bomzix

On Patchwek style fights ( Static fights with lots of HPS required for safety ) it's easy to see how much u need and how many healers u need to provide it. Haste increses HPS but decreases spam longevity, modeling this relationship for generic fights doesn't sound very usefull... For a particular situation it's very easy and usefull.

On the whole I'd say u need/should have some haste and some regen gear to swap around from boss to boss. Hybrid Bags anyone?

 

galzohar

"Spam longevity" is meaningless. The easy example to show why it doesn't tell you anything is that I could show you a gear and heal rank spam where adding 1 mp5 would multiply your longevity by a factor of infinity. Obviously 1 mp5 is never worth infinity, and since there is 1 example that the "spam longevity" is supposed to be relevent to that is obviously incorrect, the whole concept is incorrect as well. That is, no matter what spell you're using, knowing that adding X of a stat multiplies your longevity by Y doesn't really tell you anything useful at all. Another proof is simply saying that healing twice as fast with same mana efficiency is not worse than healing normally, while your "Spam longevity" would be 1/2.

Casting a HL every 3.5s is not even close to spending most of the time casting. That's 43% of the time doing absolutly nothgin. While I can't really tell how much of that 43% you really couldn't be doing anything and how much you were simply not healing as good as you could be, I really doubt during 43% of the fight the tank is sitting at 100% HP.

Keep in mind while FoL spamming is about 1/2 the HPS of HL spamming, so spamming FoL for the full duration alone would probably get not much worse results. And since you'd have so much mana left at the end, during bursts swapping that FoL to HL to the point where you still use all your mana, and voilla you did more healing and spent much more of the fight actually doing something. While you can say that you only need big heals as he either takes damage or doesn't take damage at all, there is no way to prove no matter how much you try and WWS it that cast-canceling HL will save the tank better than always having a FoL incoming with LG up to react with HL after a big hit.

This also has nothing to do with how I calculate itemization as you could use the same calculation with any healing style, you just put in different numbers and get different results (and not much different at all especially if you assume that you're not completely unable to cast FoL).

I also don't see why you're letting a feral tank on a fight when your "burst" includes a 10k fire damage hit even if the boss can't crush. There had been enough discussion in other threads saying it's debateable enough on bosses that deal 0 elemental damage and never crush wether you should tank with a feral or not, so a boss with extreme elemental damage burst should obviously never be tanked by a bear. Not to mention bear tanking has other costs on top of the "who is a better tank" question, like keeping TC (and possibly demo depending who's job it is to keep it up in your raid) becoming the duty of another warrior, and if you don't sit out your prot warrior you're subbing a cat DPS with a prot DPS which is very bad.

 

Palados

Well, isn't working for Brutalus. Basically, this boss can, in principle, twoshot tank. Therefore you do what I think those palas did - cast HL about as fast as boss hits. And let them land. Thats it, no stopcasting, no FoL/reactiveHL combo. As soon as your mana allows to cast HL a LOT (remember, that you won't fight brutalus more than 6min and 20sec or so no matter what) you do it. Mana of those guys allowed them, therefore they did it. You should agree, that as soon as you are definitely sure that effectiveness won't be an issue, you should maximize your HPS aka burn all your mana in that 6min to make a kill safer.

Pala job is to make spikes smoother and let big heals land/hots tick. We did it with FoL earlier. But if we have to do it with HL - I don't see why we shouldn't. We never were pro in reactive healing. We have chanse to be with a lot of haste, but we will see how it will work out.

 

• alcaras

For what it's worth, here are some target gear sets I worked out for myself, optimizing around an Overall build, a HPS build and a HPM build, and including Sunwell and not Sunwell loot options.

 

Palados

Hm, you assumed what gemming, when you choose between glimmering and lightbringer shoulders? I woulnd't go for Glimmering over Lightbringer in overall set tbh, but it's just me. I would use sunwell crafted chest over t6 one (it owns by all parameters) and use T6 shoulders in almost every sets you listed as 4th T6.

 

• alcaras

Gemming is done according to these links (in the particular example you noted, both would have 2x Teardrops; it's a pretty close call). I wanted to get just the 4 piece bonus from T6 off the strongest pieces (Chest, Bracers, Belt, Boots).

