. 100% haste means 100% more casts. The value of haste is that it is percentage based, unlike +heal which adds an absolute amount of healing to each spell.
[edit] Since there are a lot of people questioning the value of haste, here is my post from the 2.3 thread.[/edit]
Here is a more general treament for Haste:
A = total amount (including crit) healed by a particular heal at a particular value of +healing
B = base cast time or GCD
c = 1/1570
T = converted cast time or GCD
S = spell haste rating
H = +healing additional to what you already have
k = constant converting +healing to actual healing for the particular spell
T = B/(1+cS)
HPS = (A+kH)/T = (A+kH)*(1+cS)/B
For H=0, S=S'
HPS' = A(1+cS')/B
To get HPS' by adding extra +healing and without haste you will need H=H', S=0
In other words the value of haste in terms of +healing is dependent upon the A/k ratio
"A" depends on how much this particular spell heals at your current +healing.
"k" is a constant for each particular spell.
The obvious conclusion is that value of haste over +healing in terms of HPS increases the more +healing you have. There will be a certain value of +healing for each particular spell above which c*A/k will exceed the ratio of the +healing/haste rating itemvalue ratio meaning that from that point on to increase your HPS you are better off stacking haste instead of +healing.
With respect to haste and +heal and the conflict between them:
The rough conclusion is: haste is good, once you have the regen and the +heal to support it. Stacking it is a viable way to increase HpS without effecting HpM, and in some cases (esp. CoH) can be a *more* effective way to scale HpS, since the spell scales so poorly with +heal.
Any time you're doing a haste argument, just acknowledge that we have 3 primary spells: Renew, Circle of Healing, and Greater Heal. Pick two typical ranks of GH (3 and 7 should work), and then base your argument off the 3 spells that actually benefit (since Renew does not). The secondary benefit of reduced GCD is rather meaningless in the grand scheme of things, although very nice for Felmyst.
We all know Improved Cast Time = (Base Cast Time) / (1+Spell Haste Rating/1570). On top of that, we assume Greater Heal has a base cast time of 3.0, reduced to 2.5 by Divine Fury; assume 5/5 Empowered Healing, and a 20/41/0 spec that includes Circle of Healing. Assume a raid-buffed Holy crit % of 10%, just for sake of argument (it doesn't change much if it's 11 or 12, and no-one runs 20% that I'm aware of).
So, as mentioned several times by previous posters, HpS scales at 0.475 of your Healing S&E, assuming 5/5 Spiritual Healing and 10% crit. Every point of +heal you stack gains you an additional 0.475 HpS.
0.0003 is a fairly small rating, but it does say that for every 100 haste rating you gain, you will also additionally begin gaining 0.03 HpS from each point of +heal you have. When many of us are running 2500 +heal, that's another 75 HpS, which is nothing to sneeze at.
The scaling is exactly the same as above. Every point of haste give a direct increase of SHR/1570 to your HpS, again *after* HSE is factored in. This, of course, is not amazing news. :p
Conclusions and Remarks: (for those who want to skip the mind-numbing mathematics)
Basically, what we're really curious about is the equivalence point. At a given level of +Heal, assuming only GH, how valuable is each individual point of Haste versus a point of Healing S&E?
100 HSE increases HpS by 47.5, assuming no haste, additively.
100 SHR increased HpS by 6.369%, multiplicatively
And that's the fundamental problem: there's no easy conversion between a percentage-based HpS increase and a flat additive increase. However, we can ballpark some numbers:
At this level, adding 100 HSE will increase HpS by:
GH3: 50.8 HpS (2.5%)
GH7: 50.8 HpS (2.0%)
Since 40 SHR = 88 HSE in ilvl points, let's add 46 SHR and see what it increases HpS by:
GH3: 55.1 HpS (2.7%)
GH7: 69.6 HpS (2.7%, as expected)
In both cases, adding 46 haste rating is worth slightly more (quite a bit more in the case of GH:7) than adding an equivalent amount of HSE. This is really not surprising, given how far a little bit of haste goes. If we were to add *representative* percentages, then it would be a much stronger case for HSE -- adding 100 HSE adds only 4% additional HSE, whereas adding 46 haste rating is adding 42%!!
This is now updated to (hopefully) be correct formulae. Haste is a tricky thing, and trying to quantify exactly how much you should have is not a simple matter.
Last edited by constantius : 05/08/08 at 2:55 PM.
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
Stacking it is a viable way to increase HpS without effecting HpM, and in some cases (esp. CoH) can be a *more* effective way to scale HpS, since the spell scales so poorly with +heal.
Its not just the scaling that matters but the Current Amount Healed over Scaling factor (A/k) ratio.
For example for CoH at 2k healing this ratio is 4209.855565, while for gheal7 with 5/5 empowered healing its 4729.72973
The bigger this number the greater the amount of +healing required to achieve the same HPS increase as a given amount of haste. I.e. the bigger the number the greater the benefit of haste.
In otherwords gheal max rank benefits from haste more than CoH does, despite both of them having roughly the same scaling (116% for gheal and 120% for 5 targets for CoH). The difference is that gheal has a larger base heal amount, which adds extra value to haste.
In contrast flash heal which scales poorly (52.8%) with +heal has a A/k ratio of 4471.744472, i.e. it scales less well with haste than gheal does.
Prayer of Healing our highest scaling spell (32% per target) on the other hand has an very high A/k ratio of 6706.875.
Showing that it benefits more from haste than any other spell in our arsenal.
