If I do incorporate this system, I will definitely use the Wavelength's suggestion.
Good to hear!

In general, you want to design your mechanics so that they are intuitive (from a game and interface standpoint, not just a realism standpoint), and you want to avoid creating situations that frustrate your player. I think this smoother "minus HP" system, always using 0 as the breakpoint for incapacitation, would accomplish both. I forgot to mention in my last post that I've played a game that actually does this - I think it was
Phantom Brave? Definitely one of the NIS games.
One thing that might contribute to it is this: whenever a party member gets knocked out in other games, you just use your revive on him, only losing one action from one person. And you almost always have the ability to revive. Your revival item stock is basically a second HP meter. But, then again, my system wouldn't help that much, because you could just spam your high-powered potions on the knocked out people.
I totally agree with you that this is a problem in many RPGs! It's dumb to make it this easy to recover from "death" (or KO or whatever) without a hefty penalty. The whole thing just becomes a slugfest/grind, with no real goal besides "hope that when it rains it doesn't pour". Easy revives also force designers to create overpowered foes that can two-shot you or even one-shot you in order to create any semblance of challenge.
WHAT IF I make it to where party members cannot be revived in battle, but can be revived outside of battle? (And not use the whole perma-death thing)
That's a good idea. I actually do something similar in most of my games - I either make it impossible to revive characters in combat (once they fall, you have to try to win without them), or I make revive items extremely rare/expensive (to the point where you'll only come into possession of a couple dozen throughout the entire game) so that they can only be used as trump cards - and then I automatically revive any KO'ed character to 1 HP
at the end of battle.
Making it easier to revive an ally, but introducing a hefty penalty for doing so, can also create fair and interesting battle dynamics. For example, in
Guild Wars when you are revived you get a 15% penalty to all of your stats each time you are revived, stacking up to 60%, which is only removed once you leave the dungeon - I think this is fine.
@WavelengthAnd if your worried about somebody irrationally only using the low-AP attacks, that's just bad strategizing. Anybody could just guard every single turn in any game, right? In that case, it's the player's fault and he can't really complain that the pacing is terrible. Of course I understand that it is always best to optimize the game to "guide" players into making the right decisions, and I'm not saying that it doesn't matter how slow the battles system is. I do tend to exaggerate.[/QUOTE]
I'm less worried that players will latch onto bad strategies (it's on them, not on you), and more worried that your good strategies might not be real fun. It's good that you have AP attacks at different levels, and even better that you give players a reason to use mid-level attacks by showing enemy HP bars (so they can take out an enemy if they have a strong enough attack). Consider, though, the player is on turn two of a normal battle, one where the most efficient move is to save up AP for their strong skills that target all enemies. What are they going to do this turn - and on turns 3, 4, and 5 until they have enough AP to launch their super-attack on Turn 6? Probably use their standard (very weak) basic attack, or guard the whole time, right?
This is actually a problem that I'm dealing with in one of the battle systems I'm designing right now (where it also pays to save up resources over time), and I haven't come up with a great solution yet. My current line of thinking is that the best way to counteract this boredom is to make the accumulation of resources more interesting and interactive. Perhaps defeating an enemy gives the character who scored the last hit an AP boost; perhaps things like being attacked while Guarding give characters an AP boost and the player has ways to influence who gets attacked. Now the player has more interesting ways to gain AP besides "wait until it accumulates", and they still get to experience their high moment on turn 6 but the lead-up in those first five turns is a lot more engaging. (This is also a good solution for your problem where Actor A can't accumulate AP because only Actor B is being attacked.)
I need to try to not make walls of text. :/ Sorry about that.
I don't get to criticize
anyone for this.
In fact, I remember one game did this, and it made me want to throw my 3DS out the window. I still think whomever designed that final fight in Shim Megami Strange Journey needs to get their head examined. NEVER put an hit 8 random target skill and then have the MC being KO'd be game over in the same game. All it takes is all 8 hits to hit the same target and game over. So make sure you avoid situations like this, else your game will be known for raising everyone's blood pressure instead of being known for fun.
With a very few exceptions, I always design "X random hits" skills so that they can only hit each target once. Any further hits on the same target are ignored. These skills seem literally impossible to balance otherwise.
I suppose
League of Legends does something similar where each hit on the same target beyond the first will do some small fraction of the original damage (for example, if the fraction is 20% then a target that takes one hit will take 100 damage, a target that takes four hits will take a total of 160 damage, and a target that gets hit all eight times will take 240 damage). That seems reasonable as well.
Also, remember that Atlus loves making you cry.