Overall:
Loot Rank

HPS:
Loot Rank

HPM:
Loot Rank

I'm an Enchanter/Alchemist, so I can't access the Sunwell crafted chest.

 

bomzix

K, when I said "Spam Longevity" it wasn't as a concept to base yur gameplay on, it's a meter of how long u last doing a particular spell on a given situation.

If you know u have to do a spell over and over for X minutes then u know how long u last with very adequate precision. Patchwerk on NAXX allowed u to stopcast a lot with minimal risk and conserve your mana. Never seen Brutallus "in person" but doesn't seem like u can stopcasting at all, if you can't then u need to know if the spell u choose to spam allows u to last he whole fight using the resources u have.

 

vorda

I agree with the t6 shoulder remark and am also wondering why in the overall set you prefer naj'entus ring over hyjal rep or Blessed band.

 

• alcaras

Probably because I didn't take into account the hyjal rep ring's proc in my weights, which now that I've noticed it, I will use once I hit exalted.

vs. Blessed Band it's really a haste v. crit trade off that's very close (139.6 v 137.26).

I wish there were more healing items out there with haste _and_ crit. Unfortunately there are just two: Items - World of Warcraft (T6 belt, T6 boots)

Note that my gear sets aren't meant to be 'ultimate best end all be all.' This was a rather mechanical effort of looking at the best items by item weights. It's more a reference for me to look at when deciding what to roll on for what purpose, and I put enough time into it I figured it might be useful to post for others to use.

That said, I appreciate the commentary and critiques.

 

galzohar

This kind of calculation has no effect on gear choices nor playstyle, and therefore is completely pointless. You're not going to pick your rank based on that and you're not going to pick gear based on that. What matters is how much burst you can put up and how much healing you can do during the whole fight using all your mana.

As for bruttalus, I wouldn't call spending 57% of the time casting a "I have the mana for it" situation especially when you end the fight with 0 mana. Spending some of that time on FoL would allow to do more total healing with the same mana (and spending more time in the fight actaully doing something). You can't say you're not canceling if you only cast a heal every 3.5s becuase it pretty much means you on average cancel every 2nd heal (or isntead don't even start casting it). Considering the boss doesn't stop attacking at any point in the fight, it means every second of that 43% remaining time was not used for anything - either the heal got canceled becuase the tank wasn't low enough or wasn't even casted in the firstplace. Both cases Adding FoLs could help.

And again, even if the paladin class did not have a FoL spell my theorycraft would remain the same, only the FoL would be replaced with some HL rank.

 

tholex

I'm wondering about your point assignments to stats on the links. You should mention that they are an opinion, rather than some kind of definitive go-to for gems.

For instance, for HL's HPS, what discussion seems to have been centered around for a while, the equivalencies are about (normalized around my HPS calculations for HL at T6 level of gear a couple pages prior)
Heal = .834
HasteR = 3.3
CritR = .988
Int = .677

Basically putting these values in would maximize HL Healing/Sec, and stacking Haste gems, ignoring gem req's entirely is the way to go, if HL HPS straight up is what you're looking for.

Efficiency for HL goes up decently with crit, so I suggest to increase the value of crit for HL efficiency, and mp5 as you wish. Increase the healing coefficient if you want to also add in FoL (.95 healing is the average of HL and FoL, and still suggests haste as by far the best gem to go for)

 

galzohar

First of all for HPS crit=0 as while it provies average HPS, you can't rely on it. It will still increase efficiency but will not increase your ability to reliably save a tank the way you're trying to do with max HPS and ignoring mana effficiency.

Then int is 0.35 healing without kings and without crit, 0.385 with kings. Why did u rate it 3.4?

For efficiency you're way over-estimating crit, and that mp5 value fits the one I have when I have a shadow priest + resto shaman + chain potting - any less of those buffs and mp5 value goes way higher all the way to 6~7. And you have haste as a negative - even if you ignore the fact that higher HPS allows slightly better efficiency it would be 0 and not negative. And HPS provides and obvious >0 efficiency benefit. I mean it's not huge but it makes +healing and haste (and to a much lesser degree crit and int) an extra benefit over mp5. In fact if you ignore that benefit mp5 should be quite higher than 6-7 healing, more like 10+ (although obviuosly not correct as you need to take the HPS->efficiency benefit into account).