What one should be looking when determining whether a spell scales well, is not the absolute scaling but the base amount healed compared to the contribution from +healing.
I do not understand what this means. How does the "time dilation" effect changes the amount of effective mana spent per cast? Do you mean you lose less to overhealing? Or do you mean you will lose mana faster therefore having less mana to gain from regeneration. In either case this is not correct.
Well, I'm probably doing it wrong, but part of your regen is IFSR, correct? Typical GHeal base cast time is 3 seconds, meaning you get x ticks of regen during the cast time. Well, if you reduce the cast time to 2.5 seconds through haste, you now gain 2.5/3.0x ticks of regen, changing the amount of effective mana spent, or the delta in mana between when you started casting and when you finished.
If you have 250 mp5 IFSR regen, that's ~50 mps. For unhasted GHeal (max rank), you end up with 825 - 3.0(50) mana less than when you started casting, or 675. However, with a 2.5 speed GHeal, it's only ~125 mana from regen across the cast time, meaning your delta is -700, rather than -675.
[E] Forgot that the FSR starts when you last spent mana, so technically, the above could be OoFSR as well, if you dropped into it before casting
I decided to look at it this way, with units of measurement h=+heal, m=mana consumption, s=seconds:
HPS = x h/s
HPM = y h/m
MPS = MP5/5 = z m/s
Looking at the units, you can relate them this way: HPM = HPS/MPS
Originally Posted by Havoc12
Can you explain the parameters in each equation? I think a general formula explaining what each number is would be more helpfull in understanding your reasonining.
Comparing my equation and one of yours (with some emphasis):
The bolded portion of my equation is essentially the numerator. Comparing that portion to yours, I see:
A = hbase
k = bct/3.5
H = h
Originally Posted by Havoc12
It looks to me that your reasoning is based on decreased mana regeneration during a cast and if that is the case it is incorrect to base HPM calculations on this. HPM is how much mana is required for a certain amount of healing and the mana you gain from regeneration is a function of the fight's length, not the casting time. Only a decrease in overheal can change HPM, but the effect of haste on this is difficult to determine and is most likely going to change non-linearly with increasing haste. It is hence best to ignore it for now.
It looks to me like you are trying to incorporate the increased mana consumption into HPM, but that is not a usefull number, the values of HPM, HPS and mana consumption are the 3 values that a healing priest is interested. The amount of mana regenerated through mp5 is dependent solely on the length of the fight and is not affected by your casting speed. The amount of time regenerated through the 5 second rule, is dependent on your casting pattern and though it may seem counterintuitive, faster casts for priest can improve your oofsr time in a similar way to +heal. I expect the difference will be negilible but we can calculate it if you are not convinced.
This is exactly what I'm trying to do. Mana consumption is directly related to the amount of mana consumed per spell cast. Due to the inherent time-based nature of every fight, mana consumption, and hence regeneration is also affected by time. However, while haste affects number of spells cast, and amount of health gained per second, it inversely affects the amount of mana regenerated across that same number of spell casts.
I'm looking at the fight as a series of casts, essentially. You can even model OFSR times as spells with 0 mana cost and increased IFSR, if you'd like. Instead of looking at mana regen as "giving you a bigger mana pool" I'm looking at it is as "less mana spent per cast." I fail to see how this is an invalid perspective, rather than just a different one. However, I don't actively play a priest, so I've got plenty of room for re-education.
The equation that I posted is HPM = HPS/MPS. The HPS equation is the same as yours:
(hbase + (bct/3.5)h)/(bct/(1+s/1577)))
The MPS equation is as follows:
(mbase - m[bct/(1+s/1577)]/5)/(bct/(1+s/1577)))
Same units as I posted before, though I should clarify that m is mp5 either IFSR or OoFSR, depending on when you cast the spell.
If you take both equations for the same spell, and take their ratio to get HPM for that spell, the denominators for both equations drop out, leaving you with the equation I had originally posted. However, since the numerator on the MPS is affected by haste, haste remains in the final HPM equation.
I guess as one final illustration, imagine it this way:
As you can see, haste gives you more casts (1 in this primitive example) but leaves you with the same number of ticks with which to regen mana (regardless of IFSR or OoFSR). For the fight, both your total mana consumption and healing went up, but not by the same amount, or even the same ratio. In fact, your mana consumption has to increase faster than your healing output, causing your HPM to go down over the fight as a whole.
Constantius are you sure about your formula? The values you are posting come up wrong in my healing spreadsheet and in the healing spreadsheet posted in google. My formula for +healing from gheal is
base + (3/3.5+empowered healing)*(Heal Bonus). It looks to me as though you are applying the 3/3.5 factor to the BASE heal of the spell which is not correct, AFAIK. This factor is applied to the bonus healing effects alone. Its far simpler to use the formula I posted.
Originally Posted by Torq
Well, I'm probably doing it wrong, but part of your regen is IFSR, correct? Typical GHeal base cast time is 3 seconds, meaning you get x ticks of regen during the cast time. Well, if you reduce the cast time to 2.5 seconds through haste, you now gain 2.5/3.0x ticks of regen, changing the amount of effective mana spent, or the delta in mana between when you started casting and when you finished.
