Even if that happens, you can still put on a few mana regen trinkets / fingers. Or put a shadow priest in your group (with the healers and optional lock). Besides, there are still more consumables that affect mana than those that affect rage. Oh the wonderful Demonic Runes...
Then it's gonna be Alliance whinning about how horde gets Seal of Blood (wich paired with Judgement of Wisdom) grants you Holy dmg similar to SoR / JoR (with a high DPS 1h) and twice the proc's from JoW. The drawback is having poor consecration / Exorcism dmg.
So this is where i plug my the suggestion i tossed in the void that was the WoW forums.. Replace Weapon Expertise with Imp. Spiritual Attutement. 2 ranks, increase mana return from Spiritual Attunement by 5/10%.
The doge loss is nothing compared to our gear having enough block rating that we can reliably push crushing blows off the attack table. Not only that but on a dodge/parry you lose a big chunk of threat compared to a block which gives you threat from; retribution aura, BoSanc, Holyshield.
Converting it to block gives us alot more bang for our buck than conversion to dodge.
Block is only better than dodge up to the point where you become crush-immune with Holy Shield up. Since this is achievable already with pre-raid gear, we don't need enormous amounts of block on raid gear.
It is technically true that a fully avoided attack doesn't generate threat, but if you're in a situation where losing an occasional block due to avoidance is putting you in danger of losing aggro, then you've probably got more glaring gear issues.
My comrades are my weapons, and I am their shield.
Then it's gonna be Alliance whinning about how horde gets Seal of Blood (wich paired with Judgement of Wisdom) grants you Holy dmg similar to SoR / JoR (with a high DPS 1h) and twice the proc's from JoW. The drawback is having poor consecration / Exorcism dmg.
So this is where i plug my the suggestion i tossed in the void that was the WoW forums.. Replace Weapon Expertise with Imp. Spiritual Attutement. 2 ranks, increase mana return from Spiritual Attunement by 5/10%.
We really don't need more mana, at least for the content I've seen. My mitigation gear is about half epics and half high-end blues and my mana bar stays pegged at full on Malchezaar (and that's a fight where I have Exorcism as an extra mana dump). For anything less dangerous than that, I can just replace some mitigation gear with spelldamage plate and downrank if necessary. (Consecration is wonderfully immune to downranking penalties, so with high spelldamage, you can spam Rank 1 for substantial threat at minimal mana cost.) I haven't seen a single fight anywhere in the game where I can't reach an appropriate combination of threat and mitigation that leaves me with plenty of mana to work with.
My comrades are my weapons, and I am their shield.
I don't see any real need for a change/improvement to spiritual attunement. I've found on any boss of consequence in Kara, I've had zero mana issues. On something like prince, I literally have a full mana bar the whole time, there is literally nothing more to spend it on. If Blizzard is in fact moving us towards endgame raid tanking, then maybe more mana dumps would actually be in order...
On trash, things are a different story, but I occasionally drop a little mitigation or...taunt more mobs. Stopping to drink isn't the end of the world, or if healers are full on mana, just ask for a little spam, mine are usually quite willing to oblige.
Last edited by goss : 03/21/07 at 4:38 PM.
Reason: Beaten on basically every point by Cathela...creepy.
Well i'm going from what i have heard / read.. Appears that Hydross OT-ing adds is a really good place for a paladin to shine.. but since the adds themselves are stunnable and with the way the fight goes the paladin is best to AoE tank them.. But those adds can spike you for a lot so you can't take more mana / mp5 gear while lowering tank stats... And since you and your rogues are stunning quite a bit the mana back from SA seems to be a bit lacking...
Potions? or a better SA... I think blizzard is leaning more to a better SA.. at least the rumored 4 pc set bonus of our T6 is 10% more mana back from SA....
I run low on mana on Hydross at certain points. Simply put, because of the nature of the encounter, the mobs are stunned often, and I don't take a lot of damage (though things can go wrong, and I'll easily take upwards of 15k damage in a matter of a couple seconds, and with 1 healer assigned to me... yeah, it can suck). This generally means less heals, which in turn results in less mana. And frankly, I go all out.
Yeah, I've read that, but I don't think you can really draw a conclusion from one guild's experience on a single fight.
Don't get me wrong; if I wake up tomorrow and discover that SA's mana yield has been increased in some way, I won't be upset. I just don't that's going to do much to address our real shortcomings as tanks.
