Many Savage PFs are barring PLD and WAR, anyone know what's going on?

[Lyth]

????

This is completely false.

The aDPS calculation discards all damage gained from single target buffs permanently. Nobody benefits from them. That's why it's an artificial metric.

rDPS redistributes the damage to the provider of the buff regardless of whether it is a single target buff or an AoE buff. That's why it ends up being equivalent to your party's actual damage, just reallocated.

[Argyle_Darkheart]

True, targeted buffs are discarded going from DPS to aDPS (it also doesn't account for Everburn, for what it's worth). Funnily enough, this actually means that tank balance is even less balanced (albeit slightly) than either rDPS or aDPS conveys, since DRK can be a valid target for those targeted buffs.

I suppose you could argue that this isn't relevant so long as DRK isn't the best target in the game at any given point, but that's not entirely fair, I don't think.

[SaberMaxwell]

rdps is good for comparing against other players on the same job, not for comparing jobs. It doesn't matter that a PLD will do more damage than DRK when hitting a dummy/ignoring other people, what matters is that in an actual group, a DRK will offer more total DPS by putting more potency inside raid buffs. You can look at adps and apply that dps to the DRK, or rdps and apply that DPS to the NIN, DNC, DRG etc., but either way, an identical group with a DRK instead of PLD will have more total dps and kill faster. adps is good for comparing jobs against each other, especially jobs that have no raid buffs of their own like tanks.

Nope, nDPS is the metric for intra job comparisons as its whole point is "how well did I do my rotation?" By cutting away both your buffs to others and their buffs to you.

rDPS is probably the best metric for inter job comparisons as it is a flat "this is what this job provided to the group" and probably the worst metric for intra because of how reliant on comp it is--a problem that fades as data points grow and get aggregated.

aDPS is pre-ShB parsing but with single target padding removed.

[Undeadfire]

Raiders and Balance will tell you aDPS is more important for tanks in dedicated Savage/Ultimate killing, and so will I, it is.

Put a Pld over a Gnb, or Drk in a fully BIS raid team and get a solid run.
Now for the original Drk and Gnb, same team, fully bis, solid run, Drk/Gnb would still be faster.

As previously stated rDPS is only a good measure for Pld vs Pld, Drk vs Drk, Gnb vs Gnb. In a raid setting for higher performance speeds, rDPS is irrelevant, it's performing the most optimal cooldown usage and 2 mins. Pld in the aDPS department falls lower than Warrior, and that's terrible for a HC raid team. Certain jobs have to take a hit in their rDPS measure for a faster kill, it's all dependant of kill time and rotation.

rDPS if you wanna log run against other players using the same job.
aDPS if you want faster kill and competitive other raid teams.

Focusing rDPS to try and competition other teams is never going to happen.

[SaberMaxwell]

It seems counterintuitive to use a metric that counts on other people using your buffs well to measure how well *you* did as opposed to the metric that only measures how well you did, but the balance has had some wild takes in the past so I guess I'm not surprised.

[Undeadfire]

For usual jobs, faster kills usually results to a higher performing parse. Pld has increased rDPS, flaw is it's losing a lot of raid buff gains and that effects kill speed a lot and difficulty defeating enrage for new kills. DRK/GNB/WAR are perfectly synced to 60s-2 min cooldowns and this benefits hugely for a whole team. Pld de-syncs on raid buffs far too quickly and loses too much DPS trying it, it's terrible for trying to push faster kills. Can release the next raid tier 6.4 Pld remaining same performance, 97%+ raiders would still pick GNB/DRK.

SHBs Pld did in fact sync with 60s-2 mins, anyone else telling otherwise is lying, or isn't a raider. Pld has 8 GCDs in Spell Speed spectrum, SHBs it had 5 so it could skip 1 Holy Spirit twice and remained with the 60s-2 min cooldowns, pushing competitive against GNB/DRK.

[Argyle_Darkheart]

It seems counterintuitive to use a metric that counts on other people using your buffs well to measure how well *you* did as opposed to the metric that only measures how well you did, but the balance has had some wild takes in the past so I guess I'm not surprised.

Well, for tanks rDPS = nDPS and aDPS = DPS (possible exceptions if you're a DRK). If there was a tank with a raid buff, then the discourse would be different.

[SaberMaxwell]

Well, for tanks rDPS = nDPS and aDPS = DPS (possible exceptions if you're a DRK). If there was a tank with a raid buff, then the discourse would be different.

That's fair I guess, but in the case of aDPS, the tank benefits in some fantastic ways from buffs like Mug, Embo, etc since they're not single target "padding" buffs. Which itself is counterintuitive to measuring a job's performance relative to others.

[Argyle_Darkheart]

Being better at funneling into buffs is relevant when comparing job performance, but not player performance (at least not necessarily).

[SaberMaxwell]

Actually thats fair, more I consider it the more that makes sense for tanks to use aDPS instead.