Pitty there arent any of those left in the current end game content. For us, our super druid almost always wins on bossfights in sunwell (kalecgos meters are off, brutallus has 2 tanks + 1-2 burns which he rolls lifeblooms on, same for Twins) the not-so-good druid score around the middle where usually our paladins are aswell.

If your shamans arent all above your paladins and in most cases, above your CoH priests they either suck or your raid assignments are off.

On healing meters:

Someone has to heal the tank. This is a rather important job. It is a job that paladins are well suited for. Unfortunately, it is a job that almost inevitably entails high amounts of overhealing due to the need to compensate for the unpredictable damage from the boss's melee. It is also a job that entails using single-target heals that will generally not have the HPS of AoE heals. Thus, however heals the tank should be fairly low on the healing meters (unless all the healing in the encounter is tank healing).

Ergo, in a typical encounter paladins will tend to do poorly on the healing meters if the paladins are appropriately assigned to tank healing and if the other healers are doing their jobs. Does this make paladins worse healers? Of course not. If the paladins weren't healing the tank the priests or shaman would be, and then they would be lower on the meters.

Ergo, anyone who wants to top healing meters should not have rolled a paladin.

You judge healing by keeping your assignments alive. PERIOD. WWS records on "who healed whom" gives a great insight as to what people were doing. As a paladin, your primary job is usually to keep the tank alive. My guess is, if a tank dies and it wasn't due to some tank failure... its because a paladin got caught on trying to heal someone else and the tank got hit with a quirky spike. Its soooo tempting as a paladin to toss a heal ... but it comes with big risks.

My biggest pet peeve comes from pallies that just spam FoL ... you can spam FoL till the tank dies. Learning how to be a pro cancel healer will be much better for you as you enter Sunwell.

Paladin healers (especially with a shadow priest) should be able to destroy meters on Brutallus. It is such an ideal fight for Pally tank healing. The only healer that can beat you assuming similar skill/gear/group is the raid healer who is healing slashes.

Here is the WWS from our Brutallus kill this week Wow Web Stats. Where over the fight I do 520k effective healing.

Tanks just need to get topped off extremely quickly for Druid healing on tanks to do anything nearly substantial. Our pure out burst hps destroys shaman and priest single target healing. With a shadow priest you can sustain it for most of the fight, especially during stomp.
If a tank dies because the paladin assigned to him was healing someone else and not the tank then that paladin is a total failure. I agree with the FoL spam, in Sunwell spamming FoL is far below your potential. But, I don't think cast canceling is the right way to go. Intelligent switching between FoL and HL (max, 9, etc) will lead to far greater results.

I strongly agree with this and will add that with increased latency canceling casts becomes increasingly dangerous. I only cancel my casts when I'm really running on fumes. This means I'm sitting on LoH, canceling every 2nd cast and watching my pot timer. Usually this situation occurs when I make an error earlier on in the fight and get a bit happy chain casting HL11.

I've actually had to beat it into my tank healers heads that we are no longer raiding MC and overhealing is what keeps tanks alive. I realise this is an over-simplification but its useful for me to tell my team this so they don't get too uppity about effective healing and minimising overhealing.

I think the gear levels you used was way too low, you don't even assume 2pc T6 for Holy Light. Remember by the time S4 comes out and people grind up to 2050, they should have a hefty chunk of Sunwell gear, if not close to all of it. With using a +healing mace you had 2500 healing and 28% holy crit buffed (reasonable for mid sunwell) it should push the pure hps numbers in favor of +haste. But I think an integral part of deciding wether or not haste is worth it is analyzing the amount of extra healing done with the extra mana a +healing mace didn't use compared to the +haste mace.

Say if you found a gear setup that did the slightly more HPS with a +haste mace then a +heal mace. How would you evaluate the worth of the extra mana used? What I would do is determine how many Holy Lights you would be able to cast instead of Flash of Lights while using the same mana as the haste mace. Then show how much extra healing that would add.

