Dual Wielding DKs: Demystifying the Myth.

[lyd]

So. The cost of dual wielding is 18.2 additional parryable attacks per minute, or a 31.9% increase in parryable attacks, assuming the same DK switches from an Inevitable Defeat to a brace of Broken Promises and assuming that Howling Blow has no cooldown in 3.0.8. This number scales upward and downward with expertise until you are fully capped against parry. No amount of expertise negates the percentage increase in parryable attacks until you have zero parryable attacks.

your bolded statement is very misleading, and devoid of any useful information.

not only that, but for all your math, you are forgetting that this mechanic results in such a small increase in damage over time on a DW tank that, frankly speaking, it's not a big deal either way. i'll explain that in detail later.

first off, 2 points:

a) 2h tanks will not be gearing purposefully for expertise because they generate, relatively speaking, a small amount of parry opportunities due to their slow weapon speed. they are likely to have a modest amount of expertise that is incidental from gear, but that's it. 2h tanks will -literally- have more rune strike procs than they can physically use, so the only genuinely parryable attacks they have to worry about are their weapon-based attacks. wasted rune strikes is actually one of the criticisms of a 2h build. but regardless of this, compared to every other tank in the game, the amount of parry opportunities they will generate is extremely low, making expertise a low priority for 2h tanks.

b) the numbers only scale if you assume that two identical tanks, one 2h and one DW, both increase their expertise at identical rates. but this is silly: in fact, this is actually useless information. it's like adding or subtracting a like amount from both sides of an equation. the net change in value on either side will always be zero, so pointing this out is basically pointing out nothing at all.


secondly, earlier in this thread, which you appear to believe has used bad math and not offered much useful information, as well as in an earlier DW thread on the general forum, the idea of using expertise to close the parry-haste gap has been heavily discussed. you saying that the percentage scales with expertise is very misleading, because it gives readers the impression that a DW tank that gears for expertise, over and above what a 2h tank would reasonably have, will not change the increased percentage of parry-hasted attacks they suffer, when in fact, all your statement is saying is: "expertise remaining constant, this percentage will remain unchanged." well, yeah.

let's review:

according to the figures i've seen, 14.25% is the accepted parry rating for bosses. with 6 expertise from talents, which most tanks will have, this goes down to 12.75. in order to close this gap completely, you would need to garner 32.79(12.75), or 419 additional expertise rating. this is possible, but fairly difficult to gear for, and takes from pretty precise planning.

using your figure of 31.9% for an increase in parryable attacks from DW, the question is this: how much expertise does a DW tank need to align their parry-haste chance with an identical tank using a 2h? this is pretty easy to figure out. assume a 2h tank has the normal 6 expertise from talents. if a DW tank with the normal 6 expertise will generate 31.9% additional parry-hasted attacks, then he needs to have enough expertise to close that gap. so we'll use your numbers again:

DW:57 parryable attacks per minute
2h:38.8 parryable attacks per minute

now calculate how many of these attacks will actually, on average, be parried, at the base 6 expertise:

DW:57(.1275) = 7.27 (rounded up)
2h: 38.8(.1275) = 4.95 (rounded up)

now increase the amount of expertise on the DW tank by 4 (i.e. 1% reduced parry chance):

DW:57(.1175): 6.70
2h: 38.8(.1275): 4.95


if we chart this out, we find the intersection between 4% and 4.25%:

DW:

+1%: 6.70
+2%: 6.13
+3%: 5.56
+4%: 4.99
+4.25%: 4.845

so to close the gap in parry-hasted attacks, a DW tank needs approximately 16 additional expertise (131.16 expertise rating), past what an identical 2h tank would have, in order to have no more parryable attacks than the 2h tank (.04/minute additional parryable attacks is negligible). note that at +17 expertise, the DW tank actually surpasses the 2h tank in terms of survivability related to this specific mechanic.

but wait, it gets better: how much extra damage over time, percentage wise, will a DW tank with no additional expertise be taking if he DW's without additional expertise?

since we're talking percentages, we can pick a nice round number. let's say a boss is hitting plate for an average of 6k. with a swing speed of 3.0, that's 2k average dps from melee attacks.

now, let's calculate the real effect of the parry-haste mechanic on a DW tank with no additional expertise from a 2h tank:

from wowwiki:

When you parry an attack, the remaining time on your current swing is reduced by 40% of your weapon speed, unless this would result in a reduction to less than 20% of your swing time remaining

now if we really wanted to, we could get ultra-complex here and calculate the average amount of time that a parry would occur when it would result in a reduction to less than 20% of the boss's remaining swing time, but that's a little more complex than we need to get to make a valid point. so let's just use the 40% haste factor.

now we already calculated that at the base 6 expertise, a DW tank will generate 2.32 more parry-hasted attacks per minute than a 2h tank (i.e. attacks that can be parried that actually will be parried and generate a parry-hasted attack). now a boss with a 3.0 swing timer and no parry-hasted attacks will have 20 attacks per minute. at 6k per hit, that 2k dps.

now let's calculate the increase in dps for a 2h tank's average parry-hasted attacks per minute (i'm going to use some ham-handed math here, so bear with me):