IFSR is 5 seconds AFTER a spell is cast not while its channelled. Lets say I begin casting a gheal. The 2.5 seconds my gheal takes to land is still OFSR regeneration. When my gheal resolves and the mana is deducted, for the next 5 seconds whether I cast or not it will be IFSR regeneration. In other words during combat I am getting IFSR regeneration UNLESS 5 seconds have elapsed since my last heal. The only way I can get OFSR regeneration during combat is if I manage to avoid landing a spell for more than 5 seconds. Now lets say that my target has to be healed at the very worst when he is 70% HP and it will take him exactly 8 seconds after my last heal. Even if my haste has reduced gheal to 2 seconds instead of 3, my next heal will be timed to land exactly 8 seconds after my last heal landed, I will simply start casting a second later. Hence during this time I will have 5 seconds of IFSR time and 3 seconds OFSR time regardless of the amount of haste I have.
As you can see, haste gives you more casts (1 in this primitive example) but leaves you with the same number of ticks with which to regen mana (regardless of IFSR or OoFSR). For the fight, both your total mana consumption and healing went up, but not by the same amount, or even the same ratio. In fact, your mana consumption has to increase faster than your healing output, causing your HPM to go down over the fight as a whole.
I am afraid this is wrong. Whether the ticks are IFSR or OFSR is a very important difference. If none of the gaps in casting are longer than 5 seconds, as it appears to be in your example, the entire length of time is IFSR and regeneration will be identical during this period. Also if your spell costs 10 mana and heals for 100 health, then if you cast 11 spells with haste compared with 10 spells without haste, then you will have spent 110 mana and healed 1100 health compared with 100 mana for 1000 health. I.e. haste has increased your healing output by 10% and your mana consumption by the exact same amount.
Each spell heals for a certain amount (depending on total overheal) and costs a certain amount of mana. The total amount HEALED over the total mana SPENT is your HPM, i.e. effective healing per mana point. This does not change by haste, unless haste increases overheal (more likely the inverse is true). This number is a very important value for a priest, because it determines how much mana regeneration you need for a particular fight. Trying to combine mana regeneration and HPM in one, is pointless. HPM you can determine experimentally from the stats and you need that value to determine how much regeneration you need for a particular fight.
The mana you regenerate for any given time, depends not on the number of spellcasts, but on the number of breaks between spellcasts that were longer than 5 seconds. Hence if someone spends x seconds chain casting (i.e. complete IFSR regen) and then takes a y second break before begining to cast again, his regen will be x+5 seconds of IFSR and y-5 seconds of OFSR regeneration. In those x seconds this person tried to heal a certain amount of damage. If another person has haste they will heal that amount of damage in x' time and then take a y' break in casting. To understand how haste changes your regeneration it is necessary to understand how having haste instead of +healing influences the relative length of x and x' as well as y and y'. Because haste increases HPS faster than +healing does, having haste instead of +healing means that you can heal the same amount of damage in less time, thus x'<x and y'>y meaning that overal regeneration will be higher with haste than without .
I think what confuses your calculation is that you are looking at a fight in terms of total length of casts. That is not a good way to model regeneration. For regenation you don't care about how many casts you have but how these casts are spaced. If for example you have only a few casts of flash heal but in regular 4 second intervals you will be IFSR for 100% of the time despite spending 62.5% of your time not channeling any spell at all. Whereas if you are landing a flash heal every 2 seconds for the first 90% of the fight and do not cast at all for the last 10% of the fight, you will have 90% IFSR and 10% OFSR regeneration, which is higher than 100% IFSR despite a higher number of casts. In otherwords the number of casts has nothing to do with overal regeneration.
Well, in my example, I wasn't differentiating between IFSR and OoFSR in on the ticks. I understand it makes a very large difference, but you can take that into account (m changes in the formula, as I made a note of in the opening statements).
However, this brings up another difficulty, then, if you're saying that regardless of haste, you'll time your heals to give you the same number of ticks of regen. If I understand what you're saying right, about timing, it'd end up looking like this:
That causes the problem where now you've added haste, but gotten absolutely nothing out of it. Your spells cast faster, but since you're keeping the time between when they land (or spend mana) a constant, you're still casting the same number of spells, healing the same amount, and spending the same amount of mana. To look at it another way: Your HPS per spell goes up, but your overall HPS stays constant. By the same note, however, what I was saying it also true: Your effective mana spent per spell also goes up.
That is to say, any place where you'd actually take advantage of haste (increasing the number of spells cast in x duration) you're reducing your time in OoFSR, decreasing your overall HPM. If you're keeping your time in OoFSR a constant, and adjusting your healing accordingly, you're removing any effect haste has, implying you shouldn't be stacking haste at all, or at best, haste is only effective situationally where you'd need to spam spells to keep people alive).
Originally Posted by Havoc12
Also if your spell costs 10 mana and heals for 100 health, then if you cast 11 spells with haste compared with 10 spells without haste, then you will have spent 110 mana and healed 1100 health compared with 100 mana for 1000 health. I.e. haste has increased your healing output by 10% and your mana consumption by the exact same amount.
So to counter this point: If you cast 11 spells in the time it would have taken you to cast 10, you've reduced the percentage of the time you spent in OoFSR, reducing the amount of mana you regained by some amount. So yes, you spent 10% more raw mana to gain 10% more healing, but in reality, across the time it took you to cast those 11 spells, you reduced the amount of time you spent OoFSR, meaning you actually spent more than 10% more mana, as you received less than you would have if you cast those 11 spells without haste.
This is where I was talking about "time dilation." It doesn't effect all things equally. The FSR is always 5 seconds, regardless of your haste; and since you're using haste to squeeze more spells into a given time period, you're reducing the amount of time you're in OoFSR. Again, however, if you're keeping your OoFSR time constant, that means regardless of haste, you're still landing spells at all the same times, meaning the effects of haste are practically wasted.
