WAR MT + PLD OT vs. PLD MT + WAR OT in terms of DPS

[Fue]

Yes people tend to underestimate sword oath.
It also scales from weapons autoattack time so faster weapons make it even better.

For example if you have 2s delay weapon just having sword oath on will add 25pot/s to your total dps.
Now if SE ever gave us 1s delay weapon it would double to 50pot/s.

[Ooshima]

Why are you adding in Brutal Swing's potency numbers but not Circle of Scorn or Spirits Within? And why aren't you accounting for the extra 50 potency per auto attack from Sword Oath into the total PLD potency production? (And don't forget PLD auto attacks more often than WAR) There's no question that WAR hits harder with weapon skills. The way PLD in sword oath potentially closes the gap is through off-gcd skills and passive damage from auto attacks. Talking about how high you can crit sort of misses the point.

This is kitru's old math post comparing sword oath PLD to non-defiance WAR.

I didn't include any OGCD etc, I was just solely comparing potency differences generated by standard combos under the effect of FoF vs Berserk. So nope there isn't any Brutal Swing included.

i haven't read thru this yet, but id bet most people will say what i am about to.

ive tested the exact gear between the two, and a PLD OT does better deeps than a WAR OT. This is with a WAR without defiance, and the PLD in sword oath. Its only a 5-9 % difference though, but hey, that can be the difference in an enrage.

however.....recently I've been rolling as paladin with dps stuff on trial roulette. i often get titan, and i run out of tp fairly quick. paladins have no way to regain tp as an off tank. shield swipe isn't available, because you aren't blocking shit. you'd need a bard in the party to play tp song. then you're jeopardizing the casters dps because that bard can't play foes for them while they are buffing up your tp.

id say its better to roll with the paladin as MT, because he won't have tp problems, and the warrior can go to town on his debuffs, so imo its better. paladin sword oath does do more damage though.

OT WAR can just consume as much TP due to fracture also.

[Fue]

I didn't include any OGCD etc, I was just solely comparing potency differences generated by standard combos under the effect of FoF vs Berserk. So nope there isn't any Brutal Swing included.

Difference is.
Brutal swing adds 5pot/s if its used every time its available.
Compare that to spirits and scorn that both add 10pot/s.

Thats 15pot/s that pld loses in calculations where you ignore off-GCD abilities

[Alphras]

Why are you adding in Brutal Swing's potency numbers but not Circle of Scorn or Spirits Within? And why aren't you accounting for the extra 50 potency per auto attack from Sword Oath into the total PLD potency production? (And don't forget PLD auto attacks more often than WAR) There's no question that WAR hits harder with weapon skills. The way PLD in sword oath potentially closes the gap is through off-gcd skills and passive damage from auto attacks. Talking about how high you can crit sort of misses the point.

This is kitru's old math post comparing sword oath PLD to non-defiance WAR.

Thats a really informative post, thank you. I guess this actually concludes this discussion. PLD OT and WAR MT (who keeps up Storm's Eye) is the best combination for total DPS.

[klops]

It actually will outdps warrior when hitting same target with slashing debuff. Without it the difference is neglible.

Myself (High Allagan Axe) in full STR with a SE>SP rotation against our PLD (High Allagan Blade) in full STR with storms eye debuff: ~345 versus ~300 over 4:00-4:15. I'll give the PLD the benefit of the doubt and say 305-310 is possible considering he couldn't shield swipe the dummy. Show me a PLD that can maintain 345+ in sword oath.

Combine that with the fact that a warrior cannot maintain SE>SP rotation against a full STR sword oath'd paladin and is forced into either BB>SE>SP rotation (where you lose upwards of 7.5 seconds of SE) or BB>SE rotation and the difference is not "neglible" in terms of net group DPS. Threat is the number one limiting factor in this discussion.

Furthermore, the "benefit" of having your OT never have to enter defiance or shield oath is negated by the fact that you shouldn't even be running a second tank in that scenario. For all fights that require two tanks currently (there are close to none), having your WAR OT while rotating Unchained will always result in more net DPS than having a PLD OT while rotating Shield Oath.

If you want your math (or this thread, for that matter) to be useful, apply it to actual encounter mechanics. Example: every group in the world is guaranteed at least two Dreadnoughts on Turn 8 with average groups being stuck with three. If your PLD is OT, he will be required to sit shield oath for anywhere from 30 to 60 seconds depending on the group. If you force your WAR MT to tank the dreads so that your PLD can sit sword oath, why did you even bring a PLD to begin with?

[Giantbane]

Myself (High Allagan Axe) in full STR with a SE>SP rotation against our PLD (High Allagan Blade) in full STR with storms eye debuff: ~345 versus ~300 over 4:00-4:15. I'll give the PLD the benefit of the doubt and say 305-310 is possible considering he couldn't shield swipe the dummy. Show me a PLD that can maintain 345+ in sword oath.

Is your parser accurately accounting for and attributing the extra sword oath strike? That used to be a problem, I don't know if it still is.

Since we're talking about the OT, you should be able to ignore shield swipe since that shouldn't be active.

Combine that with the fact that a warrior cannot maintain SE>SP rotation against a full STR sword oath'd paladin and is forced into either BB>SE>SP rotation (where you lose upwards of 7.5 seconds of SE) or BB>SE rotation and the difference is not "neglible" in terms of net group DPS. Threat is the number one limiting factor in this discussion.

true and important. Depends on how much of a lead the WAR can build up in the beginning with the BB > SE > unchained + berserk > BBx3 combo, and how much time the PLD spends away from the boss (tanking something else).