20 hits in 60 seconds at 6k = 120,000 damage

4.95 attacks at 3-3(.4) = 4.95 attacks at 1.8seconds instead of 3.0

this means you will have 4.95 attacks per minute with 1.2 seconds shaved off, so that means you will have an equal number of attacks in less time. in this case, that time is -5.94 seconds. so when accounting for parry-hasted attacks:

20 hits in 54.06 seconds = 120,000 dmg

so if every 3.0 seconds = 6k damage, 5.94/3 = 1.98*6k = 11880 additional damage per minute. 11880/60 = 198 additional dps, or an increase of 9.9% in dps from parry-hasted attacks.


now let's do the same using the DW number:


20 hits in 60 seconds at 6k = 120,000 damage

7.27 attacks at 3-3(.4) = 7.27 attacks at 1.8seconds instead of 3.0

7.27(1.2) = 8.724 seconds shaved off.

20 hits in 51.276 seconds at 6k = 120,000 damage

so if every 3.0 seconds = 6k damage, 8.724/3 = 2.908*6k = 17448 additional damage per minute from parry-hasted attacks. 17448/60 = 290.8 additional dps, or an increase of 14.5% in dps from parry-hasted attacks.

so assuming two equal tanks, one uses a 2h and one DW, 14.5-9.9 = 4.6% more damage per second as a result of parry-hasted attacks.

but let's stop to think about what these numbers really mean, because percentages have a tendency to obscure the true nature of a relationship between two variables in real-world (or fantasy world) terms.

17448-11880 = 5568. a DW tank will take 5568 additional damager per minute from the parry-haste mechanic on a boss that hits for 6k every 3.0. assuming all other things are equal. 5568. are you kidding? your healers, and again, this assumes all things are equal on both tanks, which they most likely will not be, will have to find a way to heal you for an additional 5568 every 60 seconds. with +4 expertise over a 2h tank (or -1% to parry), this number goes down by ~1392 less damage per minute, until at +4%, the difference between the two becomes negligibe.

what about a boss that hits harder? this is where expertise actually scales, because no matter how hard the boss hits, every 4 expertise a DW tank has over what an identical tank using a 2h has will reduce this number by ~25%, until at +16 expertise, it becomes negligible.





conclusion: a DW tank needs +16 expertise (+131.16 ER) to close the gap in parry-hasted attacks between a 2h tank and a DW tank. and if you assume that 2h and DW tanks will be grabbing many of the same pieces anyway, the amount of real damage over time that a DW tank will be taking from additional parry-hasted attacks scales down as the base expertise rating you work from on both tanks goes up. so if you assumed that both tanks have similar gear and started from 12 expertise instead of 6: the percentage gap will represent a smaller amount of real additional damage than at a lower base expertise score, because the amount of parryable attack that both tanks generate will be universally lower. so even if the DW has no more expertise than his identical 2h counterpart, the amount of real additional damage taken will be lower at higher expertise ratings, even if the percentage gap stays the same. this is because what we are really measuring here is the additional burden DW tanks place on healers. that burden scales down as base expertise scales up. in other words: does +4.6% of an increasingly small number make a difference in relative survivability? the answer is, at even modest levels of expertise, no, it doesn't.

EDIT: also, compare these numbers to warrior tanks in terms of overall survivability. warriors generate more parryable attacks per minute than DW dk's do, due to the nature of their instant attacks. yes, they can block: but a deep-frost tank will have at least 2 mitigation cooldowns (IBF and UA) to throw in their rotations, not to mention the inherent armor bonus of frost presence itself and the 10% additional avoidance from BB assumedly makes up for whatever benefit a shield provides a warrior. and if you really want to get technical, we should be adding in the 20% attack speed reduction from talented IT, which wouldn't affect the percentage gap, but would affect the real total damage being put out.

assuming a DW tank has closed the expertise gap, and a 2h tank has avoidance in place of expertise, you are looking at an effective difference in the two tanks of 2-3% avoidance. that means for every 100 boss attacks, the 2h tank will dodge or parry 2-3 additional attacks. if a boss fight lasts 5-10 minutes, and a 3.0 swing time = 20 attacks per minute, that means a 2h tank will dodge or parry 2-3 additional attacks every 5 minutes, or an average of <1 additional attack per minute. it is ludicrous to consider this anything but a negligible difference.

[Satorri]

2 things, as Visual defended me (thank you), I was not assuming that all DKs are Unholy, my goal was simply to offer a scope of comparison. Compare an Unholy DK with a 2-hander to a Frost DK dual wielding allows me to compare my current spec (and not an uncommon one) with the proposed "best" dual wield spec. My "napkin math" was simply to show that there was a scenario in which wielding a 2-hander wasn't so different than the dual wielding in another form, since you will not actually use the same rotation, abilities, or spec for changing weapons. It's nice to set those as statics in the equation, but not accurate. Point of fact many 2-hand Frost tanks will favor Oblit because it is better threat (better single target damage with a good epic 2-hander, and less chance to fail to land even with parry chance) and they will not avoid it because of the parry liability. Similarly other tanks were put in to provide a bit of equivalency, however cursory to set the stage for DK parry liability next to other tanks.

