No one have done the maths for stats equivalence for pallys?
... i really cant understand why all classes have a "consensus" about stats equivalence, besides holy pallys.
I think you will find there isn't really a "consensus" for other classes either. Stat equivalence is just broad generalization and fairly useless. There are too many factors that vary.
The universe is run by the complex interweaving of three elements. Energy, matter, and enlightened self-interest.
In fact... for DPS there is little to no variation. X of each stats gives you Y of DPS so thats all that matters.
For healers its more complex since you want to balance Mana efficiency with HP/S, and this two stats have no simple mathematic relation.
But all other healing classes have stats equivalence.
A 22 Heal gem is better them a 10 crit one? Or a 11 heal 5 int?
Im sure you can prove with the right math, that one is superior to the others.
Not in all cases, but for the typical ones:
You need to varie party composition, and fight leght and calculate the best stats for that.
The only real problem is how to create a mathematic relation between Mana efficiency and HP/S.
Edit:
Btw, if you dont believe in a stats equivalence, them how you gear yourself?
Why some choose Heal, others Crit or Mp5?
Probably cause they think that stats worth more them the others...
Having "some of everything" is useless... (like people that gem 4MP5 on Girdle of Hope cause "It lacks MP5, i gem MP5 it becomes a great belt!")
Actually efficiency is not hard to calculate (although not easy either, the thread explaining it seems to had been deleted so I can't refer to it).
HPS is not hard to calculate, either.
However the stat equivalence you'll get from calculating HPS will be extremely different than the stat equivalence you'll get from calculating efficiency. So while you could make a spreadsheet that will tell you how much burst and how much efficiency every item gives you, choosing which item is actually better will depend on how much you really need that extra efficiency and how much you need that extra burst HPS. On addition the amount of mana regen you get (mostly from shadowpriest vs no shadowpriest but also from resto shaman and gear) will make a big difference on your stat equivalence when looking at efficiency.
If you look at other healer threads you'll see they don't have agreeable stat equivalences either, for the same reasons as here.
It is easy to prove, however, that 22 healing is better than 10 crit rating or 11 heal 5 int. It's even probably true in pretty much all cases that 18 healing is better than 10 crit rating. 18 healing vs 11 heal 5 int though depends. If one stat provides both higher efficiency and higher HPS then it's obviously better, however in many cases you'll be comparing a stat that gives higher HPS with one that gives higher efficiency (say, haste VS +healing, or 18 healing VS 11 healing 5 int in a no-shadowpriest situation).
Why some choose Heal, others Crit or Mp5?
When people have no clue what stat is best, they go for whatever feels right, which is 99/100 wrong, but also very often (as long as your choices aren't totally dumb) more than good enough to get you all the way through the game. Even in some cases of totally dumb choices you'd still manage.
You'll see even on the mage/lock forums, where it is quite clear how good every stat is compared to others (with a very small margin of error), you'll see the odd "I like crit" mage or "haste>>all" (it's good but not really noticeably better than other DPS stats if at all better) warlock with all kind of wierd arguments that make no sense. Guess what if they focus in raids they'll probably still clear the game anyway, but their DPS will be quite inferior.
No. Crit multiplies the amount of healing you can do with your mana. Casting faster by iteself doesn't make you able to heal more with same conditions, so crit doesn't give you any more total healing done in the fight when you have more haste. That's why your whole crit->mp5 conversion is just plain wrong (it greatly overvalues crit if you assume it's for HL spamming or undervalues crit if you assume it for pure FoL spam).
Casting faster does allow a bit more efficiency since you'll less often run into needing to use less efficient spell, but that effect is much much smaller than your "10% haste -> 10% more gain from crit".
Another way to explain this is that you can't assume you chaincast non-stop yet still care about mana. Those 2 situations cannot exist at the same time, as if you need mana you already cannot cast that spell 100% of the time and if you already have the mana to cast the spell 100% of the time additional mana will not change it, meaning your whole "I use X mana/sec and get Y of it back per 1 crit rating and therefore 1 crit rating is worth X*Y MP5" is completely wrong.
I see that we always make different assumptions and look at things from a different point of view. I agree that haste itself won't let me heal more with my mana. However this is irrelevant information for me as a raid healer. I don't really care how much I can heal with my mana, I am interested in maintaining certain lvl of HPS through the fight. Question about efficiency ALWAYS comes after the question about HPS. If you can't maintain HPS that is required from you, your efficiency doesn't matter. Now, haste is more or less centered around HPS. As you mentioned, it doesn't really increase efficiency and I agree with you. Now, imagine that my spell pattern is fixed and nothing disturbs it. Will 10% haste increase extra HPS from crits by 10%? Yes, it will. As it will increase extra HPS from +healing by 10%. Will 10% haste increase MPS gained from crits by 10%? Yes, it will. I DO receive 10% more HPS and MPS from crits. Will it influence my mp5 stat? No, it won't.
