Well, I just thought of something, a potential test to see if a mod skill is overpowered. I wanted to hear what you guys thought about this test: "What would happen if I picked this skill for all seven slots?" If the result would be ludicrously overpowered, the idea is that your skill is overpowered. If the first-level buffs for your skill would add up to making you an unstoppable tank for the first few levels if you took the skill tree seven times, that's an issue. If having seven times the uses of your skill's cooldown spells lets you spam a skill combo that makes you literally unkillable, that's overpowered. Since a single skill is only 1/7 of your character's abilities, it may become difficult to tell if a skill is overpowered unless it's REALLY overpowered, and if the above argument holds water, this could help. Any thoughts?
Can't really see myself agreeing for the simple reason it doesn't account for interaction factors; this reminds me of once-factor-at-a-time experiments people would do in high school labs. A skill that has anti-synergy with itself would naturally ping up as negative, even if as a skill itself it could be overpowered or balanced. Same for skills that has synergy with itself or with other skills instead: if you took Blood Magic seven times over it'd probably ping on this test as negative, but it only becomes powerful with "kill lots of things" skills like promethean. Heck, that would probably ping negative on every single support skill ever. And of course, balance on things like buffs and such are going to be contingent on them having a limited number of stacks. If you could stack zenzenzaic 21 times, that pretty much makes things moot. If you took master of arms and could proc walk it off all the time for 1/hp regen, would that make it overpowered? No, in short, I don't think this is a good test. Part of balance in DoD is working out good builds, and this test just ignores that.
I don't mean that it would be the ultimate test, just one test. And also, this would be assuming that the procs, buffs, etc, could all activate separately from eachother, so that you could have 7 sets of berserker rage going at once, for example, or potentially proc blood magic 7 times.
Well, the thing is, that would make almost any skill overpowered. I don't see this test working out at all. Just stack seven Combat Rip and Tears, whatever it's called, That's overpowered. Stack seven This translation is all wrong, that's not overpowered, because, it doesn't give you any boosts, and multiplies x7 the chance of being lost in your own museum, which is like the worse nerf of a skill ever. So, yeah, there will be only a few skills that would not be overpowered if stacked 7 times. Most of the Smithing tree falls under this, except for the Trap Affinity.
The +14 damage from having 7 copies of Axe Skill would seem pretty sick on the first floor, as would the damage and stat boosts from having any one weapon skill maxed out 4 to 7 times. Yet most people would not consider any of the weapon skills to be over-powered. So, neat idea, but probably not actually a good yardstick of balance. Hmm... There may be some design space for modding in a second weapon skill for each weapon type. Something supplemental to encourage keeping with the weapon you chose skill in instead of ditching it for that EvilChest Lirpa with the insane damage boost. I'm actually kinda surprised nobody has tried anything of the sort... oh wait, that would make dual-wield that much more insane. Nevermind.
There's at least one skill mod that grants a chain that does absolutely nothing for the first four ranks, followed by a rank that gives an insane boost to a lot of stats (I don't think there's a single number lower than 10 in that rank) and two ranks with extremely strong cooldown abilities. Stacking seven of them is a good way to die on floor one, but nobody can say that it's not an extremely potent chain.
I can't get behind this. There are too many instances when having a skill even twice would be pretty dangerous, but having it once is totally OK. For example, taking the first level of Wild Magic nets you a 10% chance for a wild burst. Taking it seven times nets you a 52.3% chance of a wild burst per spellcast -- and those bursts are pretty damn strong, so even casting a spell that did nothing at all, you could potentially walk through the first floors of the dungeon just wild bursting and make it out with hundreds of boozes, bolts, eggs, and so forth. Particularly when it comes to percentage-based effects and long-cooldown effects (Spirit Bomb?), the sevenfold test starts to break down because it's totally OK to have powerful events happen rarely and/or uncontrollably, but not OK to have them take place commonly or consistently enough that they'll happen when you need them. For strictly stat-based levels, I think it's probably a decent idea -- but I don't really believe in levels that give you stats and nothing else unless those stats are cool ones that are decidedly out-of-flavor for your archetype AND they come in a decent chunk (see Bushido.)