There is an uncertainty on Rune Strike that will have a major impact on your math as well, and it is one we haven't demonstrated yet. Namely, will Rune Strike replace an off-hand swing? If it does the dual wield DK will be able to make much better use of their Rune Strikes. Your math appears sound on opportunity for RS usage, but the numbers don't ring true in calculation. You're suggesting that with a weapon that is 30% faster, you'll only be able to use 10% more RS's, when really the thing that will be obstrusive with the 2-hander is the awkward windows have having a swing timer that will be so out of phase with a faster boss swing.

On top of that, neither of our math solutions take into account the interesting compounding effect of our being avoidance tanks. We are avoidance driven and the more our gear improves the more value we get from Rune Strike, and the more parry haste becomes a reduced liability. Bear with me through some numberless logic:

Boss swings, we dodge. At this point we've avoided one attack and reduced our vulnerability on the next melee swing since it cannot be dodged or parried, which means better hit chance and no parry liability. Now, we swing, we are parried, the boss gets to squeeze more swings into the same span of time. However, the boss's extra swings are ALSO still avoidable. More swings from the boss actually increases our opportunity to avoid and proc additional RS's. So, in a way there is a decay. Rather than parries simply increasing how much we get hit, it increases our opportunity to use Rune Strike, not just our ability to use it more, and it doesn't increase the damage flatly. You say 30% more parriable swings, but I flavor that with of the swings that are parried, 15% say, if we had zero expertise, 15% of 30% means 4.5% more parries, and if we still have 50% avoidance we're further cutting the fallout in half.

In fact, dual wielding actually has a self-compounding protection factor to compensate for the increase in parry liability by reducing parry liability and simultaneously buffing threat output.

[A Man In Black]

AMiB, your math looks sound, but you make your numbers out to be bigger than they are. Slapping together some gear, gemming it, and enchanting it, using chardev (linky, if interested), I come up with 43 expertise.

You're gemming lots of pure red, without any sign of any sort of mathematical examination of whether stamina or avoidance would be better in those sockets.

Additionally, 2h tanking would also benefit from that expertise, assuming gemming expertise is superior to the alternatives.

The biggest gap in my reasoning here is the fact that I'm trusting chardev to calculate the expertise correctly, but assuming that to be correct, I hardly feel we should get all excited about one more parry per minute. That's not going to kill anyone.

The biggest gap in your reasoning is that it's acceptable to assume that any amount of additional damage is acceptable without a reasonable justification for doing so.

Plus, one extra swing in the middle of a three-drake Sarth's buffed breath and you just died and wiped. Content mostly isn't that hard yet, but you'll note I'm not specced 0/0/61 just because it's good enough.

Also, I think that it is rather unfair of you to criticize Satorri for using an unholy DK as her 2h comparison, because let's face it: most DK tanks are unholy at the moment, due to Bone Shield being so good. This may change in 3.0.8, but it may not either. That's anyone's guess. At the moment, though, unholy is the status quo.

It's bad math and it's bad science. Control all of the factors you can to make a more even comparison.

Besides, if 2h unholy is better than any sort of frost regardless of parryhaste concerns, then DW tanking is strictly inferior regardless of if it's better than 2h frost and all of this is moot.

Finally, you say that we shouldn't compare to other tank classes, as presumably, they are balanced around their increased parry generation. This is faulty reasoning, however, as Ghostcrawler has directly said that they want death knights to be able to effectively tank in either style, and their hope is for there to be both dual wielding and 2h tanks.

Ghostcrawler says lots of things, including claiming that DKs can effectively tank specced into any tree (currently not true). If DW frost is as effective as tanking blood, then GC's comment is true, but it doesn't make DW frost not inferior.

I'm not concerned about DW frost's viability. Right now, fury warriors and ret pallies can OT anything in the game, and MT many things. The margin is very large. I'm concerned about what is ideal.

Now, you might say that the balance isn't currently in place to make that possible, and that may be the case, which is why we're discussing the matter. However, if we are willing to assume that DKs might not be balanced for the kind of parry generation DW provides, we can't assume the other classes are balanced for a similar level of parry generation. If we doubt one class, we need to doubt all.

Start a "Warriors/druids/paladins are not viable because of parryhaste" thread with some good math and you'll get a heck of a discussion. It is, however, a different discussion. (The math on other classes here is Not Very Good, on top of everything.)

In the meantime, we have two nearly identical cases, one which takes more parryhasted attacks and one which doesn't. It behooves us to examine carefully the gains of dual-wielding, if any, and the opportunity costs of dual-wielding, if any, and see where the balance lies.

Not only has nobody shown that DW tanking isn't strictly inferior, nobody has even suggested any reasons it might be superior, save for the occasional handwaving about threat.

2h tanks will not be gearing purposefully for expertise because they generate, relatively speaking, a small amount of parry opportunities due to their slow weapon speed. they are likely to have a modest amount of expertise that is incidental from gear, but that's it. 2h tanks will -literally- have more rune strike procs than they can physically use, so the only genuinely parryable attacks they have to worry about are their weapon-based attacks. wasted rune strikes is actually one of the criticisms of a 2h build. but regardless of this, compared to every other tank in the game, the amount of parry opportunities they will generate is extremely low, making expertise a low priority for 2h tanks.