Furthermore, the "benefit" of having your OT never have to enter defiance or shield oath is negated by the fact that you shouldn't even be running a second tank in that scenario. For all fights that require two tanks currently (there are close to none), having your WAR OT while rotating Unchained will always result in more net DPS than having a PLD OT while rotating Shield Oath.

If you want your math (or this thread, for that matter) to be useful, apply it to actual encounter mechanics. Example: every group in the world is guaranteed at least two Dreadnoughts on Turn 8 with average groups being stuck with three. If your PLD is OT, he will be required to sit shield oath for anywhere from 30 to 60 seconds depending on the group. If you force your WAR MT to tank the dreads so that your PLD can sit sword oath, why did you even bring a PLD to begin with?

The point isn't that the PLD never has to enter shield oath, it's that he's going to be in shield oath *less* if he's OT. If he's MT, he's pretty much in shield oath the whole time, which is a big loss in DPS for the PLD for the entire fight. The WAR loses less when turning on Defiance.

The two big issues (in my mind) come down to whether or not the WAR will have troubles with threat, and whether or not you want the WAR's better DPS while tanking (and better AoE) to be utilized in burning down adds instead of the boss.

[Dano]

it's almost always better to have WAR take care of adds because of WAR's burst and AoE damage to burn adds down quick, unless the add requires special care like chain stuns / silence. And PLDs have so much cooldowns, they can tank the main boss without special attention from the healers while they heal up the off tank and party.

[TemariVtwo]

The solution to this problem is War MT + War OT 8)

[klops]

Thats a really informative post, thank you. I guess this actually concludes this discussion. PLD OT and WAR MT (who keeps up Storm's Eye) is the best combination for total DPS.

Because the numbers didn't add up to actual in game experience (where 110 STR WARs are demolishing 110 STR PLDs), I investigated the math in that thread. Unfortunately, it is hilariously inaccurate on multiple levels. One of the most glaring examples is its assumption that pacification stops auto attacks. There is a lack of distinction between weaponskills and autoattacks in berserk's formula. Berserk is not calculated correctly at all. There is also no brutal swing in the warrior's rotation while there is spirits within and circle of scorn in the paladin's. I understand that most of that post was made in september, but I am speaking to the people who are referencing it as a source here in this thread. You are using incorrect formulas to back up your arguments.

[Giantbane]

Because the numbers didn't add up to actual in game experience (where 110 STR WARs are demolishing 110 STR PLDs), I investigated the math in that thread. Unfortunately, it is hilariously inaccurate on multiple levels. One of the most glaring examples is its assumption that pacification stops auto attacks. There is a lack of distinction between weaponskills and autoattacks in berserk's formula. Berserk is not calculated correctly at all. There is also no brutal swing in the warrior's rotation while there is spirits within and circle of scorn in the paladin's. I understand that most of that post was made in september, but I am speaking to the people who are referencing it as a source here in this thread. You are using incorrect formulas to back up your arguments.

Missing 5s of unbuffed auto attacks out of every 90s is a very small portion of overall DPS. Auto-attacks are normalized to something like 100 potency every 2.8s. So that's 5 / 2.8 * 100 or 178.6 potency lost. 178.6/90 ~= 2 potency per second or ~5 potency per GCD.

Brutal swing wasn't forgotten, or ignored to slant the numbers. This was done back in September when Ifrit and Chimera were major bosses and the ability to use stuns (like brutal swing) at will wasn't always guaranteed. So it was left off. This isn't the case any longer, so it should be added back in. If you include it, that adds 100/20 = 5pot per second or 12.5 pot per GCD. To ballpark Berserk, it's an extra 20s of damage at like 40% every 90s. 20x.4 / 90 ~= 9%. Internal release is 5% for 15s every 60s, or 15 x .05 / 60 ~= 1%. So lets call that 10% extra from the buffs, so Brutal swing gives something like 13.75 potency per GCD.

Between the two that adds a grand total of 18.75 potency per GCD. That still puts the WAR damage below that of a PLD. But it doesn't end there.

There are also a couple of other important notes in there that you sort of didn't pay any attention to. The math was done with SE>BB instead of SE>SP that was used in your "in game experience". So the math is actually OVER estimating the damage for the rotation you used. SP does 40 potency less total over the 3 move combo. So over 6 moves (SE>BB) that's a difference of 40/6 ~= 6.7 potency per GCD. It's a little higher because of berserk + internal release and a little lower due to frac usage every 30s. Either way it still cuts a chunk of damage that basically more than nullifies the "hilarious" mistake in auto attack damage calculation.

The other important thing is that the PLD damage is also underestimated. It calculates the sword oath bonus only once every GCD, but every PLD weapon is faster than the GCD. That means that the sword oath bonus should be higher. My PLD using a break blade attacks once every 2.16s for 50 potency. So that's 50 x 2.5 x 2.16 ~= 57.9 potency per GCD. The math underestimates about 8 potency for PLD baseline and about 9 potency if you include FoF.

So if you take the damage underestimated for the WAR by the "hilarious inaccuracies" (~18.75 per gcd) vs. the differences between the math and the scenario you tested (~17potency) it still ends up at almost the same result.

So the math says the PLD should be doing ~9% more damage than the WAR does with the SE up, however, your tests show that the WAR is doing 15% more damage than the PLD. That is a MASSIVE swing between what the math shows and what you are seeing in your in game tests. An minor mistake and a small omission doesn't even come CLOSE to explaining that difference. You're going to need to find much, MUCH larger inaccuracies.

The other possibility is that the parser is wrong. I don't use 3rd party programs that can get me banned so someone else is going to have to test the accuracy of that.