The actual overall stat ratings would be somewhere between max HPS and max efficiency, depending on how much you need both. Just remember that most stats help one by a lot and the other by nothing or not much, while +healing is either 2nd best or best stat (depending on your (shadow priest etc) standards) for both efficiency (2nd to mp5) and HPS (2nd to haste), and not too far behind. So if you need both about equally, +healing is by far the best stat. On the other hand, you sometimes don't "need them both equally" but rather need "as much HPS as I can get with not a lot of need for efficiency" and sometimes (although a lot less often) "as much efficiency as I can get while not really caring about HPS becuase I'm going to need all my mana sustaining heals rather than handling bursts".

 

Palados

Well, gal, let's look again at real, not theoretical fight.

Imagine me assigned to tank doing usual job - FoL spam with reactive HL after burts. You say that with extra +healing you can FoL more and HL less for same healing done and thus increase the efficiency. I say that for any realistical +healing you are still going to spam FoL and reactive HL. Thus you will follow the same healing pattern (doing bigger heals ofc, which is safer). I can hardly imagine a situation where you added +heal enough to FoL after burst and thus not casting HL. The nature of MT pala healing binds us more not to our mana, but to certain cast sequence. And I am NOT (I think you don't as well) gonna cast FoL on tank less often only because I have more +heal. Since FoL isn't reactive healing and only a mean to create a stream of small heals, thats why you cast it more or less non stop on a single target. If you have 1900 healing or 2500 you gonna do it on about same pace and thus use same amount of mana per minute. Do you agree with me?

From a practical point of view, I would rather look at what stat is more efficient to have assuming constant cast sequence/healing pattern (say cast each 2 sec, HL or FoL depending on tank HP; or FoL spam with HL after each burst). It will definitely change the weight of different stats.

Once more in short - for most fights assuming usual assignement (MT healing) extra +heal will makes healing safer but will hardly add efficiency. And this fact, imho, should be reflected in calculations.

 

galzohar

Wether you use your extra +healing to cast less or do more healing, it helps either way. What you really need to get is that gear changes make very small difference. That extra +20 healing is not going to be gamebreaking. It's less than +1% to your efficiency, which means 100 FoLs will become equivalent to 101.

Sometimes when things are real easy I actually do cancel FoL if the tank is full HP and I know all healers that are supposed to be on him are alright (not running from doomfires, not feared etc).

And as for using less mana because of higher HPS, even if you end up using rank9 HLs instead of rank11 and then rank7 instead of a rank9 and then a FoL instead of a rank7, it's the same as if you used a FoL instead of a rank11. Granted there's some room to play around with it here, but the end result is that if you ever cast FoL, no matter how you break down the spells you will more or less end up gaining the same HPS->efficiency benefit as if you were using it to replace HL11s with FoLs. If you really want to be more accurate you can use your spell distribution and actual downranks you use, but the difference in accuracy of the calculation will be very small.

At the end it's also not really impossible to have it happen once in a fight where you casted 1-2 FoL instead of max rank HL because your HPS was high enough to afford it but slightly lower HPS wouldn't allow it. Remember at the end your HPS determines a breakpoint of required HPS that once crossed you stop the FoL and cast a HL - having more HPS will help you cross that line slightly less often. This principle can be applied to any combination of spells and ranks, and while I agree it's more realistic to do a calculation based on more "continous downraking", the actual results are too similar to simply calculating HL11->FoL to bother (again, as long as your lowest HPS heal used is FoL, although I bet you wouldn't get extremely different results even if your lowest rank is, say, HL7 or 8).

 

• alcaras

I generated the point assignments using this php script which I wrote:
Alcaras' Paladin Stat Weighting Calculator

I may have made mistakes in some of my calculations in terms of how the formulas worked, though I did my best to cross check it with an existing spread sheet.

In any case, I've released the source if others want to play around with it:
http://subcreation.net/pal/source.php

 

tholex

Ok. But I think you still see my point - mana vs. healing done is valued differently based on who you ask. Some pallies like more mana efficiency via mp5 and FoL more, some like more crit and HL more, etc.

Basically, mana efficiency is a stat that differs in value depending on the person and the situation. Yes, your equivalence point between healing and mana efficiency may be valid for a lot of people, but yea there are other ways of valuing it.

For instance, I add an arbitrary 20% more to my haste coefficients just because I feel like that's what I'm lacking at the moment.