OK, now my main point. How we should compare mana returned from crit and mp5? From HPS-centered point of view both are only means to increase the time over which you can maintain your HPS (till oom basically). Let's say without any haste 1 mp5 will make my required HPS longer by x sec and 1% crit makes it longer by y sec. With haste 1 mp5 will increase my time by x1 sec (obvioulsy x1<x) and 1% crit will increase my time by y1<y. I am now claiming that y/x < y1/x1. That means that crit effectiveness scales better with haste for me.
And once again return to your words:
so crit doesn't give you any more total healing done in the fight when you have more haste
Your statement is completely wrong unless you change it into "so crit doesn't give you any more total healing done with your mana when you have more haste". You don't take into acount fight length and the amount of mana I have during the fight. If I always do have half mana bar after some certain fight, I clearly would benefit from haste that allows me to burn it faster (and replace 1 healer by extra dps for example). Lets imagine after adding 10% haste I would still end at 10% of my mana bar. In this situation crit will add more total healing done in the fight after I added haste. Since part of that extra mana that I spent was causing crits that actually increased total healing done in this fight.
Also in RoS P2 maximal available mana goes down faster than paladins could burn it (I don't consider stupid mindless HL spam). In this situation crit surely would add to total healing done, since it's not your mana determines how much you can heal but the rate at which your mana pool shrinks.
P. S. I valued my crit using my averaged T6 wws graphs. So that number means that adding 1% of crit would give me back about as much mana as X mp5 added for an averaged T6 fight. I don't see why it shouldn't be a reasonable emphirical estimate of crit vs mp5 value for me.
I never said haste was bad, I said haste was bad for efficiency. Obviously you need a certain level of efficiency (total healing done with your mana in a fight) and a certain level of HPS. If you fall short on *either* of those you will fail the fight. In most normal cases you will have enough of both, really, so it's just a matter of what's more likely to go wrong - you needing to do some extra heals and using more mana because someone took an un-needed doomfire or you needing to pop some bigger burst because too many healers took watery graves at the same time.
You can't look at the value of efficiency if you have mana left at the end of the fight. If you don't use all your mana, increasing your mana will add nothing. Adding more mana efficiency will only help when you actually have something to do with that mana, or in other words finish the fight dry, or at least have a chance to finish it near-dry if things go reasonably bad (unreasonably bad would be a wipe anyway). It's up to you to define how much efficiency is required in a fight. Almost exactly the same kind of arguments can be repeated concerning HPS.
Now once you're already using all your mana, having more haste will have little effect on how much healing you did with that mana over the fight, and thus the benefit of crit to your efficiency will barely change compared to a no-haste situation with same gear.
Also crit doesn't increase "burst" HPS as by definition you need that burst in a worst-case-scenario and by definition on that scenario you're not going to crit (becuase if you did crit it's not a worst-case scenario). It's like a tank stacking dodge over stamina - it reduces your average damage taken but doesn't reduce your ability to handle worst-case scenario, same way crit increases your overall healing done if you use all your mana but doesn't increase your ability to handle a worst-case-scenario.
You're assuming you need to maintain a certain level of HPS over the entire encounter. This is both not favoring crit (bad no-crit string and you fail so crit doesn't help get through the fight in terms of HPS requirements) and not even true to start with. Most fights require some pretty modest HPS for the most part and have periods during which you need higher HPS. The "HPS I can maintain over the fight using all my mana" is completely irrelevent and even if it was relevent, it wouldn't change anything about how you looked at haste, as you just used the same amount of mana for the same amount of healing meaning haste did nothing.
I'm not saying haste is useless, not at all, but you need to look at what it actually does and what it doesn't do. Haste increases your max burst and gives a tiny efficiency increase (doesn't increase healing done with same mana but allows a slightly more efficient spell distribution thus saving SMALL amounts of mana per point of haste). Haste doesn't make crit better because you cast more spells/sec nor becuase you use more mana/sec, because at the end of the day your available mana for the fight is NOT dependant on your haste.