I will run the numbers on this when I get home tomorrow. To make it equal, I will compare using my gear with full 22 healing gems and with full 10 haste gems.

You are wrong, since fight length doesn't enter your calculations. And it makes probably the biggest impact on comparing crit and mp5. Use spreadsheet where you can choose all parameters including fight length, there numbers are correct (first post has averaged numbers).

iLvL wise 1 mp5 = 2.5 crit rating.

Since Galg isn't visiting these forums anymore it seems, I am going to do 'his' job and remind that crit is simply increasing your effective mana pool. Reasoning: you are only interested in crit vs mp5 conversion if you go oom during the fight or can go oom. Thus, crit potential is used fully for such comparison. Crit influences your mana pool by:

new_mana_pool = non_crit_mana_pool/(1-0.6 x crit_%)

thus crit adds 0.6 x non_crit_mana_pool x crit_%/(1-0.6 x crit_%) extra mana. No matter how big fight length is, only assuming you use your mana fully. So you can see, that in 6 min fight crit is twice as good if we look at his comparison with mp5 compared to 12min fight and 3 times as good compared to 18min fight.

Mp5 on other hand is better with increasing fight length. So to compare those two you should choose the desired fight length and then do your calculations.

META Gem questions here. Someone had mentioned to me that I need to be extra careful socketing my gems; DUH, but that over socketing would also result in the gem NOT working. Meaning if the requirement is 2Y, 2R, & 2B, if I have 3B gems that it wont work either. That the requirements are absolutes NOT minimums...??? this true?

IIRC this was the case in 2.0 but got quickly changed.

[Insightful Earthstorm Diamond] is generally what I use and the requirements are minimum gem numbers as indicated by the "at least" phrase.

Uhh, no. Yes, crit effectively increases your mana pool by that amount (or reduces heal costs by .6*crit) but your mana pool increases as the fight goes longer. You can use more mana potions and 250mp5 from gear/buffs. The only part of your mana pool that doesn't scale with fight length is your base mana pool from int.

Only if you purposefully cast in such a way as to exhaust your mana pool on every fight. Otherwise, all you're changing is the amount of mana you have to spend.

Yes, I was referring to Palados' response specifically. He makes that assumption ("assuming you use your mana fully").

There is no one number that gives an exact relation between crit and mp5, the reason is the more crit you have the more effective more MP5 will be because it will give its normal amount of mana from the MP5 and then the current crit you have will increase that amount as well. If you have a ton of crit and not much MP5, MP5 starts to look really really good. The reverse is true as well for pretty much the exact same reason.

I dont think that its impossible to show the relation between the two stats, but for it to be 100% accurate (in my eyes thats what TCing is all about) any equation would have to include current MP5, current crit, current base mana and even additional stuff like shadow priests and shamen will make a difference as well.

I'm knew to posting on these forums and love them btw and have learned quite a bit tbh. I know there was some talk quite a few pages back about spell haste and someone said they where going to do some testing with it, has the test be done yet?

Wish you guys where on my realm so hard to find skilled raiders on Aussie servers. Anyway going to keep reading these threads.

Thanks for uploading what you currently have. Are the throughput/longevity filters working as intended? I'm kind of confused how they are graphed. I'm really excited to see this module in its finished state. There doesn't seem to be much ongoing discussion on the module here, so if there is somewhere else where this is being discussed, can you point me in the right direction?

Upon doing calculations +healing will always do more total healing then haste under most realistic gearing situations. Some exceptions:

• You have a really low amount of mana so you are spamming FoL most of the time.
• You have a huge amount amount of mana so you never Flash of Light
• You have a huge amount of +healing, ~3500 depending on your mana pool

Here is the excel spread sheet I made that can easily calculate this for your gear level haste-vs-heal.xls - FileFront.com. Fill in the values for your gear level in the orange cells. Crit chance is base crit, not counting any talents. Mana pool is how much mana you have for the fight, without illumination. Base mana pool, mana gained from mp5, mana pots, etc. It assumes a few things:

• Target has BoL
• Using Libram of Souls Redeemed
• 4 piece T6
• Use max rank Holy Light

To determine how it uses the mana it computes the maximum amount of Holy Lights you can do while going OOM when the fight is over. It uses this formula:
HL_Time = (Mana_Pool - Length * FoL_Mps) / ( HL_Mps - FoL_Mps)
And then computes the total healing done based on that. So even though typically +haste increases the Hps more then +healing, you can cast 10-20% more Holy Lights with the +healing setup.