My comrades are my weapons, and I am their shield.
Yeah, I've read that, but I don't think you can really draw a conclusion from one guild's experience on a single fight.
Don't get me wrong; if I wake up tomorrow and discover that SA's mana yield has been increased in some way, I won't be upset. I just don't that's going to do much to address our real shortcomings as tanks.
Oh by all means, i wasn't getting you wrong.. i think >.>; But what i was pointing is that on our gear we all agree we want more tank stats, (dodge, block, def, armor and stam...) But a boost on Spiritual Attunement (specific to prot, wether through 2 points of talent deep in tree or as a set bonus on our t6..) might just be a key to allowing more tank stats on our gear...
Most of the changes are to Protection obviously, but with several practical buffs to Ret and something I think Holy Paladins need in Purifying Power.
Personally instead of buffing the high end of each tree (Which is sort of needed, considering the spec I'm using...), I'd much prefer if the early talents in each tree would get some 'Hybridization', you can take a look at the first few talents in the Druid's Restoration tree to see what I'm talking about.
Basically, I'd say something like combining Deflection into Improved Seal of Righteousness (And keeping it in Holy) would be a nice boost; while it wouldn't directly improve a Protection Paladin's ability to tank, it would improve the hybrid option, you wouldn't be stuck with the 10 points in Retribution.
You could argue of course that Druids can't get the bonuses of those talents at the same time, but to be honest that's true for any class; gear makes too large a difference usually to effectively allow you to fill both roles anyway.
buff /bÊŒf/ Pronunciation[buhf]
–verb (used with object)
- to reduce or deaden the force of
Given the attention currently being given to Paladins by the Blizzard development team, I figured we ought to bring some more solid figures into play so we can better pinpoint the problem.
I pilfered a rudimentary tanking simulator from a gentleman on the Burning Crusade Beta forums back in November. I can't remember his name and the code was kind of buggy, but some friends and I managed to iron most of them out (and create a few new ones with the Druid tanking simulator, not that it matters for this discussion). The code can be found here, if anybody wants to take a look at it and double check the numbers.
Into that calculator, I fed some Paladin and Warrior equipment sets detailed on this spreadsheet. The results are there, too. I did each simulation over a period of 1 million seconds (or about 11.57 days of solid combat) so margin of error should be extremely low (<0.1%, even accounting for rounding errors). The sets I selected I believe represent three different schools of thought on Paladin tank equipment: a pure "this is built specifically for Paladins" set (where possible), a set where you wear "Warrior" gear in all slots except where not possible (Tiered sets), and an imaginary set where you can wear items that are Warrior-restricted. The Warrior equipment is static throughout all tests.
I ran them both up against two targets that deal an outgoing 10k DPS pre-mitigation: one that dual wields at 1.0 swing time, and one that has a single weapon with a 2.0 swing time. This is to simulate the supposed "Paladin-favored" dual wield scenario where Redoubt can really shine.
I found the Protection Paladin vs. Protection Warrior results pretty unsurprising. In the pure Paladin equipment against the dual wielder, we get our shit ruined: you die 45% faster, and take 29.6% more damage. In the hybrid equipment it's a bit better, but still a lot more of the same: 29% faster death, 17.8% more damage taken. Only when we test the impossible Warrior-only equipment do things start to look better: 7% faster death, 0.5% more damage taken. I would also detail the two handed weapon simulation, but the Paladin just gets dominated on all levels, so I won't bother other than to say I think it's an acceptable margin between a Protection Warrior and a Protection Paladin. You can do the math if you're so inclined.
What was surprising is how the Paladin stacked up against the Arms Warrior. Take a look at the data from trial 1.2 (realistic Hybrid Paladin gear, dual wield target). The Protection Paladin barely eaks out The Arms Warrior in total damage taken (2%) and is still behind in time to death (4%). On the two handed weapon test, the Arms Warrior handily beats the Protection Paladin in both areas. In the "Warrior" gear the Protection Paladin is clearly superior.
I think this pretty clearly shows an itemization gap we need to somehow account for. There are two possible ways of doing this: reitemizing Paladin tanking equipment, or changing Paladin mechanics to allow for a broader range of required stats. I'll explore both of those another time, but for now let's chew on this information and hopefully somebody can troubleshoot this simulator to see if any bugs exist.
