beta testers?

At this point, I’m still not entirely sure if I understand what you’re trying to say with your descriptions, but I’ll try to critique it anyway…

I’m assuming that magic, in your game’s world, would be rarely used? If it were to be used often enough to allow for “mages” in a traditional sense (as in, people who primarily use magic) to be anything other than rare or nonexistent, one would need a far simpler system, with far greater predictability. I get that you’re trying to create a system in which spells can be designed on-the-fly, and that as an idea is both great and rare, but I don’t think this is the right way to do it. You’ve got players having to make charts for their spell components in which they try to define the properties of their spell by picking and choosing traits from their components – that in itself is far too complex – and then choosing an undesired trait for each desired trait, thus increasing complexity further. The fact that an insufficient roll makes a spell partially succeed by only applying some of the desired effects puts far too much work on the shoulders of everyone involved, both on the part of a player who must plan ahead with each spell in order to minimize the risk while maximizing the reward, and on the GM who must, for each bad roll, consider which of the player’s decisions should be kept and which shouldn’t be kept, and also consider not only the immediate effects of what gets kept and what doesn’t but also the long-term effects and how each would influence not only the campaign’s plot but also how the players themselves would react.

If I were to play this game, I know I wouldn’t ever want to play a mage. Being the mage would mean that you’d probably take far too long above the table to cast the spells, thus making the other players resent you for hogging too much time which should, ideally, be evenly divided among the players; being the mage would also make the other players resent you because, well, when the fighter rolls bad, he simply misses when swinging his sword (I’m assuming), but when the mage rolls bad, the game could easily end with a localized (or maybe not) apocalypse. If the flavor of the world were to reflect this, practicing magic would certainly be done by only the most nihilistic of villains, and any mage who would for whatever reason not fall under that category would conceal his identity as a mage, for knowledge that a mage was nearby would cause everyone to immediately try to A. kill the mage, B. run to safety, or C. commit suicide so at least they could die on their own terms.

Of course, this is all assuming that I’m interpreting what I read correctly. Now, just so that there’s more of a degree of constructiveness to my criticism, I actually had my own idea for instant spell engineering which, unfortunately, would not be implemented in my own system for various reasons, but which might be useable in a different system (not necessarily yours, but you might be able to use or adapt it). It’s actually based on the flavor of how magic works in the setting of my game, which is why it’s quite shameful that I can’t think of a way to properly implement it mechanically. Anyway…

The system I’m proposing for instant spell engineering is that each overall “spell” is actually a combination of multiple smaller spells. Say you wanted to cast a fireball to throw at your enemies. You’d first cast a spell to flood a volume containing air particles with arcane energy (to act as fuel), then you’d cast another spell to increase the rate of collisions between those air particles in order to drastically increase the heat in that volume. The oxygen is already there, so you now have a fire, floating conveniently in the air. Assuming the fire was big or hot enough to satisfy what you wanted to throw at your enemies, you’d then cast yet another spell to move the effects of the previous two spells toward your enemies at a high velocity.

This system would allow you to mix-and-match spells to produce different effects based on the spells you use. Players would, of course, have to learn spells to apply certain effects. It would be a lot simpler, because the effects of each spell are already determined, rather than having to think of each and every effect that would be associated with a component. It would be less risky, as you would roll for each spell, and if you failed on any individual spell, you’d know exactly how your “overall” spell “partially succeeds” by simply not including the smaller spell that you failed to cast, and if you had enough time and energy to do so, you might even try again. Unpredictable effects would only happen on a critical failure. It would still be more complex than what most people are accustomed to, but it would also still be easier than what you were going with. I hate to make it sound like I’m insulting you, or that I’m bragging, but ultimately (and again, assuming I’m understanding what you’re saying correctly) I think a lot of people would prefer the system I proposed to the system you’ve got going as a system of instant magical engineering.