Does this mean that 2h tanks will be gemming/enchanting more stats that are more generally useful than DW tanks will be, while DW tanks will be expending resources to offset their disadvantages?

Doesn't this imply that DW tanks have significant disadvantages due to dual-wielding, and need to compensate for them? Either parryhaste is enough of a concern for DW tanks to cause them to gem to compensate, and thus it is of enough concern to discuss here, or it isn't a concern and both specs can be assumed to gem/enchant the same way.

As for your math, you just said that the DW tank is inferior by enough gems to take 4 full expertise points. That's kind of damning if your math is good, and it kind of smells fishy.

There is an uncertainty on Rune Strike that will have a major impact on your math as well, and it is one we haven't demonstrated yet. Namely, will Rune Strike replace an off-hand swing? If it does the dual wield DK will be able to make much better use of their Rune Strikes.

No. This is trivally tested. Go pull many weak-hitting mobs one by one, put a very slow weapon in your mainhand, a very fast one in your offhand, and Rune Strike them all to death.

Your math appears sound on opportunity for RS usage, but the numbers don't ring true in calculation. You're suggesting that with a weapon that is 30% faster, you'll only be able to use 10% more RS's, when really the thing that will be obstrusive with the 2-hander is the awkward windows have having a swing timer that will be so out of phase with a faster boss swing.

You have estimates based on your feeling of what might be right. I have done some (fairly crude) modeling, and it's obvious that a linear increase in weapon speed does not result in a linear increase in RS opportunities. Do superior modeling or point out an actual flaw in my model.

On top of that, neither of our math solutions take into account the interesting compounding effect of our being avoidance tanks. We are avoidance driven and the more our gear improves the more value we get from Rune Strike, and the more parry haste becomes a reduced liability. Bear with me through some numberless logic:

Boss swings, we dodge. At this point we've avoided one attack and reduced our vulnerability on the next melee swing since it cannot be dodged or parried, which means better hit chance and no parry liability. Now, we swing, we are parried, the boss gets to squeeze more swings into the same span of time. However, the boss's extra swings are ALSO still avoidable. More swings from the boss actually increases our opportunity to avoid and proc additional RS's. So, in a way there is a decay. Rather than parries simply increasing how much we get hit, it increases our opportunity to use Rune Strike, not just our ability to use it more, and it doesn't increase the damage flatly. You say 30% more parriable swings, but I flavor that with of the swings that are parried, 15% say, if we had zero expertise, 15% of 30% means 4.5% more parries, and if we still have 50% avoidance we're further cutting the fallout in half.

No, no, no, no, no, no, no.

Being an avoidance tank does not somehow make taking more swings acceptable. In fact, it makes it less acceptable.

In progression raids, tanks do not die because they took so much damage over X period of time that the raw healing per second could not keep up with it. Tanks die because they take unexpected spikes of damage that overwhelm the healers' ability to compensate. This is a natural problem of avoidance tanking, because no amount of avoidance (short of 100% avoidance) can guarantee that you will never get n+1 hits in a row without an avoid, where n is the maximum number you can possibly survive. Increasing your chance to randomly take an extra attack is even more foolish given this.

This is partially why comparisons to other classes are distractions; DKs rely on their innate avoidance in ways classes lacking that innate avoidance do not.

In fact, dual wielding actually has a self-compounding protection factor to compensate for the increase in parry liability by reducing parry liability and simultaneously buffing threat output.

How on earth does increasing your number of parryable attacks reduce parry liability?

And where's your math to show that dual-wielding is more threat?

There's lots of handwaving here about "Oh, the DW tank will just get expertise to compensate." All stats come at a cost. What are you sacrificing for that additional expertise?

[visualdeity]

You're gemming lots of pure red, without any sign of any sort of mathematical examination of whether stamina or avoidance would be better in those sockets.

There are only 3 red gems that this really applies to. I gem to the color of the socket, period, unless I *really* need a stat. Now, one could argue that I could stick 8 expertise/12 stam (I believe that's the number?) gems in there, which is reasonable. But I'm not going to stick pure stam in, nor does anyone consider it worth gemming for parry, ever. Dodge gems are an option, but on the whole, one can't say they're more valuable than expertise gems, given the extra threat that causes for you, in addition to the slight extra avoidance. Equal? I'd give you that. Superior? Hell no. Expertise is just plain an excellent stat to gem for, as any spec, because it increases your survivability slightly while increasing threat. You're seriously picking nits if you're trying to take issue with gemming for expertise, which is universally considered to be a good gem choice.

But hell, let's go without expertise gems. That still leaves 29 expertise from the gear and talents, which translates to 1.4 more parryable attacks per minute. No. This is still not a big deal. 1 more attack per minute is MINISCULE, and not worth worrying about in the least.

Additionally, 2h tanking would also benefit from that expertise, assuming gemming expertise is superior to the alternatives.

