Yoshi P talks about Warriors- from Producer Letter Live @ TGS

[C-croft]

Dumb question. Has anyone tried stacking DEX to increase parry rates?

[Rensumas30]

lmao thats funny @ the let me tank maybe, im lvling both, mara and glad , maybe stacking str and dex ups damage for the heals and dex for parry instead of stacking vit

[C-croft]

lmao thats funny @ the let me tank maybe, im lvling both, mara and glad , maybe stacking str and dex ups damage for the heals and dex for parry instead of stacking vit

I only ask because I am wondering if we are expected to get some ridiculous amount of Parry to try and get our mitigation from the 20%-21% DMG Parry. Wasn't sure if the more dex you have the more useful stacking Parry would be. Kind of like reducing the diminishing returns of stack parry. If it even diminishes.

[HiirNoivl]

Hate to break it to ya, but dex does pretty close to 0 for parry.

[dandelions]

Another thing that really hasn't been talked about much... resource use and resource pools (TP mainly in this case) .I'm assuming this is similar on Paladin, but just using a basic rotation on Warrior will eventually deplete your entire pool; Overpower's added at any point expedite your journey to low TP.

Low TP affects damage output, enmity gain, aoe threat capability, etc... on both classes BUT low TP hurts Warriors far more than Paladins. Passive mitigation always happens, whereas active/reactionary dmg based healing happens far less frequently when TP is exhausted. Warriors ability to build Wrath stacks and therefore Inner Beasts, rely on consistently chaining TP attacks to build the resource.

Sidenote: This is also why Hiir is again full of shit: Increasing Skill Speed leads to low TP scenarios more speedily.

Anyone done comparisons of TP use between the Warrior/Paladin? Between my buddy and myself, I have a feeling his Paladin is more efficient in the long run on TP use given they mix MP abilities in more frequently. It's also the reason I rarely use Overpower besides short bursty large add pulls knowing I will be out of combat shortly as you will not see that TP again anytime soon otherwise.

[Rensumas30]

invigorate gives 500 tp would be useful for both types of tanks

[MillieVanillie]

So I've been following this thread for a while (read all 33 pages, INCLUDING Kitru's mathematical equations and formulas. Which I have to give him/her massive credit for crunchin them numbers like like it was just another day at the office) and I can fully say that WARs do need something. I just currently put mine off to the side to level a PUG because I can tell even by playin around with it for 2 days that the class seems to be missing a vital part to it. Either it needs a buff in defense or the HP regen skills need to be juiced up a bunch. So yea, no way is it "Just fine" as is.

[Kitru]

Another thing that really hasn't been talked about much... resource use and resource pools (TP mainly in this case) .

Which class has the easier resource construct depends upon how you play, but resource pools are large enough that TP costs only matter in extreme circumstances (the only common one of which is brought up later). The rotation that they use (using the one from my previous post) costs 650 ((70+60+60) * 3 + 80) over 10 GCDs, or 65 TP/GCD. The WAR rotation is slightly more expensive at 860 ((70 + 60 + 60) * 2 + (70+60+70) * 2 + 80)) over 13 GCDs, or 66.15 TP/GCD. Base resource gain is 60 TP every 3 seconds or 50 TP/GCD. As such, the net loss of a PLD is 15 TP/GCD (~63.33 GCDs before you reach 50 TP and are forced to have an open GCD) whereas a WAR has 16.15 TP/GCD (58.82 GCDs until empty GCD). The difference isn't major, but it's present. Of course, if you factor in Berserk, where a PLD loses 2 GCDs every 36, cost goes down to 62.48 TP/GCD and the net loss becomes 12.48 (76.12 GCDs before out). Honestly, most fights have some degree of downtime so that you're not attacking constantly. It doesn't really take all that much to allow a tank to have infinite resources for all intents and purposes.

When you bring in AoE threat, the difference becomes rather starl. Overpower costs 130 TP/GCD, which is a net loss of *80* TP/GCD whereas Flash is effectively free but limited to 7 chain casts before you run out (every 22 seconds you get MP equal to an additional Flash; 2% per 3 seconds; 1474 max mp with Flash at a cost of 212). Flash is also a net gain of 50 TP every time you use it. If you string Flash into your rotation, you can have a PLD that, quite literally, never runs out of resources. Each use of Overpower, on the other hand, is, effectively, 6.4 GCDs of TP consumption right off the bat. Overpower is expensive as *hell*.

