A QoL/Balance Shortlist for DPS (Tanks/Healers only as additional context)

[Cetonis]

I agree that it's fine for songs to be a bit imbalanced, but not... unfun, if that makes sense? I hear from many Bards that AP is not only numerically shit, but unfun, for instance, and I'd like to fix both issues there. I disagree though that Bard feels already like a polished product; a lot of it feels just as jagged as Machinist to me.

Sure, individual mileage varies, but it's a pretty broad consensus that Bard design is in a good place right now, and it's one of the (or the) most popular jobs.

AP is 'downtime'. That's maybe not 'fun' in and of itself, but from a game design perspective it's not valueless to give players a breather from watching for procs and such. Players do shy away from 'high effort' jobs to some extent, and while I won't claim SE is being super cognizant of that in relation to AP, it is a thing.

Personally, I do think going-fast is fun enough, but AP doesn't quite get fast enough to feel it. Wouldn't mind 6% per stack or six stacks max or the like, the problem is that if a BRD actually built SkS a 24% redux would run into the soft barrier they seem to have where they don't want it to be readily possible to get the GCD much under 1.75s (save for Rapid Fire because /shrug). It'd be nice if they just made that a hard cap so that they could do more speed things imo.

I can't fully tell what you mean here. Could you clarify? Piercing / crit buff? Is that one thing or two separate issues? What is the "crit buff" problem?

The crit buff problem is this - so long as Bard generates an outsized benefit from certain party comps, relative to the benefit other jobs and namely Machinist get from party comps, Bard will always and forever be the 'meta' choice over Machinist if the jobs are correctly balanced.

If you're SE, you really can't and shouldn't seek to balance your game around one exact party comp. And SE very clearly doesn't; they make concessions here and there when the meta situation is hurting players' impression of a job, but by and large it's pretty apparent that SE balances with random JP raid finder groups in mind.

That's why we get silly statements like "if we gave BRD/MCH piercing they'd need to be nerfed". Because the notion that a DRG may or may not be present is actually part of how they tune numbers.

With respect to BRD vs. MCH, what this means is that SE is going to (and rightly should) tune BRD under the presumption that there may or may not be Litany, there may or may not be Chain, may or may not be Spears.

But if BRD is even with MCH on that metric, it will always pull ahead in a static 'meta' context, provided that crit comps remain optimal for speedkills (seems likely, as you want to maximize your high roll potential).

While this is of course produced by BRD being the way it is, it's moreso produced by Litany and Chain being what they are. Buffs that, aside from messing with the value of the crit substat, are no more interesting than straight damage buffs - save for what they do for MNK/BRD. If SE made those more plainly 'boring' damage buffs, balancing BRD vs. MCH with respect to meta would be much more plausible.

An increase of 20% RA generation from HS, given that RA has exactly twice the damage of HS, would produce a 40% damage buff to HSs used over AP. AP can include up to 14 GCDs, with 1-2 spent on IJ, leaving you with 12-13 HSs. That would produce roughly 720 potency, less than two Pitch Perfect. A good WM will go 4-5 3-stack Perfect Pitch casts. That leaves us about 1000 potency short. That's better than what I've included so far for raw ST potency, but we still need more...

I have no interest in making AP and MB equal WM. There's no reason to do that, having a 'burst' song is fine, preferable even (why have three songs if they're just different paint jobs for the same outcomes).
Making AP close to, or a little better than, MB does strike me as intriguing though, because MB would still have merits for AoE and for better leveraging running crit-buffed DoTs (assuming SE doesn't actually change much on the crit buff front, which seems like a safe guess).

I haven't the faintest clue where you get that 40% notion from - your average HS/RA gcd would go from 175p to 192.8p, so about a 10% boost. Ballad is worth ~26.67 pps at 38% crit; current AP is hard to calculate for reals due to the broader 80 vs 90 second context re: EA and BL, but just naively looking at autos and HS/RA would place it at about 27pps at max stacks. So a bit better than MB but mostly at early expac crit.

On the chance that this is in reference to my own suggestions, everything here has been a universal change. The job-by-job mentions are merely to clarify that change (ability damage being affected by Speed via damage, unless already affected by Speed via rate as per Empyreal Arrow).

Taking one job's section by itself, it looks like you're just arbitrarily adding SkS scaling to various skills. If there's a universal thing there it'd be better to just make that a separate point.

Though I think making SkS even more nonsensical isn't really the answer. The DoT/AA damage scaling is already hamfisted; I'd prefer some alternate solution.

The effort to make every substat desirable has largely just been an exercise in homogenization. DH, Det and Crit are simply math problems where you work out which one increases all of your damage by the most, plus a couple cases of job-specific factors with Crit (and WAR DH). SkS tries to be different but still ends up only being wanted when it increases near-all of a job's damage, at a rate akin to Det but self-scaling instead of diminishing.

I'd prefer that SE back off of this, and instead simply decide to be fine with having some stats being more or less wanted by different jobs. So long as each stat has its places to shine, none of them are truly awful, and there aren't any MCH situations where a substat threatens to break a job functionally.

Sure, this adds some dependence on available gear to job balance, but I think discussions of this vs. that piece are healthy on the whole. It's not bad to have players discussing and debating your game, or to have folk looking for community platforms to ask questions on etc. It leads to stories about 'did you see the numbers xyz got with their meme build'? None of this is bad.

And griefing over questionable substat selection has never been that big of a problem really. I don't think it warrants abandoning the positive elements that having more distinct substats can bring. Including also, they would be able to add third sets at base+10 or base+30 ilvls, without it being wholly disinteresting. And lack of reward options is certainly a real problem right now.

<crazy half-serious half-baked ideas follow>

In this vein, I think they could rework skill speed to affect auto-attack delay instead of damage (if it makes NIN want it, that's fine), and even also reduce cooldowns a la the old Spear. This would return it to something that matches its namesake, and could provide some interesting options for optimizing even on a fight by fight level (grab a SkS set if it makes cooldowns fit in that awkard phase timing etc).

I know they balance tankbusters and such around cooldown length, but realistically tankbusters haven't been scary in years, and if a PLD is stacking a substat to improve their defensive options, then great! SE has never been able to get folk to do that.

Then I'd turn to Direct Hit and delete it. Keep Direct Hits in the game, but make it a chance only accessible via skills like Battle Voice and Inner Release. Maybe swap Chain and Litany to DH buffs re: the crit problem referenced earlier. Then you'd replace the DH stat with proooobably one that increases non-GCD damage, kinda the reverse image of speed. That would obviously vary wildly in how desirable it is, but this is okay in my book.

So you'd have the flat all-damage stat, the rng all-damage stat, the weird stat, and the stat that covers the weird stat's holes (while also amplifying certain parts of it). So long as there's no more than one undesirable stat for a given job, nothing too awful can result really, but you'd make your substats meaningfully different and interesting as a conversation piece.