And? That doesn't have anything to do with it. This topic has nothing to do with trying to prove DW to be superior to 2h, it has to do with whether or not it's suicide to DW tank. Any effect it has on 2h tanking to gem for expertise is irrelevant to this discussion.

The biggest gap in your reasoning is that it's acceptable to assume that any amount of additional damage is acceptable without a reasonable justification for doing so.

On the contrary, I would say that what you just said is the gap in your reasoning. WoW is not, and has never been, a game where you need to min/max to get ahead. The raiding game is almost all about coordinating your players, and understanding how to down the boss. Squeezing an extra 1% out of your raid is going to help you down the boss a little sooner (maybe), but it's hardly required. Skill can more than cover any gap min/maxing covers.

Min/maxing is the silliest thing raids have ever done, and continues to be silly to this day. Taking a miniscule amount of extra damage (one whole extra hit per minute, oooooooooh) is not something anyone should care about in the least.

It's bad math and it's bad science. Control all of the factors you can to make a more even comparison.

It is neither. The only reasonable comparison, whatsoever, is between the most common way to play each variant. Unholy is currently the most common way to 2h tank (by far), thus, it is the ONLY reasonable comparison. Comparing to 2h frost is setting up a straw man, because you're comparing against a situation that doesn't actually occur in common practice.

Ghostcrawler says lots of things, including claiming that DKs can effectively tank specced into any tree (currently not true). If DW frost is as effective as tanking blood, then GC's comment is true, but it doesn't make DW frost not inferior.

The blood tanks tanking 25-man content would beg to differ. I'm not one of them, but it's rather dishonest to say that it can't effectively be done, when people are, in fact, effectively doing it.

But this misses my point: if you wish to say that other tanks are balanced around their increased parry generation, you MUST accept the same for DW death knights, because this is a role Blizzard intends them to be able to play, thus they should be balanced for it. You can't have it both ways. Either accept both without question, or question both. End of story.

I'm not concerned about DW frost's viability. Right now, fury warriors and ret pallies can OT anything in the game, and MT many things. The margin is very large. I'm concerned about what is ideal.

On the contrary, as I've stated using YOUR math, the margin is tiny (if you mean between DW and 2h). You throw down a percentage like 30% like it actually means something, but the truth is, percentages mean very little when working with small numbers. If I have a penny, and you give me another penny, that's a 100% increase in my wealth, but it doesn't make a damn bit of real difference. The percentage is misleading, as your cited percentage is.

And you are, to your credit, honest that you are concerned with what is ideal, not what is merely acceptable. However, in that case, accept that the game you're talking about is an imaginary game where such extremes are necessary. In the real game, ideal is never required in the least, only acceptable is.

Not only has nobody shown that DW tanking isn't strictly inferior, nobody has even suggested any reasons it might be superior, save for the occasional handwaving about threat.

By and large, no one gives a damn if it's inferior or superior. We care if it's close. There's a theoretical best spec for DPS at the moment, but the fact of the matter is, it doesn't matter if you do 2000 DPS, or 2010 DPS. The performance has, at this point, been shown to be so damn close that it doesn't make any meaningful difference. The difference is academic. I stand by my statement: if you are truly worried about one extra parry every minute, you are entirely too focused on min/maxing, and need to see the forest, not the trees.

Those who oppose DW tanking speak of it as if DW tanking is the equivalent of taking off half your armor and trying to tank that way. It is nothing close to such an extreme scenario, and that's all this thread is trying to get across.

[A Man In Black]

Min/maxing is the silliest thing raids have ever done, and continues to be silly to this day. Taking a miniscule amount of extra damage (one whole extra hit per minute, oooooooooh) is not something anyone should care about in the least.

If this is how you feel, feel free to stop reading my posts on this subject, as I am speaking to those who feel that taking extra damage every minute, that doing your job in less than the most efficient manner, is not acceptable.

DW tanking is the equivalent of trying to raid tank with 539 defense. It's possible, but you're going to randomly die to spikes a bit more than someone who's doing it correctly. WOW is a math problem; there is a right answer.

[GravityDK]

I think Visual has a good point. Min/max raids culminated and ended in Sunwell.

Now WotLK has raid-wide buffs, heterogeneous buffs, boosted "utility" class' dps, no drum-stacking, no heroism stacking, etc.

I think min/max philosophy is only mandatory for the very top 5% or so of raiding guilds, where the lowest common demoninator is no longer player skill but intead gear and tiny optimisations.

Some people adopt a min/max approach to their personal behaviour, which is their choice.

However, everyone else can tolerate 1 extra hit per minute, given that's all it appears a DW tank will receive when they have added the minimal required 16 expertise (=131.16 expertise rating) beyond those 5 provided in talents.

I think the maths is strong that DW tanking does not equal instant doom, which was prreviously the widely-held opinion.

From what I read here in Lyd and Sartorri's analysis, DW tanking is only very marginally worse than 2H tanking.

[Satorri]

And thank you Gravity, as that was all that was trying to be demonstrated.

