figure out the exact multiplier that strength should add to magic based physical techs from the sword-caster
1.337x, obviously.
or exactly what color the panties of our heroine will be when the clothes start coming off in those new and popular Bioware sex scenes.
Turquoise, to match her skin tone.
Well. Let's look at this in more detail. What do I think makes RPG combat fun?
1. Interactivity: The game requires me to actually be present during combat and contribute regularly. The player is necessary, not just for decision making, but for actively doing things. This is a game, after-all.
This is the part that makes or breaks a console game for me. For a handheld, it tends not to matter because I'm only ever half paying attention to those anyway. But, let's pretend we want our game design to be engaging. I personally would like to see some kind of combination of Grandia's turn system (IP supression and combos are fun!); and some kind of Timed Hit system (a la FF8/SMRPG, or we could go all-out SH-style Judgment Ring). I cite these two example specifically because these are really engaging systems, but they can be ignored in favor of good planning should a player not be skilled at timing-based games (it IS an RPG, after all).
Barring that, some kind of VPDS-style SRPG hybrid would be awesome. SRPG maps, but combat involves multiple units in range, ARPG-style (or we could use something other than ARPG-style, but I like the idea of combat still being team-based with any PCs that are in range). This may be a bit harder to balance.
2. Potential for Challenge: I like RPGs to be challenging. I'd like more RPGs to be challenging, honestly. You should not be able to breeze through most battles (or, at least, there should be a mode where you can't). You should have to use your brain. Mindless grinding should not be an out. Fights should take some planning and some strategy. And while the obnoxious difficulty of SMT games has some good ideas (nasty surprises, etc), a game should definitely NOT be chock full of those.
Pretty much anything we choose can be tweaked to be difficult just based on what kind of enemy AI we choose.
3. Depth: Kind of ties into two, really. You can put some thinking into the system.
One thing that I've found particularly engaging in Suikoden were the multiple modes of gameplay. Not just exploration and battles, but also a story-driven tactical RPG and duelling system. These are arguably mini-games, but there's no reason that multiple battle systems can't be present in a single game. Especially if we're just designing it, not making it.
4. Variety: A variety of PCs is always cool. And a variety of PCs with a variety if abilities (and styles to those abilities) are cool.
Lots of PCs. Lots of Temps. A world thrives on characters. This was one of the few things I really liked about FF4 - if you met a character, chances were that they'd join you, at least for a little while. I like the idea that important people in the story will join up, and then leave once their role isn't important anymore. It gives them a chance to show off for the player, but you can still have your core group of mains. I'd personally recommend about 8 permanent PCs, and 20 temps and optionals. Also, fun to draw and gives the multiple writers a lot more flexibility with adding their favorite personality types to the game. Big fan of making final bosses post-game optionals (but I'm also a big fan of N1 cameos, so feel free to ignore that).
Now, trying to make about 30 different PCs with completely different playstyles may be difficult. ...but somehow I doubt it with this group. We know more about how to make characters VERY different than most. In particular, I'm a fan of different characters having completely different growth/learning systems, a la SH2, SaGaF. Simple levelling is boring, but I'm also not too crazy about uniform Sphere Grid/Liscense Board style levelling. Mostly because even if each character gets different skills, they're still all learning them in basically the same way.
I find it more interesting if different characters have to do different things to improve. Some characters might learn skills through plot points, others might have to use old skills multiple times to learn new ones, still others might have to equip certain combinations of equipment, others might have to complete fetch quests or fight specific enemies, some characters might get larger stat bonuses from equipment than others but level much more slowly, some characters might start the game with really powerful abilities and long charge times but the get faster as time goes on, or perhaps a character whose entirely based on using a limit-break-like system. Maybe one group of characters can work on a job system and swap skills with eachother, but the rest of the cast has a different method of levelling. Maybe temps stop levelling after a certain point, but the player would still have access to them if s/he really loved that temp?
Mana Khemia: Fun, fluid, natural. Would have required thought if the game was tuned better. PCs were varied and the game encouraged you to use that variety.
Also, swapping PCs mid-battle is an awesome feature, even better when it comes with an interesting effect.
Final Fantasy 3/6: Standard ATB shenanigans, but the PCs have a lot of variety in what they do and how they do it.
I still prefer Grandia's system to FF6 ATB, but depending on how complicated it would be to implement, ATB isn't bad.
All told, I'm thinking ARPG style combat is definitely out. My ideal system for ARPG is some odd 2D Fighter/RPG hybrid, and that's a bit impractical.
ARPGs are fun to play, but they don't make for quite as easy to explain interesting IAQs. Also, probably far too hard to program/do art for should we ever decide to make something with this.
I think the best bets are definitely in the SRPG or "standard" RPG styles. Ideally, I'm thinking using the Mana Khemia as a rough base (the "card" system, consider the benefits of an assist system [although something different that would better support a larger PC]) and then making some changes additions (PCs with unique mechanics, etc) might be a good launching point.
Thinking on it, Grandia's IP supression and Mana Khemia's Time Card-delays are roughly equivalent, and MK's is easier to visualize. I'm all for this system, provided we still have G3-style combo options.