Negative resistance question

[Littlewoodenboy]

as of now my sorc has -30 light res for dmg. with convciton god who knows how much its decresed by. but is there a cap to what negative res you can get. i was planing on switch out eschutas for a crescent moon and replacing coh with up'd per vipermagi with a 5-5 facit. this puts me at -65 with over 117 fcr. still maxed res in hell. but i lose out on 8dr. for a litt more killing speed. would this better for me to do. an extra -35 is alot of dmg being delt. skills go down by 4, 3 from eschutas 1 from coh. is it worth it??? cant decide suggestions on people who use this before?

[sirpoopsalot]

1. The conviction from an Infinity will lower monster resistances by 85... or only 17 against broken immunities (there's a 1/5th effectiveness penalty that is applied when breaking an immunity).
2. Your -30 will always work 100%, except it cannot break immunities on its own.
3. The lowest that an enemies final resistance can be is -100%.


After understanding those points, the math becomes relatively simple. Let's say you do 10000 lightning damage (a low amount, but easy to work with). We'll assume your current setup with Infinity and an additional -30% Enemy Lightning Resistance (probably from Griffon's Eye & Rainbow Facets).

- Against a monster with a starting 0% resistance, your conviction (-85) and existing -30% will push the monster down to it's lowest possible resistance: -100%. So, when your 10k lightning attack hits, it will do double-damge: 20k

- Against a monster with a starting 75% resistance, your conviction and existing -30% will push the monster down to -40% resistance, so you'd do 40% more damage: 14k

- Against a monster with a starting 120% resistance, your conviction only lowers the monster to 103% (because of the penalty), and your -30% doesn't work against an immunity, so the monster will remain entirely immune.

- Against a monster with 110% resistance, your conviction will only work at 17% while breaking the immunity, leaving the monster at 93% resistant. However, luckily your -30% will always be 100% effective, so that pushes the monster to 63% resistant. Now your 10k attack is resisted down to 3700 damage (just think though - without that -30% you'd only be doing 700 damage, which is why it's always a good idea to have at least a little -Enemy Lightning Resistance).

- Let's look at that last example again, but instead of your current -30%, we'll say you have a Cresent Moon on top of that, for a total of -65% resistance... So the monster is at 93% after Infinity/Conviction, and your -65% works fully, leaving the monster 28% resistant. So you'd do 7200 damage.


... You can plug in different numbers and different resistances pretty easily, but what it boils down to is that against monsters with poor or medium resistances, Cresent Moon is not significantly better than some +skills from a wand/HotO (in fact, CM is worse against low-resists monsters). It's when fighting a broken immunity that cresent moon really starts to shine.

... it's up to you do decide if it's worth it. Why not try timing 5 or 6 runs with both setups and learn which is faster?

[AnimeCraze]

That's why I heard of people carrying 2 setups. 1 for common monsters and 1 for immunes (which includes a hand held infinity for -55 resist).

[Borje]

Poops has sumed it up very nice, I just want to give a small comment. One of the key arguments that I find speaking for the Crescent Moon is that no matter what you hold in your right hand your damage will be so high with that infinity against any non-LI monster, it's when you are facing broken immunities you will be slowed down. Thus, loosing some damage against the non immunes (maybe 10%, you will gain up to 100% (immunity broken by infinty to 93% -30% to 63%, another -35 from CreM is 35/37=95%). So, you loose some damage against the monsters that are allready falling ridiculusly fast but gain a lot of damage against the monsters where it will make a bigger difference. As allways when using CreM you have to asume that you hit the same FCR BP as if you where using the Temper/HOTO, in other case it's a some what other kind of math.

Ofc, this is the all theretical argumentation, the idea to time a dozen or so runs and to try out diff equipment to find what you like the most is a pretty good way to find what you find best!