[mononoke]

just wanted to say thanks for this thread, it encouraged me to try dual wield tanking whilst everyone else was calling me a nab

now im maintanking naxx with my guild my guild up to Sapphiron, to whom we got today with food (Rhinolicious Wyrmsteak) i have 26 expertise and number of parries is really low (on side note i don;t get parried at all in heroics). healers say its no different healing me from when i was using 2h and i personally see dualwielding more responsive when i need to switch target and start building threat on it fast+rune strike went from 4th place to 2nd in my damaging abilities on Recount.

here's link to my armory
The World of Warcraft Armory

thanks for thread and keep up the nice work here, your vid's+forums are a great help to starting tanks as myself (death knight is my first tank char, i gave up on 59lvl warrior, was quite painful to lvl pre-patch 3.x

[Krenian]

Well one thing that I'm happy to have done is to have this thread open up ideas and discussions about the ability in which people have been curious about. We'll just have to see how well this idea flies or how long for that matter.

It's well noted that I don't play my DK as a tank, as I am much stronger in a DPS rotation. However with members like Satorri, it helps a lot of the causes for people who need information about the idea. Glad to see you guys can use this information and have a logical discussion of a spec.

[A Man In Black]

However, everyone else can tolerate 1 extra hit per minute, given that's all it appears a DW tank will receive when they have added the minimal required 16 expertise (=131.16 expertise rating) beyond those 5 provided in talents.

Also assuming you have two 2.5+ speed tanking weapons, of which there are exactly three. (A quest item, Infantry Assault Blade, and Broken Promise. I assume you're not foolish enough to use a green of the Champion.) If you aren't, the situation is even more grim.

I think the maths is strong that DW tanking does not equal instant doom, which was prreviously the widely-held opinion.

Until it does equal instant doom. Until you take autoattack + spike ability + hasted autoattack and fall over stone dead. It happens on Malygos, it happens on multi-drake Sartharion, and it will happen in Ulduar, the Nerub raid, etc. It's not that parry haste is a marginal increase in overall damage; it's that parry-haste causes unpredictable spikes in damage. Amortizing spikes over the overall damage of a minute misses the point of why spikes are bad.

If you'd like another way of looking at things, one extra hit a minute from that two-second swing-timer boss is like losing 2.8% avoidance from your gear.


On top of this, there's lots of people asserting their God-given right to be suboptimal, without bothering to articulate why. Dual-wielding only benefits in DPS because of powerful talents like Killing Machine and Blood-Caked Blade, the latter only increasing your parry liability. Without those, you're ending up generating less threat than a 2h-wielding DK, regardless of spec.

So. So what advantages does dual-wielding give that are worth downgrading your Naxx ring to a level 78 blue, or losing three talent points worth of avoidance talents?

[mononoke]

... it happens on multi-drake Sartharion, and it will happen in Ulduar, the Nerub raid, etc. It's not that parry haste is a marginal increase in overall damage; it's that parry-haste causes unpredictable spikes in damage....

dude, maybe you ned to try things before saying untruths. i have maintanked Sartharion while dualwielding without any problems, you just got this idea in your head that DualWielding is bad and dont wanna listen to people who proves you wrong.

simple as that, you're not right, try to deal with it it will happen many many more times in your life, i can guarantee that.

Dual Wield tanking is fully viable for raiding, same as 2h tanking is, neither one of those is superior or inferior to other, tried both for maintanking raids - only persons who could have trouble with that are our healers, which didnt notice any change in increased need of healing me when i started to dual wield tank.

[Satorri]

A Man in Black WANTS it to not be reasonable, that clearly doesn't include trying to make it work to test his mathematically supported ideas.

I wouldn't hold it against him though. We're dealing with a setup that has so far been sketched with more reasons not to than reasons to, and testing something sub-optimal is only for the people who want to do something that others don't think can be done.

People are doing it, they're not exploding in terrible spikes of damage. It is not wildly unreasonable to do so, and no one is questioning that you are opening yourself up to certain vulnerabilities.

We're past the point where continually crying wolf and doing your own napkin math will talk people out of it. If you really want to discourage people (not sure why), go take any set, take your current standard DK tanking gear, grab a couple tank swords, slow/slow, slow/fast, fast/fast, whatever you have available, and go tank something. Post the WWS and we can deal with raw data (a screenshot of your equipment, stats, spec, etc would be helpful).

Let's take this out of the realm of the "I don't like the idea," and start posting some real information.

For anyone who doesn't know how to WWS, when you're in game type "/combatlog" into your chat line, you'll get a confirmation message. When you're done collecting, type it again. When you log off (won't be able to remove the file until then), go into your WoW directory, into the "Logs" folder, there will be a file called WoWCombatlog. Take that file and drag it to your desktop. Go to:
Wow Web Stats.

I created an account so we can put all of the information under one heading:
Username: DKDW
password: idualwield

Pleast use it only for this data collection. When you go to the site and log in there is big button to "Start Client." This will open a Java applet. You enter the path to your combat log (on your desktop or wherever you put it). On one tab you'll need to enter the website login, though it will fill on future use automatically. Please enter in the comments who created the data, and a short description of where, such as Naxx10, ObSanc25, etc.