 

needle

Zurm mentioned that Divine Intervention may be the short end of the stick in terms of wipe protection. It does give you options however, and if used in a combat situation it may also function as an advanced dispel.

At Kaz'rogal for example you repetitively get a non-dispellable debuff which drains mana. Once your mana pool is empty, you blow up. When you are completely drained with mana pots/demonic runes on cooldown and mana tides/innervates unavailable, you can use Divine Intervention, not as a wipe protection, but as a means to remove the debuff from another caster like a priest and give them time to regenerate mana out of combat, ready to reenter the fight when every other healer is drained or dead.

While this certainly is not your everyday use of Divine Intervention, it is nice to at least be aware of this option. You never know when it might come in handy.

 

galzohar

Assumption - the more efficiency you have the less likely you are to use all your efficiency, and thus the more you have the less you need it.

FoL users have more efficiency than HL users, and thus actually don't need it as much and can afford more "pure burst" gearing that more and more people seem to swear by.

HL users have a lot less efficiency and thus actually need it a lot more.

Neither FoL users nor HL users have any different max burst (assuming same gear). Since mp5 and crit neither really affect your max burst then the people you claim who "FoL with mp5" and "HL with crit" must be doing it for efficiency, right? But hey let's look at the actual efficiency you get - even HL will get noticeably more efficiency with mp5 than crit even if you already have a lot of crit, at least as long as you assume any realistic amount of crit overhealing (as in, as long as you don't assume each crit heals for 1.5x - because it's not realistic to get full advantage of the extra healing from crits). The only argument I seem to hear from HL users for equipping more crit is "I use mana faster and therefore crit returns more mana", which is completely irrelevent if you followed any of this and/or the other thread.

Since both "styles" use the same kind of "burst" to heal on an emergency, they gain exactly the same burst from the same stats (healing/haste).

For bread&butter healing the only gain from having HPS is the more HPS -> can use more efficient heals to keep up -> more healing done with same mana. This is because you never use rank11 HL for bread&butter healing - there's no fight in the game where you can do all the healing in the fight with rank11 HL (unless you're way overgearing it and can allow to burn mana stupidly like that). On top of that you should look at the actual DPS tanks get and the actual HPS of FoL and HL and the number of healers usually on the tank. HL11 is always an overkill when the tank isn't getting unlucky and/or other healers are unable to heal at that moment for whatever reason.

 

Mirkael

First, you are wrong when you think that adding HP/S just unables you to use FoL more.
Everyone can heal Najentus, Rage, Teron, Azgalor... But the intensive heal fights are a different story: RoS, Healing the Tank of the Paladin on Illidari, healing the Flame Tank on Illidan, and sometimes on Archimond if you have less them normal number of healers for the fight.

People arent trying to gear to do the easy BT/Hyjal healing fights, but to maximize the efficiency on the difficult ones, and mainly on sunwell right now.
For what we know now about sunwell, the burst damage will be really big, even more them the hard fights on BT/Hyjal. And most of the fights have a low enrage time.

When you increase your HP/S via Hast/Heal and to some extend Crit, you lessen the chanse the tank will die.

HL builds gears for Crit because considering the total mana pool (Mana+Pots+Mp5+Shadowpriest+etcs...) and adding the 60% cost reduction/1% of crit, you will discover that for 6mins fights, Crit is better them Mp5. (even longer fights depending on party composition, and gear level)

Nowday pallys are trying Hast because in 6min fights with a good party setup, we are efficient enough, we just need to cast faster so we can save the tank.

If people that gear for HP/S was trying to do it because they wanted to use more FoL or Downrank, they should have used pure +Heal, since the extra HP/S from hast comes with a huge maka sink.
No. Most of people that are gearing with hast arent looking for downrank or FoL Spam, but to use FoL and HL on the best combination possible to maintain and save the tank. (Yes, your pots will need to be on CD all the time, you will expend alot with consumibles, you will need a good party setup with atlast a SP, but in the end of the day, your tank will be alive, and your guild progressing...)

 

galzohar

When did I ever say "gear haste/healing purely becaue it helps your efficiency"? I never said that. The main purpose of haste/healing is bigger burst to help heal the tank when he takes a burst. I'm just saying the extra HPS also has a small increase to your efficiency when healing outside of a burst, on top of it. I never said haste is a great stat for efficiency becuase it isn't.