I can modify this later to also compare to spell crit if people want later.

Well, I agree with your results, but what kills a tank usually is not a lack of heals in average, but lack of burst HPS.
Haste helps with burst HPS much better than +healing or crit. Let druids/priests/shamans do more total healing, while we make sure MT never dies while we are alive :P.

This is ~impossible to model though, as reactions to incoming spike heals cannot be mathed.
But it is a very valid point that compared to higher raw +healing, higher spellhaste might offer less output, but better reactiontimes when necessary.

Also, might I remind everyone that healing meters are at best a loose indication of who heals badly. They cannot indicate who heals well. Mathmatical modelling of healing is fine, but comparing simple healmeters to the "quality" of someone's healing cannot be done, depending on the encounter it can be compared to the quantity however.

You know what also kills tanks? Casting Flash of Light instead of Holy Light when they need it. With the +healing setup you can cast ~10 more Holy Lights over a 6 minute fight, that is a much bigger on demand burst HPS increase then +haste provides. Sure, on a fight like Brutallus where it is known almost every time your tank takes a lot of burst haste might be better for that. But, I would call that fight an exception.

How often do tanks die when you spamming max rank Holy Light on them?

I've done some calcul about illumination, i'd like to have your opionin... i hope maths/hypotesis are correct.

If i spent 840 mana with a 20% crit, the mana i get back is not 840*0.20*0.6=100.8 (average).
i have then to calculate that, casting a spell with the mana i get back in this way, i can crit and get some other mana back from illumination.
In this case, i'll get back 100.8*0.20*0.6=12,096...and then i have also to calculate the amount of mana i get back from crits average in this 12 mana, and so on.

So the mana i get back is: summ of i , 1 to 99999 of : used_mana*(crit_chance*0.6)^i.

the result of this is: mana_used(1-crit_chance*0.6/(1-crit_chance*0.6)).


so if i assume that i'll use all my mana available i have:

mana_used = starting_mana_pool + mana_regen + mana_used(1-crit_chance*0.6/(1-crit_chance*0.6))

so:

mana_used = starting_mana_pool + mana_regen / (1-crit_chance*0.6/(1-crit_chance*0.6))

1 / (1-crit_chance*0.6/(1-crit_chance*0.6)) is the mana pool multiplier from illumination.

this is:

crit chance//mana pool multiplier from illumination
0,05 1,031914894
0,1 1,068181818
0,15 1,109756098
0,2 1,157894737
0,25 1,214285714
0,3 1,28125
0,35 1,362068966
0,4 1,461538462
0,45 1,586956522
0,5 1,75
0,55 1,970588235

so what i think is that, if one want to go for crit, he has to stack it because the benefit for crit on mana efficency is not linear.
unfortunatly this is only a multiplier and one need to have a good starting_mana_pool and mana_regen (which one is better depens from length of boss). if one can't get a good mana regen from outside (shadow priest, flask, pots every 2 min) then best thing is to leave crits and go for mp5 / heal / haste.

Read back for 5 or 6 pages this was covered already. One problem with formula like yours though is that it doesn't take into account the different crit rates of Flash of Light or Holy Light. So it is much easier then to think of Illumination as reducing the mana cost of heals by .6 * CritChance. But yes, crit scales geometrically not linearly.

The correct formula is 1 / ( 1 - .6*CritChance) since it is the infinite geometric series. Sum from k = 0 to infinity of R is 1 / 1 - R