EDIT: It should bear mentioning that I assumed away the use of Demoralizing Shout and Thunderclap. If those were included in the Warrior simulations they would absolutely destroy the Paladin in all scenarios, with all talent builds, and in all types of Paladin equipment. As a result I don't have a problem with working towards having a Protection Paladin equaling a Protection Warrior only in a bubble-world comparison like this, because they will always have access to tools that significantly reduce incoming melee damage, while we will not. The addition of those two abilities, plus Shield Wall and Last Stand will always mean Warriors are the tank of choice on raid encounters, even if Paladins are their relative equals. Or at least close to it.
Interesting work, Theras. I was thinking of doing something like this myself, but you've done a far more complete job of it than I ever would have.
One thing I notice is that for the two-hander simulation in "warrior" gear, the Paladin takes more crushing blows than the warrior. This really shouldn't be the case in practice for that gear, since the paladin in that case has enough avoidance to eliminate crushes with Holy Shield alone, and 4 blocks per 10 seconds should be at least as good as 2 blocks per 5 seconds.
I think the problem is that the sim doesn't intelligently conserve the GCD to guarantee having Holy Shield available. For example (assuming I'm following the code correctly), if the sim reaches a point where the GCD is free and Holy Shield still has 0.5 seconds left, the sim will go ahead and do something else (refresh SoR or consecrate), triggering the GCD and guaranteeing that HS will be down for one second. (The warrior portion of the sim doesn't have this problem, since Shield Block doesn't trigger the GCD.)
This should be easy to correct by adding another branch right after the holy shield trigger, like so:
My C is a bit rusty, but I think that should do the trick. Technically, you could allow the sim to judge SoR but not refresh it during the "wait period" but that's probably more effort than it's worth, since we're concerned with mitigation here more than threat.
Or we could just petition the devs to take Holy Shield off the GCD.
My comrades are my weapons, and I am their shield.
This is why, though announced changes to itemization are heartening, I don't think we'll really see much in terms of becoming more wanted as tanks until the talents or base skills of the class are given a noticeable boost. Even if they itemized things perfectly, which is doubtful, there's just too many gaps(7% dodge?!) to catch up with at the moment.
This is why I take issue with statements like:
Originally Posted by Tseric
While there may be a flavor distinction between tanks, we don't want to (or want the players to) find some rigid hierarchy by which classes are measured in their tanking potential to the third decimal point.
I'm not sure what this means, unless they give the paladin some sort of truly unique mechanic that gives substantial mitigation in certain conditions(a lot of our abilities fit that description in theory, but none so far in practice). Otherwise spreadsheets like this expose just how easy it is to do exactly that.
With all the content they are trying to stuff into the next patch, let's just say I am not feeling particularly optimistic at the moment.
EDIT: This codelooks nearly identical to php code I wrote for doing the some time ago, though I was more concerned with avoidance and crushing and I never really got around to putting damage values and actual gear in. I'll see if there's anything I had in there that maybe exposes an error but it looks spot on.
Last edited by Snow : 03/22/07 at 6:23 PM.
Reason: additional info
Aha! That explains a lot. I also seem to have neglected to consider that Mark of the Wild will also give you 459 Armor, but I doubt that will change the results. But for the sake of completeness, I'll re-run the tests in a bit.
Numbers have been rerun with improved GCD usage by the Paladin (and proper MotW stats), and I yielded the results on this sheet. I won't bore you with a full analysis, but I'll just post a summary of each trial using the Paladin as a baseline:
Two Handed target, Warrior Gear
Prot Warrior lives 10% longer, takes 3% less damage
Arms Warrior lives 11% shorter, takes 15% more damage
I'd like to see this tuned to the point where the Warrior's advantage in damage taken vs a 2-hander is the same as the Paladin's advantage in damage taken vs a dual-wielder. Still a little ways to go, but we can hope.
Then of course they actually need to give us decent gear.
My comrades are my weapons, and I am their shield.
I've got a few ideas to possibly solve the problem, but it's taking time to properly formulate and proof them. Then I need to do regression tests to previous tiers and types of equipment, too. Tough stuff.
Great work Theras. Perhaps you should let the arms warriors have something like a 41/5/15 spec as I think that the large majority of raiding warriors have improved shield block.