Ocourse if you want to be more efficient you don't gear haste. While the efficiency increases are there, they're much smaller than the efficiency increases you would get from +healing/mp5. This is probably the 50th time I'm saying this, I don't know why it didn't get through the first time.

Remember to seperate the "consistent healing" from "burst healing":

You cannot use max rank HL nor do you need to use max rank HL for the consistent healing. This means that if you increase the HPS of whatever spell you're using during your consistent healing, you will be able to use slightly more efficient spells during that time as you can't use any more HPS effectively, as in without adding massive overheal and going oom. If you could, you'd be using a higher HPS heal in the firstplace and go back to the beginning of this paragraph. Here the HPS of haste, +healing (and even crit to a lesser degree) help you be more efficient - not by much, but it gives you something so when you compare items for efficiency you need to add these values. This means if I get enough +healing to get 1% bigger heals, my efficiency increase will actually be bigger than 1%. This also means that I will get some small efficiency gains from haste - much smaller than other stats would give, but far from 0 and definitely not negative. Even the small HPS increase from crit can help a little but it's a much smaller factor as the majority of the crit benefit is the mana return and the little bigger heals on average provide an even smaller HPS->efficiency increase as an added benefit to what you would normally consider crit to give.

During a burst, you don't care about mana, you're just casting your biggest meanest heal to save the tank from dying. The threshhold of tank HP depends on the fight, phase, and general raid status and is up to you to decide when it happens, how often and how important it is to be able to handle it better. This is where haste, and to a bit lesser degree +healing, are the only stats that really do anything useful.

The problem is that the more often you need to use your max burst in a fight, it'll also require more efficiency and vice-versa. Determining how important each aspect of healing is isn't simple, although it seems to be mostly leaning more towards HPS than efficiency considering a raid will, in general, see a lot more tank deaths than oom healers during its progress.


As for HL vs FoL, the MP5 vs crit in terms of efficiency really doesn't change as much as you claim it to be changing. The people that say HL=crit and FoL=mp5 generally calculate their crit value based on their mana/sec usage which is just wrong. While HL does have a bit higher base crit it's not enough to make crit all that great compared to mp5 or even +healing for efficiency. And the fact a lot of top-end paladins use 22 healing gems makes your "but top paladins gem crit and succeed so crit was tested to be great" claim doesn't mean anything. And this is in addition to the fact that you cannot really "test" the effect of those stats by WWS/healingmeters/whatever, as there are so many other factors that are a lot more significant than what gear you're wearing that affect what you see on WWS/meters. Again it's like comparing who has his car tuned to drive faster by driving them through a high traffic town at different hours with different traffic and different luck on traffic lights and each driver testing differently. Not only I've yet to see someone who had everything gemmed with crit one day and then the other gemmed everything with 22 healing and said what the difference is, and to be honest even if that was economically feasible I doubt someone could really feel the difference unless he swapped a rediculessly high amount of gems.

 

Palados

Gal, how do you calculate efficiency of +22heal gem btw? In some other thread you wrote that it is about the same as 4mp5 efficiency. Let us take average MH fight - 5min - as a base for calculations. I know how to compare crit and mp5 gems, but I am interested how you calculate efficiency of +heal.

 

galzohar

When I actually calculate I use the baseline spreadsheet which calculates efficiency as total healing done using all your mana.

Each point of +healing makes each heal bigger for same mana cost, and you cast X heals and this healing applies to each of them. This portion of the calculation was already covered in the original spreadsheet.

Then each point of +healing also provides more HPS, which allows you to save a little bit more mana. In the other thread I explained how to estimate the effective mp5 gained by +healing, by calculating the +healing needed for your lowest heal (generally max rank FoL) to catch up with your HL11's HPS, and then how much mp5 would be needed for HL11's efficiency to catch up with FoL's efficiency, and divide the healing by the mp5 and you get the healing->mp5 conversion value - you could actually do HL11->HL9 + HL9->HL7 + HL7->FoL or something of that nature but your results would be very similar to simply doing HL11->FoL. Then 22 healing would give you +healing like the spreadsheet tells you (X extra healing added to every heal) and on top the mp5 gained by dividing the +healing by the HPS->mp5 conversion. This extra mana will then translate on the spreadsheet into more healing done and give you the actual efficiency gain of +healing. This value is *NOT* the same for different party setups. The more mana you have, the more mp5 you need for HL's efficiency to catch up to FoL.
The whole basis of this conversion is assuming HPS effectively gives you mp5 by wasting less mana on same heals while keeping the same average HPS up, and taking the simple case of full FoL vs full max rank HL to see how much would be needed to switch from 1 to the other altogether and then saying that if I only get a small part of the +healing required then only a small part of my HLs will become FoLs.