Ignoring HPS->MP5 benefits (which you should take into account if you want to accurately make calculations) but good enough to show how it works, the healing you can do over a fight if you use all your mana is:
(mana without crits) * (HP/mana of your heals) / (1 - 0.6*crit)
As you can see this does not depend on casting speed.
If you still have mana left at the end of the fight and want to use it, remember nobody can spam rank11 HLs for the full fight duration and not go oom. This means you could've casted more spells than you did without going oom, and haste wasn't needed to do that.
However if you lost someone due to not having enough burst, yet had mana left at the end of the fight, increasing HPS at the cost of efficiency would be what you want. But start by increasing HPS by using higher HPS spells (more spells, higher ranks of HLs, more HLs over FoLs) before resorting to gear changes that will end up losing more efficiency per HPS gained than simple spell changes would. "Only" where max rank HLs fail is where haste (and healing) over mp5 (and healing) wins. With how fights are made nowadays, though, that "only" seems to be quite important and that's why haste doesn't suck at all.
On a side note evaluating crit based on WWS is compltely flawed, as it doesn't take into account the extra mana you would've gained if you actually used the mana you had left at the end of the fight. Would you have used that mana, crit would've returned more than WWS showed, making crit better than what it looks like on WWS (if you didn't use all your mana, if you used all your mana then WWS will show exactly the same as theorycrafting would). And if you had a noticeably amount of mana left at the end why do you care about how much extra mana crit gives?
For healing output increase of haste rating vs healing based on current amounts of the other (average values of max rank FoL and R9HL used), BoL up and the [Libram of Souls Redeemed] equipped:
Started off using my usual stats: (I'll call this normal, and it's slightly unrepresentative since I'm pulling 130 mp5 unbuffed)
HasteR=0
+Heal=2000
CritChance=.25 (Crit HL = .3)
BoL=yes
Libram=yes
Haste Constant = .01/ 15.76 = .0006345
Then, I compared the effectiveness of +Healing vs Haste vs Crit rating in terms of HPS ONLY. This in no way takes in healing efficiency, or any of that stuff. I took the epic gems equivalency of 10 crit rating = 10 haste rating = 22 Healing (although it is suspected healing is actually 22.5 due to blue gem equivalencies). I multiplied these values by 5 to increase effect. (50 rating = 110 +heal)
SCENARIOS are as above, but with one of the three stats added.
As you can see, Haste is better than +heal for both spells' healing output, but more so for HL than FoL. Considering that haste also decreases your mana efficiency by the same % as it increases your output, I would recommend stacking it >>only if you're safe for mana consumption and badly need to increase your healing from HL<<. Otherwise, + heal seems better. Also, this shows that crit is >bad< for healing output, as it's about half as effective than +heal for HL, and 1/3 as effective as +heal for FoL.
ON ANOTHER NOTE: Intellect is almost as good for healing output as +crit for FoL.
since +50 int = 50*1.21 (BOK and the 10% talent) * .35 +Healing
And 50*.014 crit % = .7, not to mention the increase to mana pool,
And since our efficiency goes up geometrically with mana pool (illumination as well as mp5 on current and illumination mana), intellect is a very worthwhile "balance" stat, although geared more toward mana efficiency / staypower rather than thoroughput.
----------------------------------------------------------------
Additional notes (not bothering to put up related data)
- a mix of two stats produces approximately the average of the two effects by %, meaning 25 haste, 55 healing produces 3.131%, etc.
- If you use FoL a LOT, and finish fights with <10% mana generally, stack +heal to improve healing output. If you're finding yourself high on mana, use [Scarab of the Infinite Cycle], and stack haste / healing mix.
- If you're using HL a lot, I'd say go with haste / int / crit. This increases your efficiency, haste is your best bet for output, and most importantly, quicker HL's means faster target switching/ quicker heals, which in my experience has been a keystone of healing.
- With a higher set of gear (2100 healing, 30 crit, 60 haste), the same 50/50/110 increases provided similar % increases. Notably, haste became slightly better (3.45% vs 3.28), the others were within .05% of the values above. At a lower gearset, not that it matters much, healing, went up in value slightly, haste went down in value slightly. In both cases, crit was just as bad.
- The breakpoint of haste > + heal (FoL) for HPS is around 1700-1750 healing, before that, +heal is better. This means that Pre-T5 geared paladins should not go for haste, rather + heal to improve their FoL
- Libram of Souls Redeemed / BoL Missing in effect dropped the data to that of one step lower on the gear, so % increases from stats became like that of a lower set of gear. This means without BoL / +Souls Redeemed, +Healing actually ties haste in output for FoL.