[E] I think I see what you're saying. You're looking at it in more of this sort of model, starting at some spike in damage:
However, again, this is a specific situation. This is the situation where spike damage is constant, and has the same interval between spikes, which isn't always the case. Haste in this case allow you to fill the health deficit faster, giving you more time for OoFSR until the next spike. Yes, this actually does increase your HPM. But this is only one situation, pointing to haste being, again, situational at best. Someone with higher +healing could also accomplish the same thing, by using fewer casts to fill the health deficit, but that, too, is situational, depending entirely on how many targets are needing to be healed. If it's a single target or a single group of targets, +heal might actually accomplish this with greater efficiency than haste; but if it's multiple targets or groups of targets, then the extra +heal is wasted if it is simply turned into overheal.
I was wondering if we could talk about haste vs regen rather than haste vs +heals for a second.
Assuming that +heals are equal, could + haste add to regen?
It appears to me that a reduction in casting time would mean an increase in the interval between casts and thus increase mana regen in the O5SR?
for more clarification..
From what i understand haste in fact increases regen with all things being equal. It increases the time between casts if you let it, and thus increases time in the OFSR.
For instance, assume constant dmg, and assume that Boss deals 25K amount of damage in 1 minute.
And assume that you can get 25k healed in 10 casts of GH.
without haste you have a 2.5 second cast and therefore you have 25 seconds of your 1 minute will be used on just casting heals. this leaves you with 35 seconds in between casts and to get regen. ASsume you wait an equal time between casts, this means you have 3.88 seconds between casts. (9 intervals between 10 casts). Assume no inner focus, etc, and you get 1.388 seconds of OFSR regen per interval and 12.5 total seconds of regen during that period OOFSR regen.
If you have haste that reduces cast times to 2.25 seconds, then you have 22.5 seconds of pure cast times, and 37.5 seconds inbetween casts and therefore 4.16 seconds between intervals, and therefore 1.41 seconds of OOFS regen between casts which means 12.75 seconds more time in OOFSR regen.
The extra regen in .25 seconds can add up significantly over time.
Haste can mean more regen and does not represent a loss in mana for priests. this is the case unlike dps, since as dps, we are a) never outside the 5 second rule, and we don't have the spirit regen that allows such a huge difference between the inside and outside the 5 second rule.
Of course, this means that in order for haste to increase mana regen the difference between IFSR and OFSR needs to be signficiant and that is done first with spirit.
Please be kind if i have completely missed something as i usually just melt faces.
I was wondering if we could talk about haste vs regen rather than haste vs +heals for a second.
Assuming that +heals are equal, could + haste add to regen? [snip]
That's sort of what's going on above you.
If you're using haste to chaincast faster, you'll at best get the same amount of oFSR time, and if you're talking about chaincasting GHeal you may actually get less (since cast-clearcast-cast can give you a tick of oFSR time, particularly if you're lagging a bit, but if your GHeal cast time drops much below 2.5s you stop having that option).
If you're just casting a heal every N seconds, even though this is highly unrealistic (bosses don't do predictable damage, in general):
0s: start GHeal
2.5s: finish GHeal
7.5s: out of FSR
8s: tick
10s: tick, start GHeal
12s: tick
12.5s: finish GHeal
versus
0s: start GHeal
2.4s: finish GHeal
7.4s: out of FSR
8s: tick
10s: tick, start GHeal
12s: tick
12.4s: finish GHeal
The only time you would gain oFSR time is if the tick fell between 7.4 and 7.5s, and again this is a fairly unrealistic example. But more importantly, in this example the haste isn't actually increasing your HPS; you'd be far better served to get +healing instead.
More realistic is the idea that you could burst-cast more quickly and then regen longer; say, in a raidwide burst damage situation like Naj'entus. But my understanding of Sunwell encounters (disclaimer: my only experience there was a couple of 95% wipes on Kalecgos) is that those aren't common; Illidan's Flame Blast is the latest equivalent, and phase 4 is a regen break for whoever's not on the warlock tank anyway.
Off topic... but thought I'd point out something you missed in your primary post, Const. You have the two mana oils listed, which is great, but in all honesty post 2.4 I don't use mana oils at all (even pre-2.4 I only used them on mana intensive fights). Instead I use Superior Wizard Oil. I know it says "Increases spell damage by up to 42" but it counts on healing as well. I'm always in a spriest+resto sham group, so I've found I can focus greatly on +heal and just chug mana pots. Raid buffed I'm around 2.7k, that's with Fishsticks, Wiz oil, heal pot, wisdom. Add in a few procs and I hit 3.2k pretty easily, and that makes for some insane CoHs. I really enjoy 1.0 second CoHs on RoS (Heroism+153 base haste rating) at over 2.7k heal.
Granted all consumes depend on group makeup and current gear, but for those of you that aren't having regen issues I highly suggest Wiz Oils.
The fundamental problem I see with haste is that it is intrinsically a negative-regen factor, at last in Sunwell. If it were the case that I could top someone up faster using Haste, and possibly squeeze an extra second or two out, putting me OO5SR for a tick, then it would be amazing. However, it is rare for that to happen with the style of fights we are seeing in Sunwell so far.
Kalecgos: if you are "up top", you can always heal the current dragon tank. Getting OO5SR ticks isn't that hard. "Down Below" it's fairly simple to top-up and then regen; if we can do it without haste, haste will only benefit this (slightly).