After you host it, you'll have the data compiled on a central page that you will be taken to, you can link your WWS back here to us on this post/forum, and we can see the exact breakdown of your damage taken (and done) as well as your amount of attacks parried.

[visualdeity]

A Man in Black WANTS it to not be reasonable, that clearly doesn't include trying to make it work to test his mathematically supported ideas.

I wouldn't hold it against him though. We're dealing with a setup that has so far been sketched with more reasons not to than reasons to, and testing something sub-optimal is only for the people who want to do something that others don't think can be done.

The frustration that I (and presumably mononoke) have with AMiB is that he's not working within the real world. WoW isn't a game where the goal is to make the most efficient character/raid humanly possible (for most, there are some who enjoy this). Maximizing yourself is a means, not an end. The end, for most of us, is to complete the raid with little hassle. Is it necessary to 2h tank to accomplish this? No. Then, there's no reason to insist that people must do it. Hell, given the theorycrafting I've seen, 2h vs DW is a completely moot point, because Death Knights don't math out to be the best tank. If we approach this from the approach of "make the best tank", as AMiB is doing, then the discussion may well already be over.

Furthermore, he uses small numbers as though they were big ones. Yes, assuming all the wrong conditions, you will have a wipe that wouldn't have occurred if you were 2h tanking. If, if, if, IF. Let's not forget that the chance of all these things lining up is relatively small. If I buy lottery tickets continually, I'll eventually win, but one doesn't bank on such a chance. Yes, the chance of winning the lottery is lower than the chance AMiB speaks of, but the fallacy is the same. Speaking of a relatively small chance as if it were a certainty is misleading at best.

The most important thing he misses, however, is the very real factor of enjoyment. This is an intangible, it's true, but it's a very real factor that each and every one of us must weigh when we consider decisions. WoW, contrary to what AMiB said, is not a puzzle which has one right solution, and that solution itself is the goal. The goal is to have fun, which is a ridiculously hard thing to define, and there is no "right" way to do that. What matters in this thread is not whether dual wielding is optimal, but whether it's viable. We have pretty neatly proven that it is, and whether you choose to go that route is a personal decision where you weigh: is the small loss of performance worth any extra fun I might have doing it this way?

AMiB's reasoning is seriously off the mark, and can be likened to getting whatever job pays the best. If your goal is to maximize the amount of money you have, blind to all other considerations, then you're liable to wind up in a job you hate... but you'll be making more money, damn it! This might be acceptable, to most of us, if the job pays $100,000 per year more, but the gap between dual wield and 2h tanking is something on the order of $700* per year more. Would you sacrifice a lot of job satisfaction for $700 per year? I wouldn't, but AMiB presumably would, based on his posts here. Which is anyone's right: because one of the factors is intangible, there is no "wrong" decision, so if someone wants to make that sacrifice, more power to them, but they'd better damn well not get on their high horse and pretend that those who don't sacrifice their fun for a small gain in performance are objectively wrong.

WoW, as with most activities in life, is not something that has a right or wrong answer. It's not a black and white game, and how far along the gradient you're willing to slide is up to you. AMiB, though, argues as if it were a black and white game, which is the frustration. He is ignoring the fundamental truth of the game (hell, life), which makes what he's saying rather silly.

*2% of my current salary, which is a rather good salary for the cost of living in my area. Your cost of living is probably higher, so adjust the number to compensate if you think my number is too small.

[A Man In Black]

dude, maybe you ned to try things before saying untruths. i have maintanked Sartharion while dualwielding without any problems, you just got this idea in your head that DualWielding is bad and dont wanna listen to people who proves you wrong.

I'm not sure if you read what I wrote. Sartharion is a kitten until you're trying to cope with taking 25-30K damage spikes. I've tanked no-drake Sartharion as fury.


Again, I'd be curious to see any argument that shows what advatanges dual-wielding has, other than "It makes me a beautiful and unique snowflake." Nobody disputes that there's a parry-haste disadvantage, or even really the degree.

So. Without constructing hilariously weird analogies, what is the advantage that is worth wasting talent points, additional gear drops, and about 3% avoidance? Or should I write "DK frost tanking without Frigid Deathplate: Demystifying the Myth of taking tanking talents"?

[visualdeity]

I'm not sure if you read what I wrote. Sartharion is a kitten until you're trying to cope with taking 25-30K damage spikes. I've tanked no-drake Sartharion as fury.


Again, I'd be curious to see any argument that shows what advatanges dual-wielding has, other than "It makes me a beautiful and unique snowflake." Nobody disputes that there's a parry-haste disadvantage, or even really the degree.

So. Without constructing hilariously weird analogies, what is the advantage that is worth wasting talent points, additional gear drops, and about 3% avoidance? Or should I write "DK frost tanking without Frigid Deathplate: Demystifying the Myth of taking tanking talents"?

The talent point claim is a lie. You take 3 points from annihilation and put them into nerves of cold steel (if you're transitioning from unholy to frost, the situation isn't quite so simple, but regardless, it's not like you're giving up clutch tank abilities to grab your DW talent). No fuss, no muss.

