Common refrain, and I think something easily agreed on. Though there's spellcasting that ignores components (and that's even called psionics - check out the gith races for that!), so this doesn't necessarily mean "no spellcasting."
Kinda why I put it as zero; it seems the bog-standard, easily agreed upon change from magic. Still, it would need some counterbalance since a psionic would basically be getting silent/still spell for free.
That's a pretty meaty distinction! The biggest thing to watch out for here is the old 2e problem of "I spend three rounds doing nothing because I can't roll for crap." Could be something like "I blow some power points to MAKE SURE this works this time."
I was really debating on the idea of "works/works better" or "works/half-effect" style; it doesn't necessarily prevent you from doing something, but makes it less effective when you fail. I don't necessarily want a "roll to see if I do something this round", but in the end it could be no different than a fighter rolling to attack and whiffing, especially if failure costs little or no resources.
This seems kind of...academic and minor to me. Like, if we have a ranking of power (in that there is a cost) and a minimum required character level, then we sort of have "power levels" anyway. Why insist at their omission?
The idea is that power point cost, activation DC, and augmentation cost sets the "relative power level" rather than spell level. For example, telekinesis is a 5th level spell for wizards with set effects, a psion has a telekinetic power that he can get at 1st level and as he gains power, he gains better ability to activate it (proficiency bonus increases), more options to augment it, and more power points to use it. The minimum class level is only for effects that are too powerful for a low level PC to have period. (For an example of what I'm thinking, look at how warlock invocations are done.)
Sure, that's pretty distinct from spells, too - usually it doesn't cost anything to keep a spell going.
That might allow personal buffs (or even limited party buffs) to circumvent concentration; burning through X power points per member per round might be a good enough countermeasure, but if not than the cost will certainly limit hour long abilities. Again, that would depend on power points gained and costs to activate/maintain (with probably some cap of power points spent per round to avoid novaing).
"Casting at a higher level," but why limit the quantity? Why not just let it happen on things it makes sense to happen on?
The idea was to make power broad and versatile. One telekinesis power can do more than just life objects; it could do what Bigby's hand, Forcecage, mage hand, tenser's disc, and such all do if you spend enough power points (and reach the DC, which as I think about could scale with augmentation).
To take a spellcasting example; imagine if Cure Wounds was a psionic power. You can add power points to heal more hp, but you could also use power points to have it act as lesser restoration (removing disease, blindness, or the like), or healing word (healing at distance), or even revivify (bring someone back from the dead) all by spending points to augment it. Makes it very broad and versatile, so you would want a natural limit on how many different effects he could actually do.
Again, a cap on power points spent per round (and possibly augments making the DC to activate scale) would be the check on such flexibility.
This might get messy - proficiency bonus doesn't apply to your ability scores, meaning that someone who is well trained to avoid assaults on Intelligence actually doesn't show it here, which kills part of their fiction - that they are clever enough to resist Intelligence-based attacks better than most.
Yeah, probably best to go with tradtional saves, but its an alternative to such a system.
Totally. And psi seems like it'd be an OK fit.
From this perspective, psi would seem less reliable than spellcasting - a spell just works, doesn't cost anything to maintain, and while your foe might avoid it, they can't really stop you from producing it. Psi would be more difficult, producing a similar effect would be more taxing.
My idea was that psionics is harder to use, but gives you a lot more flexibility in how you use it. Spells (even those you cast at higher level) are rigid; they tend to do one (or a small group) of effects where psionics is much more fluid at the cost of reliability.