Brutallus: it's chain-casting with some cast-cancelling, haste isn't going to effect regen time at all over the course of the fight, since you almost never *stop* casting, and cast-cancel works the same way with or without haste.
Felmyst: Phase 2 is regen time, who cares.
Twins: chain-cast for 5 minutes, have fun. There is practically NO OO5SR time on this fight whatsoever. If you've done the fight, you know what I mean ... the damage is constant, steady, and intense.
M'uru: running with 6 healers, you might get 3-4 seconds every minute in which you could not cast. However, it's not guaranteed, and there's usually something you should be doing with that time (Renew, PoM, etc.) -- this is not a OO5SR fight.
So as much as I'd love to be able to say that haste = more time OO5SR, it really doesn't seem to hold true in practice. Perhaps on some fights in Black Temple / Hyjal it would -- I can definitely see having haste being valuable on Naj'entus, Mother Shahraz, maybe Council (depending on your job), and possibly the odd Hyjal fight ... but none of those are fights where we really care about getting those OO5SR regen ticks anyway.
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
Any time you're doing a haste argument, just acknowledge that we have 3 primary spells: Renew, Circle of Healing, and Greater Heal.
"We" - who's "we"?
I have Greater Heal, Flash/Binding Heal, Prayer of Mending, Renew, Circle of Healing and Prayer of Healing as my primary spells. CoH and PoH depending some on spec and fight.
M'uru: running with 6 healers, you might get 3-4 seconds every minute in which you could not cast. However, it's not guaranteed, and there's usually something you should be doing with that time (Renew, PoM, etc.) -- this is not a OO5SR fight.
Doing the fight now, healing Sentinel tank all p1 + assisting on one side. I'm on spirit regen between EVERY Sentinel and end p1 on 80-90% mana. 6 healer setup btw.
Edit: No spriest, chain-potting, if you wanna know.
Our setup doesn't seem to be covering raid heals all that well, and I almost always have to toss off CoH during my down time, or people start spiking dangerously. We're going to tweak it some more tonight ... I'm tired of watching my mana go dooooown.
As far as the "we" comment above: let's be realistic. No-one specs or gears to make their Binding Heal more efficient. It's a nice spell, and we all use it, but I'm not basing my gear choices around it. I don't use it THAT much. I also use PoM a lot ... but it doesn't scale that well, and I really don't care to base my gear choices around it.
You consider the spells you use a *lot* that are effected by your gear choices. CoH is huge, because of how it scales with haste. GH is huge, because of the setup with Spiritual Healing, a 3.0 base cast time, and Empowered Healing. And Renew is ... Renew.
If you look at total healing done over the course of the night, and over 25% of your healing done is PoH+Flash Heal+Binding Heal combined, I think something is wrong with your approach to healing in a TBC raid. In a 4 hour Sunwell raid through the first 4 bosses, my breakdown (including trash) was (10 million effective healing done, total):
CoH+GH+Renew = 75% of my effective healing for the night. Those are the spells I'm going to focus on. And I haven't seen too many WWS reports that have anything different. In fact, some people have much *lower* Flash Heal usage. Most of mine comes from trash, where I just mindlessly spam as we kill equally-interesting-yet-no-loot-dropping-mini-bosses.
If you want to link some WWS reports where you have a significant chunk of your healing done by PoH or Flash Heal or some variant thereof, I'd like to see it. Almost no-one uses Flash Heal extensively on bosses - it's way too inefficient.
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
Something important to consider is that haste does not mean you cast for a longer time it means you cast more spells within the same time. However because haste at a certain level increases your HPS more, you actually get MORE healing out in that same time, than you would if you had +heal. What does this mean? It means you can find time to regen where previously you had none. It means that when you get a clearcast, its MUCH safer to cancel and start your cast again with max rank gheal, or you can delay your next cast just a little longer, because your higher HPS will guarantee that you can chase the damage up.
Overall of course I would say that the effect of haste on your mana regeneration, is going to be small enough to be of no importance. An extra OFSR tick here, or one less there is not going to change things much. What is important however is the increase in mana consumption. Because you achieve the higher HPS by casting more spells within the given time frame, you will spend more mana, also unlike +heal your HPM stays the same, which means that the maximum healing output is reduced.
Another important question to ask is when do you actually need that higher HPS. How often is it that you find yourself spamming max rank gheal with abandon and having it just race the incoming damage.
Haste however has additional advantages. Faster casts means you heal more targets per given time. It also means that its easier to save a DPS from a vicious special ability that ticks for craploads very fast. It also means you can apply your instants faster. With 1.25 sec GCD for example it takes 1.25 seconds to set up a PoM/PWS combo and only 3.25 seconds to land your gheal after this. This is somethign that +heal cant give you
Our setup doesn't seem to be covering raid heals all that well, and I almost always have to toss off CoH during my down time, or people start spiking dangerously. We're going to tweak it some more tonight ... I'm tired of watching my mana go dooooown.
As far as the "we" comment above: let's be realistic. No-one specs or gears to make their Binding Heal more efficient. It's a nice spell, and we all use it, but I'm not basing my gear choices around it. I don't use it THAT much. I also use PoM a lot ... but it doesn't scale that well, and I really don't care to base my gear choices around it.
You consider the spells you use a *lot* that are effected by your gear choices. CoH is huge, because of how it scales with haste. GH is huge, because of the setup with Spiritual Healing, a 3.0 base cast time, and Empowered Healing. And Renew is ... Renew.