Based on your math, you can reach a stats equivalence to HP/S.
If you do the same to HP/M, (using total mana pool as base), and them compare the value of each attribute in both "charts", we may be able to reach a descent stats comparison to holy pallys.
I really dont know, why no one tryed this yet... or everyone failed miserably?
There had already been a spreadsheet (by Darion) running around to calculate efficiency and a thread I posted (that seemed to had been deleted) explaining how to adjust it to take into account estimated crit overhealing as well as HPS->mp5 conversion. It also included an explanation how to calculate your burst HPS you have available to save people.
You have to remember that increasing the HPS of FoL doesn't help (much) to save people from dying, it only saves you from using less HLs and thus providing very small efficiency. The only (and important) situation where haste helps is when someone takes large amount of damage and even your max rank heal has a hard time catching up to the damage and you want to have the max possible to safely save him. In any other case you can just use a higher rank instead of gearing more haste, which would give a lot bigger HPS increase per efficiency loss than gearing haste. Crit does nothing in such a situation as if you run into those situations where it's so borderline wether he lives or dies that you need to squeeze every bit of healing speed, you want to have the best healing possible when you don't crit so he doesn't die when you don't crit (and not critting happens extremely often even with full crit gear and crit gems).
Bottom line is average HPS calculations are completely pointless, as you never need to put up "average HPS" throut the fight. You need to put some minimal to average HPS (which is generally greatly lower than the HPS of your max rank HL or else fights would've been impossible) most of the time and on occasions some major HPS burst.
While haste will help slightly with efficiency during periods that require minimal to average HPS by allowing use of more efficient spells while keeping the HPS requirement up, it's not much as haste does *only* that. Crit also gives a little bit of that on top of the mana returns and bit of extra effective healing but due to the low effective healing of crit it's just a side effect (which I was taking into account but it's rather small compared to the normal crit benefits). +healing and haste are the only stats that give mention-worthy increase to efficiency through increasing HPS, although the HPS->efficiency is not big. Since +healing already provides decent efficiency by increasing HPM, the additional HPS->HPM conversion makes it an extra-good efficiency stat that also increases HPS by not a lot less than haste does. If you're lacking shadow priest mp5 will still win for efficiency, and haste will always win for burst HPS, but if you feel you need both on an equal level +healing is by far the best as it loses the least of both worlds (yet it doesn't mean it's optimal, as we all know if your burst is too low your efficiency is worth nothing, while if you go oom all that burst goes to hell as well).
If you're going to do pure HPS calculations, they should either be the "how much HPS can by emergency HL do" or "how muc HPS can my pre-casted FoL + reactively casted emergency HL do".
If you're going to do pure efficiency calculations, you'll have to estimate your crit overhealing and have to take into account all raid and party buffs and available gear, as well as how often you're forced to HL to boost your HPS and how much increasing (average) HPS will reduce your need to HL and increase your efficiency.
Average HPS has no use other than a small increase to efficiency by requiring less max rank HLs. It will not make you save people better. Of course having more healing/haste will increase your max rank HL (and precasted FoL + reactively casted HL HPS) which will help you save people, but the actual calculation of FoL and low rank HL HPS has no meaning when it comes to saving people with emergency heals.
Running dry at the end of the fight with higher healing/mp5/int will do more total healing than running dry at the end of the fight with higher haste. And with healing more aggressively you can always use more mana (although it'll naturally go to overhealing if the raid doesn't take the damage...), so "I need haste to do more healing because I have mana left at the end of the fight" is an invalid argument.
A more valid argument for max HPS gearing would be "most of the time my raid doesn't need healing but sometimes you just need to get that tank back up ASAP". That argument works much better as a reason for gearing for max +healing and haste as that's what that kind of gearing helps with. Any other reasoning and you're probably better off with max efficiency calculations.
I believe threads that haven't been posted in for x get archived and are only accessible by benefactors. I'd say this is where your threads gone, so you may have to talk nicely to one of the admin if you want it resurrected.
Holy Librams! appears to have been updated for 2.4 and is a nice way of taking advantage of the 2.4 Libram changes. I’m not sure if the developer have a more official link, but the PTR link has been updated and works fine on live. SourceForge.net: Files
Bubblemasta:
The problem with this sheet is that it overvalue Mp5 over anything else.
And healing seems to not worth really much, same with hast.
It even considers 4mp5 gems as the best possible in every slot.
(It undervalue sockets too, if u delete the sockets, and add 0 sockets but add to the iten the attributes of a gem of that color, you will get a higher "points" iten.)
