Buff stacking

[Calatin]

Here's something that bugged me in Live.

When you had three different items/abilities/whatever that reduced your vulnerability, they added together directly. For instance, 40% + 40% + 10% protection = 90% protection. Soon there were combinations that gave numbers too high (e.g. 100%), so they instituted some artificial caps, e.g. "75% is the max". The problem with caps is that you optimize to just barely hit the cap; and furthermore anything that might help you once you've reached the cap is worthless. At times this even included bufs provided by players: "We don't need your help because everyone's already at the cap with their gear, sorry".

Instead, I always thought such items should add together by reducing the remaining vulnerability. For instance a single 50% item gives you 50% coverage; a second 50% buff reduces the remaining vulnerability by half: 75%. A third 50% item would halve the remainder again: 87.5%.

Mathematically it's the one minus the product of (1-buf) over all the bufs; for instance 40%+40%+10% = 1-[(1-.4)*(1-.4)*(1-.1)] = 1-(0.6*0.6*0.9) = .676 = 67.6% protection. Ordering doesn't matter -- multiplication is commutative.

With this scheme, it always helps to have more (though the returns are diminishing) and you eliminate the need for an arbitrary cap.

[tredway]

This seems such mathmatical nonsense hehe. Actually this is technically how it should be as you described imo, but I think EA couldn't do simple percentage calculations and stuck with basic math and added the numbers sliding the % sign over haha.

But yea Im for your scheme no doubt.

[kjeron]

Some Criticism, hopefully viewed as contructive:

The additive stacking is to allow you to overcome any debuffs mobs will have, if you just aim for exactly 50%, mobs can strip that down to nothing, but if you buff yourself beyond 50%, you can still have a decent amount, or even still be at 50%, after being debuffed. Something mobs will hopefully and eventually have as they did in live.

How would a debuff work using your model? Would it still subtract the given amount, or modify it by the same (1-debuff) - this would make debuffer's almost useless as mob/player resists got higher.

Also, under your model, a PW in a balanced group could reach 80% resist, and eveyone else 70%, and mobs already have a hard enough time killing us at 50%.

Groups wouldn't not want you because they already had maxxed resists, but because you didnt have any that would stack with what they already had, and the stacking rules would have to remain in your model because otherwise you would be getting >90% resists for whatever type of damage you planned on taking.

[kemperor1]

This is how it works in EVE and it seems to work great. It'd be a far cry from where we are now though and it would throw off the balance of how much damage mobs do vs how much we can tank. Idk if it's worth changing with the amount of work to code it then balance out damage but I'd like it.

[rhobix]

This would work for debuffs too if you calculate buffs using susceptibility.

If you have a buff of say 25%, it would mean your susceptibility would be 75% (you take 75% of the damage you would have taken without the buff). A debuff of 40% would mean your susceptibility would be 140%. This could stack by multiplying everything. If we take the two modifiers above and add another 10% buff (90% susceptibility), it would stack to 75%*140%*90%=94.5% susceptibility which is the same as a buff of 5.5%.

Note that buffs gets more "bang" for each percent because they work towards zero and debuffs gets less because they work away from zero. A debuff of 50% in the old system would be a 100% debuff in this system.

[lpalgarvio[IS]]

complicated but interesting.

if implemented, i think it will likely be in a postponed way, because it involves changing many things and the devs IMO seem already hands full.