Yes, that's indeed true. In fact, the methods jump from one summon to the other so wildly that it's quite hard to weigh them all.
As far as I'm concerned after a little test, current categories are:
- Limited stack (doesn't benefit past certain point because of iframes): Slime, Vampire Frog, Imp, Optic, Deadly Sphere
- Unlimited stack (pretty much a multiplier based on slots): Finch, Hornet, Spider (?), Pirate, Blade, Sanguine, Pygmy, Raven, Sharknado, Xeno, Stardust Cell
- Special stack (one minion gets buffed; both current implementations give 33% of first minion damage per re-summon): Desert Tiger, Stardust Dragon
So, I was mistaken as well considering more minions at early game a solution. A lot of early-game summons simply won't redeem any viability with that, only Finch -> Hornet for preHM.
However, it's not "very few exceptions" either (although prior to 1.4 that would be definitely correct as most minions did inflict regular iframes).
Yeeup. That's the main gripe for players, seems like. And that's why Sanguine staff and Xeno staff feel so powerful for when you get them: they can always hit something you intend them to, no major derping around. May not be the greatest damage output on something like Golem (Target Dummy 2.0), but in realistic conditions they will perform almost as well while tricky summons like Sharknado may fail miserably.
Manual targeting to buff summon
IQ sounds pretty neat, actually. New summons clearly show that the limitations are for the most part artificial, so might as well give some means to improve their performance.
Not so sure about speeding them up with modifiers, though. As far as I'm concerned, the AI code cannot be really "sped-up" without manual/hardcoded instructions for each minion, aside from extra updates (maybe; plus that is too excessive buff as it is multiplying by a natural number). Definitely sounds cool on paper and could be the soil for some unique items.
imagine if crits actually doubled the attack animation speed for any chosen attack according to the chance