Another problem is not having a option to add a ShadowPriest or a shaman buffs...
Idk, it seems to lack too much to be really useful for stats comparison...
But there are fights where average healing/s matters. Sure, you're not putting out that healing/s all the time, but HPS matters.
For instance, spirit bolts on Hexlord - FoL / HLx3 or so is a good answer to it, and your performance during the spirit bolts is basically your HPS, which increases with both +heal and +haste.
I do agree that crit and the average HPS values are a worthless comparison, but I just put them there to make sure people don't start trying to debase my stuff with "just gear crit" or something.
Also, I did mention it in my bigger post above, but haste is very useful in keeping up a raid. having a 1.2s FoL with scarab is far far more utility than just a % hps increase. It means you can heal up more targets in a certain timespan in clutch times, such as Naj'entus.
And the difference between switching to HL and dropping more FoL's is usually just 1k damage, it's not like you're going to wait until the tank is 4k under your usual switch limit to switch to HL. Which means a few heals healing for +200 more earlier will help you now. Of course this doesn't affect situations such as massive spike / parrygib damage, but even then, the HL values are a good measure of how good of an improvement in those situations will those stats provide. It is still a worthwhile measure.
I wasn’t endorsing the content of the thread or spread sheet (I stack crit so I’ve been told and proven enough times that I’m wrong). I was simply helping Galzohar find the thread again since he thought it was deleted.
Can you start contributing instead of posting empty and useless comments like this? Its pretty much all you have done so far here and keeping it up will get you banned quickly on these forums.
I'm giving my opnion. If you think my opnion isnt relevant, well not my problem.
I justifyed why i said that:
1- The chart doesnt take into account the minimum HP/S you need to keep a Tank in Hyjal/BT
2- The chart have wrong values for socket slots. If you change them by the actual gem stats, the iten gains "points".
3- It overvalue Mp5. If you follow it, you should be using 4mp5 Gems on all sockets.
I'm trying to do my own math, but im sure they are far from correct. If by any chanse i feel that i find something right i will post here, but im not gonna post anything im not sure about.
The only thing i dont understand i why people keep the Crit/Heal/Hast?/Mp5 fight, and dont use real mathematic models to reach a value for Crit/Heal/Hast HP/S (based on a static value of "gear level") and another model for total healed by your mana pool on a fight of X minutes (based too on a static value of "gear level").
I'm trying to reach that models, if i find they are right i will post them.
If someone with better math want to try, i would be grateful.
Honestly after healing the first three fights in Sunwell, I think haste is pretty much the most important stat paladins should be stacking once they get to that level of gear. The only thing you bring to the raid that other people don't is burst healing on the tank. If you can't sustain the heals that are necessary to keep your tank alive then there is a problem with your healing strategy, most likely. In general as a class we need more ways to dump our mana to keep people alive, rather than ways to increase our longevity, and haste is really the most reliable way to do that at present.
Honestly after healing the first three fights in Sunwell, I think haste is pretty much the most important stat paladins should be stacking once they get to that level of gear. The only thing you bring to the raid that other people don't is burst healing on the tank. If you can't sustain the heals that are necessary to keep your tank alive then there is a problem with your healing strategy, most likely. In general as a class we need more ways to dump our mana to keep people alive, rather than ways to increase our longevity, and haste is really the most reliable way to do that at present.
They seem be become more rare as time goes by, but you are one of the people that still understand how you should equip your paladin & how to socket your gear etc. Personaly I'm playing with the idea of just socketting 2 11healing 2mp5 gems & socket all the rest of my sunwell gear with 10spell haste gems. Dunno yet if I'm gonna go through with it or cap spellhaste at a certain point & just go healing from there but in theory socketting spell haste will give me more HPS then any other gem. Also paladins are mostly main tank healers, with the spellhaste you will get alot more healing ticks on your tank then you can get without any spell haste. And from my point of view it is still better to heal a tank for 1500 every 1.5seconds then to heal it for 2000 every 2seconds just to give an easy example.
Having killed Brutallus yesterday, I can only say, it is so incredibly sweet to have 1.4second passive cast time on flash of light & 1.87sec on holy light (112 spell haste). The more spell haste you stack the easier it is to generate a high amount of HPS with just FoL & thus at a very low mana cost. Brutallus will most likely be the most healing intensive encounter in this instance & honestly I did not rly have any mana issues yesterday & I rly didn't need any mp5 gear for that. Like Amera said... a paladin needs ways to dump mana rather then ways to increase longevity.