It's for this reason, the difference between Flash and Overpower's relative costs, that it's really hard to properly gauge the AoE effectiveness of either class. WAR gets better burst AoE threat, but it comes at a steep cost. PLD has better *consistent* AoE threat, which also has a larger area and is easier to use because it's a PbAoE rather than a targeted cone, that is, effectively, free. If there is a long period of time between add packs, WAR gets to shine. If there *isn't* or the packs is more spread out, PLD wins by a mile.

[Tronic]

Great Calculations Kitru. It might serve to label each factor independently so that people know where each comes from; it took me a little digging and reading back and forth to see what was what.


DPS Out of Tank Stance

WAR:
(213.85 * 1.0556 + 83.33 * 1.011) * 1.2 * /.9 * 1.02375 = 423.12 potency per GCD

That being said, you made a little boo boo in this equation. "1.011" should be "1.11". Down in the in-tank-stance analysis section you write the number correctly.

Also, I'd include WAR's stun for completeness' sake in this equation. It'll give you 50/(30/2.5) = 4.17 extra dps per gcd and it's off the gcd, just as silence and spirit are.

Given these two changes your optimal warrior dps calculation climbs from ~423 DPS to ~440, beating PLD by 3%.

Another point: since you're assuming "Auto-attacks provides a baseline 83.33 potency per GCD (2.5/3)", shouldn't you also assume that Sword Oath should be multiplied by 2.5/3? If so, PLDs optimal dps goes down to 417 from 423.


DPS in Tank Stance

You state "Internal Release actually drops its contribution to 3.57% increase". But it was 2.375% in your DPS analysis, so it actually went up? In either case you are including both numbers in your equation (2.375% and 3.57%). I think you forgot to replace 2.375% wit the DPS from Wrath V (4.76%). Making this change ups the DPS to 345 potency per GCD.

Also I think this section needs to include the effect of Unchained, given it's trying to determine optimal dps in tank stance.


Enmity Generation

There's a part I don't understand - you multiply the effect of Internal Release and of Wrath V by 1.5 (btw it seems this equation has the same problem with the 2.375% and 3.57% as the equation in the preceding section) because "Crits generate an extra 50% enmity", but that extra enmity is already factored into those two percentages, 2.375% and 3.57% because they already include the effect of the extra 50% damage they cause.

Or are you saying they create an additional 50% enmity bonus ON TOP OF the already existing 50% extra damage bonus? If that's not the case then the enmity potency per GCD of that combo would be 1192, almost half of what the pally's combo gives. Edit: just to make it clearer, are you saying that a crit under defiance gives SkillPotency*1.5 damage, but SkillPotency*1.5*1.5 enmity?


P.S. Just realized through another post you should include the DPS from vengeance as well in all 3 sections! Add it as a modifier to autoattack, perhaps.

[dandelions]

large post in reply

I had edited my post quite a bit, my point was just to convey TP constraints hurt Warriors far more than Paladins due to part of the Warrior mitigation coming from "offense" and constant use of TP to build Wrath for things like Inner Beast. I haven't been to Coil, so I don't know general timers on the fights, but I've reached low TP in fights outside Coil (and without using Overpower).

Having a Bard use the TP song seems like a waste when both melee DPS and bard have access to the Dragoon's Invogorate (which seems to be fine for sustaining constant TP use on my bard).


Was just bringing up another point (more minor given it is situation dependent on long fights) to add to the Warrior inferiority vs. Paladin list in hopes the devs planning changes in the future.

\

Also, I'd include WAR's stun for completeness' sake in this equation. It'll give you 50/(30/2.5) = 4.17 extra dps per gcd and it's off the gcd, just as silence and spirit are.

You never use the Stun unless it is your turn to interrupt something (due to Stun DR build up). Using this ability outside of set perimeters on most bosses will essentially wipe the raid or at the very minumum cause far more trouble than 50 potency would ever be worth.

TLDR: It's not a free off GCD 50 potency 30 second CD attack; using it like so would get you kicked from any raid group.