Yeah, make better assignements and stick to them. We've had holy priests in several roles on the encounter, all leaving us to regen those small seconds inbetween making us heal forever.
Regarding "most used spells" - if you read closer I wrote: Flash/Binding Heal. A 1,3 sec heal hitting for 3k on an amped tank is one of my primary heals, yes (remember, not me having mana issues ) Binding Heal is huge any time we take damage and can't count on others - ensuring that we are the healer with the best survability and ability to heal others while under heavy damage.
I don't understand this statement:
"The fundamental problem I see with haste is that it is intrinsically a negative-regen factor, at last in Sunwell."
How is haste a negative regen? I read that statement meaning all things being equal the more haste you have the worse your regen is?
If you mean mana/ time, then that would ONLY be the case if you are casting more spells over that same time, which would presumably mean you are healing much more over that same time.
I just saw the post above me making the same point. Increased haste gives you the flexibility of either
1) casting more heals within a timeperiod
2) casting the same amount of heals over the same period, but with more regen.
It is true that you will have less mana in scenario 1, but that is only because presumably heals were needed. If they weren't needed, then you would be in scenario number 2.
Const you make the point that you aren't OFSR much in sunwell, but that still doesn't mean that haste is causing you to lose regen, it just means you aren't able to take advantage of the increased regen from haste. (ie. you don't lose anything from haste, you just don't gain regen, though you do gaini more heals in that same period).
It seems to me that if you are constantly spamming heals in Sunwell, then haste isn't just good, but is required since you can then fit more heals in.
I know that you don't get haste for free, but still i think i am missing something when people say, "haste is negative mana returns."
I don't understand this statement:
"The fundamental problem I see with haste is that it is intrinsically a negative-regen factor, at last in Sunwell."
I disagree with the statement. I'm now at 190 haste (12%) while still having 1190/439 mp5. That makes it easier and more viable to cancel-cast on tanks. Actually I feel I'm in more control over healing and manaregen the more haste I get (while still having "enough" regen and healing power).
I'm not saying I heal reactive with 190 haste, but I can allow myself to be just a little bit more reacting to damage, rather than being on the safe side and always let my heals land. It's of course impossible to value this in mp5 though.
I think its important for anyone considering haste to disengage it from mana regeneration. Past 2300-2500 healing haste starts producing double and tripple (for PoH) the HPS extra +healing does, so it is the way to go, as extra +healing no longer results in significant % increases. There is realistically no other way to go forward BUT haste.
Haste benefits the following priest spells in decreasing order
PoH
gheal
Flash/binding heal
CoH
It also strongly affects the burst one can get from instants, as it has a big effect on the time it takes to apply instants and follow with a chanelled spell. This makes it easier to apply PoM, PWS and Renew in between spellcasts, thus still increasing overal HPS despite the fact that these spells are not directly affected.
A side-effect of haste, is that the higher HPS comes with an increased burn rate.
However it needs to be kept in mind, that no one maintains max HPS for any length of time during any raid encounter. Hence the extra HPS from haste, is not used all that often. For spells like CoH, which are AoE and hence have lowish HPS per person haste is very important because you want max HPS every time you use it.
Yet the decreased cast time itself is also a very important property. Faster heals are safer and easier to use and gheal still has a 2.5 second cast time, which is in fact quite long given that you want to use it as much as possible. Reducing the cast time of gheal should be a priority for any priest whose +heal value is already well above 2k. Further increases in +heal are counterproductive due to low % increase and the added utility of faster heals is a highly important and desirable property.
My oppinion is that stacking a certain amount of haste along with high +heal values (2.3-2.5k) will quickly become the optimal build for priests in the t6 level and beyond, not simply due to the higher maximal HPS, but mostly due to the increase utility of faster casts and especially instant casts.
For the record flash heal max rank has higher mana burn rate than gheal max rank. Nearly 30% more. Someone with 700 mp5 total regeneration is still bleeding mana at 900mp5 from spamming flash heal max rank and will go oom in under a minute. It is physically impossible to maintain more than 900 HPS with flash heal. It is not really possible to cast more than 1 flash heal every 2 seconds and still have mana after 4 minutes, unless you get innervates.
For the record flash heal max rank has higher mana burn rate than gheal max rank. Nearly 30% more. Someone with 700 mp5 total regeneration is still bleeding mana at 900mp5 from spamming flash heal max rank and will go oom in under a minute. It is physically impossible to maintain more than 900 HPS with flash heal. It is not really possible to cast more than 1 flash heal every 2 seconds and still have mana after 4 minutes, unless you get innervates.
I share your views on haste, but is the quoted part correct?
FH9, mana cost 470, cast time 1,5 sec: 313 mana/sec
GH7, mana cost 701 (talented), cast time 2,5 sec: 280 mana/sec
If my math is not flawed, that is 11% faster burn on mana casting FH9 over GH7 (it is late and a few years since school so feel free to correct me).
I'm not a big fan of Flash Heal, but I prefer Flash Heal on raidhealing over downranked Greater Heals. With a combination of PoM, Flash Heal, renew and well-timed Greater Heals, an IDS-priest is able to keep an impressive high HPS given a few mana pots I find myself only casting downranked Greater Heals on amped tanks when I have a very good idea of what kind of damage to expect.
We swapped the healing assignments around tonight, and instead of solo healing the Void Sentinal tank, it was myself + resto druid. And with massive consumable use, yes, I had mana to spare. I was burning every single cooldown I had, but the tank was alive, and I had mana on transition to P2. Which is significantly better.