Socketting for MP5 instead of raw healing or spell haste is just like Vorda mentioned a few pages ago: a sign of slacking on mana potions or bad healing assignment.
Amera:
Right now i dont have acess to much hast gear, and i would really like to hear about how bad its for the mana efficiency. (I know, i know it doesnt make you less efficient, its just that it dont make you more efficient either, and you are trading efficiency stats, (Mp5 and Crit) for it.)
I understand what you say, but if you go OOM the tank will dies too.
How much MP5 and Crit u stacked before being able to heal, using hast, a 6min long, and extremaly heal intensive fight?
When doing Illidari and Illidan, did you swap much pieces of gear to Mp5 or Crit?
Right now, my calculations shows me that:
Hast is totally underated by most pallys.
We need to just use +heal, to reach 2200-2300 +heal, them Hast becomes the primary stats to HP/S.
Crit need to stand at astlast 20-25% unbuffed in holy, and them you can stack Mp5, probably 200+, but atlast 150 unbuffed.
Since im not yet doing Sunwell, im still farming BT/Hyjal, i would like to know how mana intensive is the fights you face there.
I know that in BT/Hyjal the mana intensive fights are:
Illidari (Healing the Tank of the Pally)
Illidan (Healing the Tank of the Flame)
Archimond ( If people keep standing in fire... )
Najentus (If u want to top meters and spam HL on raid when his bubble breaks) (i know, not really relevant xD)
The others are pretty easy on mana, mainly if u chain pot.
Since we still dont know some of Sunwell fights, it will be useless to try theorycraft what is better for sunwell now.
If kiljaeden proves to be a long mana intensive fight, all the theorycraft of stacking Hast for sunwell will become useless. Well not useless, its still better them +Heal once you hit the 2300+ healing, but we will need alot more of Mp5, and we will be unable to really STACK hast.
real mathematic models to reach a value for Crit/Heal/Hast HP/S (based on a static value of "gear level") and another model for total healed by your mana pool on a fight of X minutes (based too on a static value of "gear level").
for MATLAB (for 8-32 processors) I can use my supercomputer time to make a good simulation.
Gal, you value your effectiveness/HPS assuming that healing should be more or less the same. I mean 'having more +heal will allow you cast more FoL' etc. But in reality leaders (me at least) decrease number of healers when gear progresses, therefore you will be forced to cast even less FoL with more +heal. For example we went from 9 to 6healing Najentus. And I know a guild that 3heals teron (shammy, priest and pala). Pala solo healing MT in teron fight means you can put FoL spam in the a$$.
The "HPS I can maintain over the fight using all my mana" is completely irrelevent and even if it was relevent, it wouldn't change anything about how you looked at haste, as you just used the same amount of mana for the same amount of healing meaning haste did nothing.
You do realize that HPS is a derivative of healing done? So second statement sounds really weird. While I would do same amount of healing, I would actually do it in a shorter time and therefore HPS WOULD be higher. Also higher HPS I have to maintain (and thus wanting haste) means that we got healer less and dps extra thus having fight duration decreased. Also, you say if I would have mana left I don't bother with mana, if I wouldn't then haste wouldn't increase efficiency. But for me haste is exactly the mean to go from "half mana pool" to ''dry" state for the same fight. For me having haste and not having haste changes assumptions (my point of view in this thread is "do not fix what is not broken" and thus i assume that if I want to change/resocket my gear than something doesn't work now), while you assume the same situation with and without haste.
Before debating let's agree on terms. For me HPS is a function of time (basically discrete derivative of healing done). Minimal level of HPS I have to maintain over fight time (let us call it iDPS - incoming DPS) is discrete variable of damage done to the target that I have to heal. If I have to heal more targets than 1, then I have to discretize damage done on a longer period of time, since I can't heal few targets at one time.
First of all I have to be able to cover maximums of iDPS - you call it bursts. In this case haste is the best stat to have per iPoint.
Secondly, I should be able to keep HPS>=iDPS over the full fight duration. You say, that I should consider as a desired scenario (for comparing different stats) the case when I go oom exactly at the end of the fight. And this is one of the points where I don't agree with you. On one hand in real T6 fight we don't want it (it means we could wipe if we would figh 30sec longer for example). But imagine you do go oom, then it means that you cast HL quite often, actually often enough to not even being close to FoL spam (and this screws your logic of stacking heal and do more FoL and less HL). If you gear for high +heal since it's more effective stat to stack and use +22heal gems in every slot you most probably DON'T EVER go oom in 95% of T6 fights.