Also I'd change alot of those gems to solid star of elune, not sure how much that would effect results but it would protect you from Gurgthock's wrath (he has a bit of a thing about it). Some of the agility in the paladin gems and enchants should be replaced though.
Perhaps you should let the arms warriors have something like a 41/5/15 spec as I think that the large majority of raiding warriors have improved shield block.
I'm taking a "purist" approach and testing against a pure DPS built Arms Warrior, kind of like this. Supposedly this type of build is the baseline Blizzard balances tanking off of, so it makes a good control. Plus if we're being beaten by a pure DPS specced Warrior, I think we've got some serious problems on our hands.
Originally Posted by Fjord
Also I'd change alot of those gems to solid star of elune, not sure how much that would effect results but it would protect you from Gurgthock's wrath (he has a bit of a thing about it). Some of the agility in the paladin gems and enchants should be replaced though.
I chose to observe socket color restrictions, even though such a restriction doesn't exist in game. Though if you ask me, it should, god damn it (I have a bit of thing about that, too). I'd be happy to re-run figures entirely with Solid Stars of Elune if you like, although I doubt it will really affect the results significantly. As for the agility enchants, I don't really think there are better tanking enchants available for those slots. I am open to suggestions, though; if we can get better figures, we can make better solutions.
Two Handed target, Paladin Gear
Prot Warrior lives 47% longer, takes 31% less damage
Arms Warrior lives 21% longer, takes 11% less damage
Two Handed target, Hybrid Gear
Prot Warrior lives 31% longer, takes 20% less damage
Arms Warrior lives 8% longer, takes 2% less damage
Two Handed target, Warrior Gear
Prot Warrior lives 11% longer, takes 4% less damage
Arms Warrior lives 9% shorter, takes 13% more damage
Swapping the sockets appears to yield very close results to before. The Arms Warrior seemed to fare slightly better, and the Protection Warrior slightly worse (at least compared to the Paladin) by making the swap. In the context of the comparison we are making, I don't feel the switch will produce any more or less meaningful results.
Now if you'll pardon my absence for a few hours, my wife is about to serve my testicles to me on a platter if I don't get my ass to bed.
Assuming a lack of demo/TC is like assuming a lack of fort/mark. Regardless of who's tanking, those debuffs will be up. Plus, what happens to the paladin's TPS when they have no spell hit or spellpower (all warrior gear)?
Assuming a lack of demo/TC is like assuming a lack of fort/mark. Regardless of who's tanking, those debuffs will be up.
I think you missed the point: assuming something away doesn't mean I assume it doesn't exist. I don't model those two abilities because they don't affect the results in a meaningful fashion, as I am not modeling real mobs, I am modeling player stats.
Let's say I decrease DPS by 15% (down to 8500) and increase swing time by 10% (up to 1.1s and 2.2s for each trial). That's swell, and it yielded different results. Now just for kicks I'm going to compare a 11765 DPS target with both a 0.91s and 1.82s swing times. Now we're back to 10k DPS and 1.0 and 2.0 swing times again.
Why didn't that matter? I am already testing swing times different by a factor of four, and multiplying them by a constant to yield different still swing times doesn't tell me anything I don't already know: fast swing times favor a Paladin, slow swing times favor a Warrior. It's extraneous information. I guess another way of saying it is that I am trying to take knowns and use them to determine unknowns, not take knowns and use them to determine things I already know.
Originally Posted by ildon
Plus, what happens to the paladin's TPS when they have no spell hit or spellpower (all warrior gear)?
It takes a dive. Also something we already know, and also something beyond the scope of this experiment. The intent of the model is to prove that an itemization problem exists by putting a Warrior and a Paladin in identical equipment, and getting nearly identical results. The only kink is that those items don't exist for a Paladin, hence the problem.
Assuming a lack of demo/TC is like assuming a lack of fort/mark. Regardless of who's tanking, those debuffs will be up. Plus, what happens to the paladin's TPS when they have no spell hit or spellpower (all warrior gear)?
Paladins can pick up a +150 spell damage sword, stick +40 spell power enchant on it and use Judgement of the crusader for +190 spell damage. That a total of +380 spell damage, before you put a single non-warrior piece of armor, gem or enchant on. Giving up any tank stats for +Spell damage makes us less effective tanks, through lack of mitigation, avoidance, and raw HP.
03/21/07, 12:28 PM