As an aside, I don't think it's functionally possible to solo heal that tank and reliably keep him alive for all of Phase 1. It's just too spiky and constant. Screw Sunwell Radiance!
As far as a negative regen-factor comment above, I'm still sticking with it in theory, anyway. The faster you can cast, theoretically, the earlier you can get OO5SR. In all practical cases, you might gain 1-2 seconds, however, and you're going to cast another spell in the middle, which means more mana used. Yes, your healing is higher; that's never been a question. The issue is whether or not it is sustainable on a true nuke-heal fight, without a shadow priest.
It really depends on group composition and setup, imo. On M'uru I'm running without a shadow priest and without Mana Spring. That's a solid 300-350 Mp5 that I'm missing. If I had that, I could basically do whatever I wanted; spam max-rank Greater Heal, stack haste, dump regen; it all becomes meaningless compared to that incoming mana.
I'm not denying that haste is valuable, and I'm not denying that it is the most effective way to scale your HpS. What I am challenging is the idea that it actually is a positive regen-effect. I just don't see it from practical healing. Unless you're doing a job and know you're going to have blocks free, and those blocks can somehow be made larger by you healing faster (which implies that your healing rate directly influences the incoming damage, which it doesn't, ever) ... it's not going to be a net regen gain. You may be able to say that it's neutral, but I think in the end, picking up haste means accepting that you are going to be doing more healing ... but it's obviously coming at a toll to your mana pool.
[side note: good discussion, nice to see some new faces jumping into the thread]
[e] I was bored, so I threw together my "end-game" ideal gear set. It's not quite perfect yet, but ... critique away. charDev Profile
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
We swapped the healing assignments around tonight, and instead of solo healing the Void Sentinal tank, it was myself + resto druid. And with massive consumable use, yes, I had mana to spare. I was burning every single cooldown I had, but the tank was alive, and I had mana on transition to P2. Which is significantly better.
As an aside, I don't think it's functionally possible to solo heal that tank and reliably keep him alive for all of Phase 1. It's just too spiky and constant. Screw Sunwell Radiance!
In all fairness I also had lifebloom supporting, which is basically the reason I can get so much time to regen. You don't need to heal the last hit from the Sentinel (let LB do the job, you're just avoiding overhealing by not healing it) and then with some clever use of trinkets/Inner Focus etc, I can keep my mana almost full all the time and then being able to throw in PoMs/Renews/PW:S on other targets when needed.
Originally Posted by constantius
CoH+GH+Renew = 75% of my effective healing for the night. Those are the spells I'm going to focus on. And I haven't seen too many WWS reports that have anything different. In fact, some people have much *lower* Flash Heal usage. Most of mine comes from trash, where I just mindlessly spam as we kill equally-interesting-yet-no-loot-dropping-mini-bosses.
If you want to link some WWS reports where you have a significant chunk of your healing done by PoH or Flash Heal or some variant thereof, I'd like to see it. Almost no-one uses Flash Heal extensively on bosses - it's way too inefficient.
We don't use WWS so I can't prove to you that I use Flash Heal on a number of encounters. So I guess you just have to belive me
Typically you have to choose between a downranked Greater Heal or a Flash Heal if you want to top up someone. Yes, it will a bit cheaper to do it with Greater Heal (even more if you go around with 2p of tier5 obviously), but I keep seeing the same thing when i take a close look at my recount after raiding nights: Overheal on Greater Heal is a lot higher than on Flash Heal. It's just easier to avoid overheal with a fast heal -- if you have 40% overheal on Greater Heal and 20% on Flash Heal, what's more efficent? There are a lot more ways of seeing efficency than just looking at HPM in the spell book.
For me it comes down to that I want to perform as best as possible. I prefer toping up my target as fast as I can and then a 1,3 sec heal wins over a 2,2 sec heal when they heal for the same. Ok I may have lost 100 mana (or whatever it is), but all of a sudden I now have 1 second I can sped on throwing a PoM on the tank which will bounce five times the next 15 seconds. A priest who downranked a Greater Heal may not have time for that PoM and just "lost" 10k healing from PoM which I got. If I run into mana issues I obviously start to downrank and play "smarter" with my mana, but I prefer not to.
As you understand, this isn't exactly theorycrafting, but more of a playstyle. You gotta do something to outheal those overrated Shamans
Typically you have to choose between a downranked Greater Heal or a Flash Heal if you want to top up someone. Yes, it will a bit cheaper to do it with Greater Heal (even more if you go around with 2p of tier5 obviously), but I keep seeing the same thing when i take a close look at my recount after raiding nights: Overheal on Greater Heal is a lot higher than on Flash Heal. It's just easier to avoid overheal with a fast heal -- if you have 40% overheal on Greater Heal and 20% on Flash Heal, what's more efficent? There are a lot more ways of seeing efficency than just looking at HPM in the spell book.
I don't see overheal as a problem, really. Certainly you want to avoid healing more recklessly than your regen can handle, but it honestly takes me more brainpower/latency to determine which is the best heal to top someone up, whether they're going to take more damage shortly, and whether they're in CoH radius of this other person than it's worth to Just Do It.
I can't speak to the M'uru example in particular, but there aren't a lot of situations I've seen where you can say "this person needs 2k healing quickly but will not need 3.5k/5k healing a bit later", and certainly not a lot of those situations you can see coming in advance. (Obvious counterexample: Naj'entus.)