The "22 healing = 4 mp5" is very approximate. Depending on your gear and more importantly the presence of shadow priest (and resto shaman) it can be anywhere between 3 to 6. If you're in the tank group for example wearing my gear it would be close to 6, while if you're in a resto shaman+SP group it'll be close to 3. Since people who care about efficiency generally don't even have a SP then for them mp5 is much better efficiency than +healing. If you have a shadow priest and no shaman they're about equal efficiency and with both 22 healing is better, but when you have a shadow priest you possibly don't care about efficiency to start with.

At the end if your burst ability matters, which in most fights it's probably more (how much more depends on the fight/setup/etc) important than your efficiency, and is not increased at all by mp5. Haste on the other hand increases burst kinda better than +healing does but gives a rather small increase to efficiency.


Using my stats as a baseline, I get effectively 1.1 +healing worth of efficiency per 1 haste rating with my gear, and it stays the same no matter how much mana you have as long as the spell distribution stays about the same). A +healing gem will give you around 1/2 of the burst that a haste gem would while giving you between 0.75 to 1.5 the efficiency of an mp5 gem at the same time. A haste gem would give 2x of the burst a +healing gem would while giving 1/2 its efficiency. Then again you need about 25-30 +healing to increase your total healing done by 1% and about 60 +healing to increase your burst by 1%. This is using my gear as a baseline and didn't cover all possible party buff options and rounded some values, so calculate with your own gear for accurate results. On a side note assuming on average 2/5 of your crits don't overheal, 1 crit rating is worth 1.31 healing for efficiency (which unlike mp5 stays worth the same

With these numbers it seems +healing is overall best general-purpose due to giving you a good portion of the HPS haste would give while still giving a good portion of the efficiency mp5 would give, but if you know you're going to face some more specific situations where you need HPS way more than efficiency or the other way around, you can gear accordingly - massive haste when only burst matters, massive mp5 if you don't get a shadow priest and want efficiency only and don't really care about burst (which is a really rare situation in my opinion), and still pure +healing if you want efficiency while having full support of SP+shaman (again a very rare situation of still needing efficiency more than HPS after you were given a SP not to mention a shaman on top). As you can see the situations where you would actually want to stack mp5 are quite unrealistic but not impossible.

 

Endoscient

I don't think you can just totally ignore crit for average HPS. It is really not realistic. Imagine this scenario. Standard tank healing spamming FoL waiting to HL when needed. In one case none of your Flash of Lights crit and after 10 seconds the tank gets low enough to warrant a Holy Light. Now what if 3 or 4 of those flash of lights crit and the tank did not get low enough to need a Holy Light because of it. On healing where it is more consistent damage or solo healing (rogue tank on council, infernal tank on anetheron) this happens all the time.

 

galzohar

I am not ignoring crit for average HPS. When calculating efficiency, which is increased a bit when your average HPS goes up due to being able to use more efficient spells, I do count the (low) HPS from crit.

When you need your max burst to save someone from dying, though, it's completely unrealistic to calculate anything average. You're looking at your ability to handle a worst case scenario and on that scenario you're not going to crit by definition, just like the tank is not going to dodge.

So crit does help your average HPS when you calculate it for the sake of efficiency, but it does absolutely nothing for your emergency burst.

 

Endoscient

I don't think it is valid to ignore it then either. If you are doing your maximum HPS throughput you are spamming HL11. In that situation most of the time someone dies it is because your heal didn't get there in time. Say it did land in time but the target still died, it is not likely it would be different it it healed for 100 more. You would have to be lucky and have the numbers work out well for the extra 100 healed to keep the target alive. While in a lot more circumstances if your heal landed and the target still died, he wouldn't have if the heal did crit. So I would say the increased chance of the heal critting and saving the target is just as (if not more) likely then the target not dying because your minimum healed was for 100 more.