Results 1 to 10 of 187

Hybrid View

  1. #1
    Player
    Lyth's Avatar
    Join Date
    Jul 2015
    Location
    Meracydia
    Posts
    3,883
    Character
    Lythia Norvaine
    World
    Gilgamesh
    Main Class
    Viper Lv 100
    There's no 'burden of proof' here.

    If you want players to be able to make up their own minds about the data set, you need transparency. It's very easy to present information in a way that is manipulative, simply because not everyone has a background in mathematics. If I can see dps contributions under Arcane Circle by job organized by percentile, I can see which jobs synergize best with me. I can also see if there's a significant discrepancy in those values that needs to be addressed. The same is true for all buffs, be it Mug, Embolden, Litany, Battle Voice, or whatever. You don't even need to make the distinction for 'single target' vs. 'raid-wide' buffs, it's useful information around which we can have an honest discussion. You don't have the same transparency with a back-of-the-envelope aDPS estimate because you don't know what smorgasbord of buffs were on offer even as a population average, and we can't tell if we're actually looking at contributions just from two minute buffs or random party wide BRD songs that are also thrown into the mix at random. The data is already all there, it's just not tabulated the way rDPS/aDPS/nDPS is. I don't think the issue is time investment. People just need to express an interest in it.

    rDPS isn't a proxy, by the way. It's just the original raw DPS numbers reallocated, with no damage discarded. And it is a relatively robust parameter based off how it's calculated.
    (1)

  2. #2
    Player
    Shurrikhan's Avatar
    Join Date
    Sep 2011
    Posts
    12,870
    Character
    Tani Shirai
    World
    Cactuar
    Main Class
    Monk Lv 100
    Quote Originally Posted by Lyth View Post
    rDPS isn't a proxy, by the way. It's just the original raw DPS numbers reallocated, with no damage discarded. And it is a relatively robust parameter based off how it's calculated.
    Yes, reallocated in a way that removes from view the team synergy that non-buffers bring to the table, just as raw DPS alone would leave out of view much of a buffer's value.

    rDPS is therefore a robust parameter only for jobs who would have more rDPS than raw DPS. It's a shit metric, though, for comparing the value of any two non-buffers, while raw DPS* (or, yet more usefully, aDPS) would instead be the obvious metric, because it still includes their contribution to team synergy.

    If I can see dps contributions under Arcane Circle by job organized by percentile, I can see which jobs synergize best with me.
    By all means, I'd love to see all that, but if you really are concerned with/by the complicating contexts of those measures, then your best bet is probably still just going to be simple relative potency maps.

    Those, in turn, won't be percentile-dependent unless different percentiles of a given job consistently progress through different rotations as rotation-affecting player error decreases. You simply look at what the job would optimally press when, in a similar fight, and calculate the relative potency. Import the optimal rotation for the fight, insert the modifier over the given actions and see what relative potency results from it. Easy. Consistent.

    (That said, not too shockingly, those early bits of on-paper theorycrafting also happened to show the very same imbalances we have already seen via aDPS, such as that 6.28 DRK would have about double the contributed synergy that 6.28 PLD would bring, because its relative potency is that much more concentrated.)
    (4)