The gear claim is rather pointless as well. One of your weapons is free, any tank is going to take one. The other isn't, but then again, it's one drop. You can live without until all the tanks who need one, have one, or make one of them wait. Either way, it's not going to be a huge burden, either on you or the tank who has to wait in line.

And to your last point, not only is your continued (and stubborn against all rationality) use of percentages misleading to make it sound bigger than it is... a 3% decrease in avoidance is still nothing to get worked up about. You will not cause undue difficulty to yourself or your raid by taking an extra hit every minute. It is strictly worse, but only in the same way $970 is worse than $1000. Strictly worse, yes, but not an appreciable difference in the least.

You discount any potential fun gained as irrelevant, when it's anything but. This is a game, and we play it for fun. If you can gain an increase in fun by taking a miniscule performance hit, that might be worth it. THAT is what this thread has been about for a while now.

It was agreed ages ago that DW is very slightly worse than 2h. That was never the concern, however. The concern was whether it is reasonable to DW. It has been shown, beyond the shadow of a doubt (for now, actual in-game data may contradict the analysis done so far) that it is reasonable to DW tank.

You've been painting this dire picture all along, as if DW tanking were akin to a warrior trying to tank with a 2h weapon - completely suicidal. It's not, and Satorri has shown it's not (hell, your own analysis of the matter shows that), but you're so hell-bent on upholding your conclusion that it's a complete failure that you've lost the forest for the trees. In the end, this all comes back, in the real world, not the math world, to 1 more hit per minute. If you die because of one more hit, either your healers are crap, or you got extremely unlucky. It is NOT A BIG DEAL.

[arthoer]

hi there, my first post.
Reason why i registered is because i felt like sharing my experience as 25 raiding DK Dual wielding tank.

This topic helped me a lot about trying dual wielding. It made me decide to give it a try and spend some dkp on dual wield stuff.

If you just look at my armory:
The World of Warcraft Armory (sometimes i am wearing my dps gear, just so you know)

you will see that i have a good setup for dual wielding. Though i still lack some expertise, but atm you could just neglact that.


So how is the dual wielding?
Well when keeping it short; I raised my TPS from 3500 to 5000 (its around the same as my MT warrior Rikp does).
- so using commen frost tanking specc and the titansteel mace gave me a base of 3500 TPS on 25 men bosses.
- Switching to dual wield specc (still tweaking and trying some speccs) and using Broken Promises and Slayer of the lifeless i raised it to 5000.

As we all know, TPS is important. So dual wielding IS an increase.

HOWEVER: comparing those 2 swords with the titansteel mace is kind of retarded. Since i dont have a very good two hander with 200dps or more i can't give a perfect comparison for you guys.

Anyway, dualwield tanking is VERY viable, you do need a good couple of swords though!


Also, i finally got def capped (kind of hard for a deathknight, as you all might know), since i got more defense now, so i could switch a lot of gems with expertise ones.

My avoidance also increased a lot, but thats just commen sense.


Still, with the upcomming patch, blacksmiths will have nice 2handers for tanking, and you will have the tanking rune enchant. So 2handers might get to be a little better. Then again, runestrike will get an 150% threat increase, and since you are spamming runestrike while dualwielding... it will prolly not matter if you use 2handers or dualwield.

All in all, try it out and see for yourself. If you have any questions or comments, i will be keeping an eye on this threat for a few weeks.

[arthoer]

ow and putting points in killing machine is really worth it, just be sure to get a nice base for crit, like 10%.

In raids you get enough crits from buffs etc. Activating killing machine once in two rotations is good enough.

Just fun to hear fellow deathknights telling me that killing machine sucks hehe.


Ow and yeah, dual wielding does suck a bit in heroics! your tps will prolly drop to 3000. Since you are less buffed and mobs less debuffed

[Ceravantes]

I think Visual has a good point. Min/max raids culminated and ended in Sunwell.

Who is to say that will always be the case, we don't know what is around the corner.

[visualdeity]

Who is to say that will always be the case, we don't know what is around the corner.

True, but Blizzard has been heavily pushing their "bring the player, not the class/spec" mantra, which is a philosophy that is going to have strong influences on their dungeon design. I'd say it's a good bet that we will not see any need for min/maxing in this expansion. Certain? No, but to me, it's the most likely outcome.

Besides that, we shouldn't try to compensate for the future too much. Let's deal with today, and if the future requires us to change our direction, we'll deal with that then.

[Ikarus]

Hi guys,
I finally registered after lurking for some time. Please excuse me if my posts are hard to read, English isn't my maternal language.

Anyway, what's the intention of this thread? For 5 pages, some of you have posted math in support of DW tankings viability. After AMiB has proven some of this math to be skewed (and he hasn't gotten all of the skewed math), you are bashing the idea of using math in the discussion.

If you want to show that DW tanking is possible (or viable), simply state that you have tanked every encounter as DW without your raid complaining, maybe post a WWS. That's enough.

As soon as you started compairing DW and 2H tanking on an mathematical base, well, you started compairing DW and 2H tanking based on math.

So, what's the intention of this thread? Success stories of DW tanking or serious theorycraft?