Now I am getting closer to my itemization paradigm. What is more important - being able to heal burst or being able to keep healing high enough during the fight? From my experience in T6 content (and I found out that healing there differs from healing in T4-5 content) being able to heal bursts on your target is more important than being able to heal long enough. Basically, your assignement bursts should be covered by you while small iDPS would be covered by other healers/hots/aoe heals etc. Therefore I now value haste quite high since it helps me to cover bursts. Healing spike damage is reactive in principle. But this creates a dilemma, since you reactive healing increase the chanses for your tank to be killed while you cast the spell. How one can solve it?
There are 2 types of spike damage - predictable and non predictable. Predictable means that you either know when spike is going to happen and can chaincast HL11 for a few secs. Or you know for sure that after that spike the probability of other spike is very small (for example mother saber lash). In second case reactive healing is OK. If you fight a boss with a lot of unpredictable spike damage, then you are more or less forced to use relatively high ranks of HL (like 7-9) as a chaincast. Or you can use high haste to make casts faster and therefore decrease the probability of tank getting killed. Also crit is nice to have in this scenario (you say I don't need crit mana if I stay half filled at the end but for me crit gives me that extra mana to stay half filled at the end and thats what I want since it means I am more flexible if usual non 'oh shit' healing leaves me with mana).
One more fact - I was looking on Brutallus fight with SP and there +10 crit gem was about as good as +22heal gem. In simplified formula '(mana without crits) * (HP/mana of your heals) / (1 - 0.6*crit)' you have to take into acount the fact that variable HP is increasing with increased crit %. You also have to take into acount that you exclusively HL there, example: 98 HL vs 0 FoL, 99 HL vs 3 FoL, 92 HL vs 0 FoL (Winternight - WWS).
P. P. S. Just an observation - most paladins in top guilds and in guilds killed 2-3 Sunwell bosses don't have 22heal in every slot. They actually do have quite some crit and haste, including both crit and haste gems. I am theoretical physicist and I know that if a good theory says that something has to be done in a way X, but experimentalists do it in a way Y, then most probably that theory doesn't work or for any reason it's more convinient to do it way Y. If theorycrafting says go +heal gems in all slots but best paladins don't do it, there should be a reason behind it.
Right now, my calculations shows me that:
Hast is totally underated by most pallys.
We need to just use +heal, to reach 2200-2300 +heal, them Hast becomes the primary stats to HP/S.
Crit need to stand at astlast 20-25% unbuffed in holy, and them you can stack Mp5, probably 200+, but atlast 150 unbuffed.
...
I've showed in my calculations above that haste becomes better for FoL at about 1750 + heal, not the 2200 range, and is better than +heal for any serious level of +heal (> like 1000).
Haste is the primary stat for HPS, AND is better than just a +heal increase, since you can switch targets and provide faster response.
20% crit unbuffed in holy is reachable with 0 crit gems at T5 gear, and T6 certainly, and unless you're >constantly< running without a shammy / shadowpriest / draenei priest in the group, and only spamming FoL is really the only time you want >120-130 mp5.
I don't have any experience in late BT / Archi, Sunwell, but it seems to me even now, with 1-2 T6 level pieces, that Haste is the best way to go to improve my utility in the raid (I'm at 2000 healing unbuffed, usually ~2150 in raids). I would have to agree with Amera, and put some bounds to it, that once you get to about 2000+heal / 100mp5 / 22-23, which is about full T5 set, haste is the way to go, keeping these three minimum bounds.
And really all this is variable, if you're using like 50/50 HL / FoL, you'd probably be better off dropping 20mp5 and getting to 27% crit or something like that, and then stacking haste. It all depends on how you like to heal.
Amera: Right now i dont have acess to much hast gear, and i would really like to hear about how bad its for the mana efficiency. (I know, i know it doesnt make you less efficient, its just that it dont make you more efficient either, and you are trading efficiency stats, (Mp5 and Crit) for it.)
How is mp5 an efficiency stat when you never run out of mana? Imo you shouldn't make use of the haste gear (Teron belt for example) since you sacrifice too much other usefull stats. But when it comes to socketting there is nothing better in game 10 spell haste gems imo. There is no other stat that will increase your efficiency more, not even raw 22healing gems.
I understand what you say, but if you go OOM the tank will dies too.
You don't go OOM...
How much MP5 and Crit u stacked before being able to heal, using hast, a 6min long, and extremaly heal intensive fight?
When doing Illidari and Illidan, did you swap much pieces of gear to Mp5 or Crit?
