WAR Stat Allocation Str vs. Vit

[Kelberos]

WAR has higher eHP than PLD now for pretty much all situations. They are roughly equivalent for most content. The number look bigger going both ways, but they are effectively almost identical.

Now im curious -
What are the primary assumptions in eHP? Warrs and Pallies out of the box or with full CD rotations?

[Kitru]

What are the primary assumptions in eHP? Warrs and Pallies out of the box or with full CD rotations?

I believe that Gamemako's calculations were based upon either reliable (i.e. stuff you can have up every time) or time averaged (i.e all CDs except RNG ones like Bulwark or Featherfoot) eHP. Defiance and Shield Oath both provide the same eHP (25% +hp == 20% DR), so the difference is the other factors. If you look at reliable eHP, it's really baseline PLD v. WAR w/ IB, which gives WAR a full on 25% higher eHP (25% +hp *and* 20% DR compared to just 20% DR on PLD) so that WAR is just disgustingly amazing for burst survivability. If you look at time averaged eHP, WAR doesn't look *as* good compared to PLD, but it still edges out PLD, once again, thanks largely in part to the new IB (though the huge buffs to Vengeance and ToB help out a lot as well; IB is 25% +eHP with 30% uptime, Vengeance is 43% +eHP with 12.5% uptime, and ToB is 20% +eHP with 16.7% uptime).

WAR is, properly played, the king of surviving the regular burst hits that the devs are so fond of. PLD is still the best option for long periods of high sustained damage (Caduceus) or single instances (or multiple instances with long periods of time between) of intense burst damage (t4 OT w/ single dread strat; UW orb eating) because the WAR CD suite is more about high uptime than awesome effects.

[Gamemako]

What are the primary assumptions in eHP? Warrs and Pallies out of the box or with full CD rotations?

eHP is only valid on a burst basis. eHP answers the question of how much damage you know you can take in a short period without keeling over. It includes only perfectly-reliable effects, and it is not time-weighted. If you add in RNG effects, you've left yourself at the mercy of the RNG, and that defeats the entire purpose of the metric. Calculating a time-averaged eHP would be meaningless; you aren't facing a continuous decline in HP over a long fight. The relevant time-averaged metric is mitigation: the amount the healers don't have to heal, thereby conserving MP. Tanks need both to be effective as MT. Which job comes out ahead depends largely on fight mechanics, since WAR excels at burst mitigation and PLD excels at mean mitigation.