I also don't see "beating the shamans" as a reason to do things differently. Boss dies, everyone goes home happy. If the shamans destroy everyone by 5%, so what?
I share your views on haste, but is the quoted part correct?
FH9, mana cost 470, cast time 1,5 sec: 313 mana/sec
GH7, mana cost 701 (talented), cast time 2,5 sec: 280 mana/sec
If my math is not flawed, that is 11% faster burn on mana casting FH9 over GH7 (it is late and a few years since school so feel free to correct me).
I'm not a big fan of Flash Heal, but I prefer Flash Heal on raidhealing over downranked Greater Heals. With a combination of PoM, Flash Heal, renew and well-timed Greater Heals, an IDS-priest is able to keep an impressive high HPS given a few mana pots I find myself only casting downranked Greater Heals on amped tanks when I have a very good idea of what kind of damage to expect.
No it was not correct, I had an entirely wrong thing in my mind and I was using a slighty incorrect mana value for FH. Your calculations are correct.
As for the question: if you have 40% overheal on Greater Heal and 20% on Flash Heal, what's more efficent? The answer is simple. Flash heal @2k +healing has a mana efficiency of 5.3 with 20% overheal overall HPM is 4.24. Gheal has an average mana efficiency of 8.2, with 40% overheal its 4.92. Gheal is still 16% more efficient than flash heal. It takes very high overheal value to start swinging the balance.
Yet the conclusion is unchanged. Flash heal has an unsustainably high burn rate. Observing my fellow fh using priests I can see them pacing themselves. They cast an FH, select a target cast another FH and so on. Anyone chaining FHs burns through his mana in 2-3 minutes. With my current regen, if I chained FH I would go oom in 45 seconds. When you take that into account and you add overheal into it its pretty obvious why its not possible for FH spam to surpass 900HPS. Rank 3 gheal on the other hand you can easily chain cast, so rank 3 produces HPS much closer to its maximal value.
The problem with flash heal is that the high burn rate, leaves you less mana for AoE and instant spam, casting time is less important than overal HPS. IF you have to heal 5 people each with a 5k deficit and you tried flash heal it would take you twice as many casts as with gheal. Then again it depends on your raid role. If you are MHing the tank, you can't take 2.2 seconds to heal someone else I use FH too or more often an instant if its imperative that I heal another target when I am on the tank.
The question is when is it better to use gheal over FH. I go through a series of mental checks before pressing the FH button. Is the damage active? Is the deficit low? Is the target in immediate danger? Do I have limited time to cast my spell? If am on a hurry to move on and the damage is not active, I will flash heal. If the target is in immediate danger and the damage is active, instant spam. Otherwise if the deficit is low renew. If its high gheal(3-6). I find that I am casting gheal3-6 most of the time and renew and flash heal very rarely. For me CoH+PoM combos pretty much replace flash heal. This is the reason why I use gheal a lot more. It gives me time to think and assess the situation.
Overall I find that FH usage is a function of healing playstyle and damage type and how much healing is available in relation to how much damage is done.
Anyone know addon for Grid that shows (pulse or something) when your heal *lands*? Would be great for CoH to see little square blinking in the corner when and where heals land.
Couldn't find anything like that in wowace.com, anyone who could make such addon? ^^
As for the question: if you have 40% overheal on Greater Heal and 20% on Flash Heal, what's more efficent? The answer is simple. Flash heal @2k +healing has a mana efficiency of 5.3 with 20% overheal overall HPM is 4.24. Gheal has an average mana efficiency of 8.2, with 40% overheal its 4.92. Gheal is still 16% more efficient than flash heal. It takes very high overheal value to start swinging the balance.
Yes, if you strictly talk about efficency as in HPM, Flash Heal will lose out to downranked Greater Heals (r1 and r2 namely if we're talking about healing 3k). I have a hard time beliving that Sunwell priests use GH1 and GH2 a lot, if we're "forced" into healing someone for 3k, then at least I want that job done as fast as possible so I can go back to better spells in terms of both HPS and HPM. (Just as a note: Most of the time I would leave 3k "unhealed damage" to CH, as a Shaman is capable of doing it a lot more efficent and it would be on their assignement to do.)
IMO efficency can't be only about HPM, then it's better to actually say HPM and we know what the point is. To me it's more "efficent" to spend 1,3 sec doing the same job you spend 2,2 sec on (given my haste) and then have 1,0 sec I can regen/heal more. Even if I've spend 100 more mana.
Yes, Flash Heal has a low HPS and I'm in no way "a Flash Heal-spammer", but the fact remains, if 1 target is 3k away from 100% hp and you want to get them up, Flash Heal is the best choice if you care about overall HPS.
Originally Posted by Havoc12
With my current regen, if I chained FH I would go oom in 45 seconds. When you take that into account and you add overheal into it its pretty obvious why its not possible for FH spam to surpass 900HPS.
A combo of Renew, PoM and Flash Heal can give 2k HPS. I've dont it a number of times. Sure you burn mana fast, but don't say it's impossible to surpass 900 HPS - hell, that's easy
Basically I don't really understand why people are so afraid of using Flash Heal and instead downrank so much. I also play a lot around the 5 sec. rule and get as much spirit regen as possible, but if I see someone in danger I don't blink before I throw them a fast heal inbetween my other tasks if I have time for it.
Again, I don't think we disagree that much - at least not regarding haste and how it effects us.
05/08/08, 5:12 AM
.
Similar Threads