No... regular maxxed out +healing gear. And yes I was spam healing all over the place.
Right now, my calculations shows me that:
Hast is totally underated by most pallys.
We need to just use +heal, to reach 2200-2300 +heal, them Hast becomes the primary stats to HP/S.
Crit need to stand at astlast 20-25% unbuffed in holy, and them you can stack Mp5, probably 200+, but atlast 150 unbuffed.
Haste is imo a new factor, since before 2.4 spellhaste didn't reduce the global cooldown & thus didn't matter for flash of light. In addition the haste gear from BT isn't really better then T6 counterparts
Since im not yet doing Sunwell, im still farming BT/Hyjal, i would like to know how mana intensive is the fights you face there.
Kalecgos & Brutallus are pretty healing intensive... so is felmyst but in a different way. Since on felmyst you will take alot of healing & thus regen a load of mana. Haven't killed Felmst yet but didn't rly have any mana issues yesterday.
If kiljaeden proves to be a long mana intensive fight, all the theorycraft of stacking Hast for sunwell will become useless. Well not useless, its still better them +Heal once you hit the 2300+ healing, but we will need alot more of Mp5, and we will be unable to really STACK hast.
It doesn't get any more intensive then Brutallus... Kil'jaeden is most likely gonna be a long epic fight much like C'thun back in the days. A control fight where tactics are the best way to succes rather then pure gear.
On top of that... untill now there has not been 1 single fight in this entire game from Lucifron over Kel'thuzad (hell I killed Kel'thuzad in full Judgement) to Illidan where a paladin needed mana regen... I don't see why all of a sudden I would need any on Kil'jaeden.
More pots... all I can say rly.
Originally Posted by Palados
P. P. S. Just an observation - most paladins in top guilds and in guilds killed 2-3 Sunwell bosses don't have 22heal in every slot. They actually do have quite some crit and haste, including both crit and haste gems. I am theoretical physicist and I know that if a good theory says that something has to be done in a way X, but experimentalists do it in a way Y, then most probably that theory doesn't work or for any reason it's more convinient to do it way Y. If theorycrafting says go +heal gems in all slots but best paladins don't do it, there should be a reason behind it.
Most paladin use other gems then 22healing in order to activate the meta socket. Most paladins socket Insightful Earthstorm Diamond in their meta which requires you to have 2 blue & 2 yellow sockets to. Usualy filled by 2 x 11healing 2mp gems & 2 orange/yellow gems. Cause eventhou mp5 is not rly important for paladins, IED reduces the mana cost of your heals by 15. Which is alot more usefull then 26 poor healing for the alternative meta gem.
What I find really disturbing these days is how people think intelect & mp5 is good for paladins... yes it's nice to have. But NEVER at the expense of other stats. Never ever...
Well, idk about you, but i go oom in some fights, and i use pots on every CD in all fights.
Illidan, illidari are the most important ones.
Maybe its cause of bad raid healing assign, that obligate me to sometime save the life of someone with a HL.
Other possibility is the fact that i need to use alot of Max rank HL on those fights, most of the time, its just me and a druid on the Tank that takes the Pally on counsil, or on the tank that take the Flame.
If my efficiency was better, on fights the tank takes little damage, i would be able to help on raid heal even more them i do now, like on Najentus.
If u never leave the Tank target, you probably will have alot less mana inssues, but my guild have weak healing, and i need to help with all i can.
And 90% of the times i raid without shaman, boomkin or SP on my party...
I've been collecting a haste set once I heard about the incoming GCD changes, but I hadn't been using it until fairly recently. During Sunwell (Brutallus in particular) on the PTR, all our paladins had the same assignment (spam the tank), and we experimented with haste gear. Obviously the math shows the HPS increase, but in practical testing, the stability of tank health increased considerably when we all dumped the mp/5 gear for as much haste as possible. It also accounted for significant gaps in healing done for which of us was wearing haste gear and who wasn't.
The only fights in BT where I feel I need to pot, barring some exceptional circumstance, are Council and sometimes Illidan. On Bloodboil I pretty much always have a shadow priest so it really isn't necessary, even with HL11 spam on the fel rage target, just due to gains from spiritual attunement. I also have an alchemist's stone if I know I'm going to be chain potting on a fight, but I haven't used that in months.
My project before the reset Tuesday is to resocket all my yellow slots with +haste and get some of the crafted/badge haste gear, and give that a whirl for our next clear. I'm expecting some pretty nice results.
03/28/08, 11:12 PM
Similar Threads
