Pet AI Rework Suggestion

[Grimoire-M]

I think you missed the part where I said I'd thought this up before BRD existed in its current iteration. Go me. I don't think anyone wants two extremely similar copies of jobs.. *looks at melee classes* .... oh.

Whether you did or not doesn't matter to me. I'd probably still feel the same way in Heavensward because I generally prefer offering solutions that require less development time based on SE's track record of putting the bare minimum effort into any changes they make, and Egi phases are substantially different from how the current Egi operate. That said, melee DPS are also not copies of one another. Combos are certainly oversaturated but the current melee DPS are not overly homogenized because of them. If you want to use that argument you'd want to point to the tanks not Melee DPS, but even in that respect they've changed quite a bit. You could argue RDM's a combo class with flexibility akin to Monk purely because of Dualcast and that MCH is a combo class that's gated by RNG too. I agree that combos have been played out at this point. Regardless, I feel you're missing the point of what I'm getting at. Borrowing aspects that are core to differentiating these two classes beyond their surface comparisons is exactly what I want to avoid.

SMN and BRD are already similar at a base level in the same way that combo classes are similar, they're DPS with priority rotations focused around maintaining damage over time abilities. However, many of their minor nuances are directly comparable too. Pets are akin to Auto Attacks, Ruin IV's proc is analogous to Refulgent Arrow. Iron Jaws saves GCDs for Heavyshot much like Tri-Disaster's reset mechanic saves GCDs for Ruin II/III. SB BRD's reliance on the crit buff stacking meta specifically exploits snapshotting, a mechanic that SMN also relied on for most of its DPS in HW. HW Bard had Feint, which at the time was an instant-cast Cross class weaponskill that allowed them to bypass their half-cast effect from Minuet at the time at the cost of DPS, similar to Ruin II on SMN now. SMN's aetherflow/aethertrail lockouts and Bahamut/Rouse lockouts also force them into a janky phase-like rotation too, only with none of the distinct filler Bard has in its song phases.

At a core level where Bard has distinguished itself from Summoner is a heavy focus on RNG procs, while Summoner leans into the consistency of their pets and Aetherflow as a flexible resource to manage. Summoner briefly gained additional resource management in Heavensward with Ruin III draining their MP for extra burst, something that was lost in the Stormblood transition while Bard got Song phases in order to fully flesh out and rework how they operate. I think we can agree Bowmage in Heavensward wasn't a good idea the same way Ruin III became a mistake at SB launch, but everyone at the time calling out the reason why it was bad: It made Bard feel like a caster. I feel Egi Phases and even the Aetherflow Lockouts are similarly bad for SMN as a whole, not only for their jank, but because it feels like Bard in a bad way.

The reason I'm more inclined toward phases than "permanent entities" is because I guess I want to feel like I have to cycle through the summons to use their power and then they are expended rather than just being basically set-and-forget. I brought up BRD specifically because that way, as you pointed out as well, you could prioritize different summons for what it is that's needed at the moment, single target vs aoe vs speed, or something of the like.

I maintain it's a bad idea to focus on phases because nothing requires you to juggle Egi phases to make the Egi themselves interesting. If you look at my proposals for SMN changes as of late, the goal of all the cooldown consolidation and lockout/DWT bonus removals is to free up time and damage potential in order for the Egi themselves to actually shine with any additions they get in Shadowbringers. We don't need to borrow from anything Bard does in order to do that. We can rely on what Arcanist brings to make SMN unique. In this case, I'd lean in on what made Ruin III work in HW without directly copying it, and what makes Aetherflow so engaging. Flexible resource management.

If the Egi were to stay as they currently are but changed to function similarly to Demi Bahamut, I'd be reasonably satisfied. They would just need to change it so you can hardcast and still get the attack rather than being forced to constantly use Ruin II.

I agree with the intent of what you're saying here but not your proposed solution. Fixing hardcasting to proc at the beginning of a cast (which is basically what you're getting at from what I can tell) doesn't fix the fact you'd want to be single weaving on them all the time and only makes the melee pets worse in the process. The problem with Bahamut's AI is he's literally just the Egi without their auto attack requirement. He's basically using the default Heel AI still, which I talked about probably being what the devs did in another post in this thread. That's why he's so inconsistent in practice. If you copy that over to the Egi they'll auto once every 3s on your cast and then run back and forth between you and your target repeatedly. The Egi are more consistent because of their auto spells in their current state. I'd prefer that consistency over his current functionality being given a choice between the two in particular.

That doesn't mean I think the Egi don't need a rework or think Bahamut has nothing to offer, but for all intents and purposes I'd consider Bahamut worse. In terms of the dev pipeline required to fix both of them I'd rather SE improve his AI and stick with the current pets till he's properly functioning, then pull his AI down to them. Getting him right first will make fixing the Egi easier in the long run. To me that's the more efficient way to go about getting the Summoner we're after. In the meantime I can see reworking the current pet abilities to have different effects, and then try to balance them out against one another.

The pipedream of egis and Demi-egis that respond to casts and basically function as turrets until the enemy is out of their range is still there, of course, just tempered with my own subjective analysis of their current nonsense.