New unique gloves for lightning builds: new version!

EDIT: New Version!

Ozone:
(strength-based gloves)

-40 lightning damage taken
Your lightning damage is lucky
Lightning damage cannot apply shock anymore

___________________________________________


Old version:

Ozone:
(strength-based gloves)

+1% maximum lightning resistance
+25% lightning resistance
Your lightning damage is lucky
Lightning damage cannot apply shock anymore

Very interesting concept; however, I'm very wary of any item that bumps MaxRes. GGG has been very stingy with MaxRes for a long time now (tree/gear nerfs aplenty).

If you are looking to afford some sort of reflect resistance, I'd rather see it implemented directly than indirectly -- just say "Lightning reflect damage is reduced by 10%" or whatever number.

Or you could go with something really cool and unique, like "Lightning reflect damage is applied at twice the rate of life leech." Or even "Your lightning damage cannot be reflected. You do not gain life leech if lightning damage would have been reflected".

All in all, though, a very interesting and well-balanced idea. Especially the part about lucky damage (which reduces RNG) for lighting (which is RNG-heavy).

I'm pretty sure some lightning/unarmed based unique gloves are coming in the expansion. They are Dominus' gloves. Hinted from a easter egg gif that was posted from a news thread.

"

If you are looking to afford some sort of reflect resistance, I'd rather see it implemented directly than indirectly -- just say "Lightning reflect damage is reduced by 10%" or whatever number.

Yes, you understood this item very well immediately. While the main characteristic is to reduce lightning RNG (and cancelling out shock to compensate, as "lucky" increases your DPS a lot), it also comes with a minor boost against reflect.

I have no problem with a more direct approach as you suggest, but I don't see the big problem with that 1% max increase. When it comes to fire, GGG has to be very carefull because RF exists. But in the case of cold and lightning, max resists are not easily overdone by granting a 1% increase.

"

All in all, though, a very interesting and well-balanced idea. Especially the part about lucky damage (which reduces RNG) for lighting (which is RNG-heavy).

Well thank you! The idea is indeed very simple. It opens up non-shock lightning builds and transforms the very essence of lightning. Because lightning has such a wide damage range, "lucky" is especially usefull on it and makes up for losing the ability to shock.

I didn't do the exact maths however. It seems rather complicated. When I have more time and there are more people enthousiastic about this design, I will try to calculate the effective DPS increase of "lucky" for a lvl 20 Arc gem.

"

I'm pretty sure some lightning/unarmed based unique gloves are coming in the expansion. They are Dominus' gloves. Hinted from a easter egg gif that was posted from a news thread.

That is cool news! Can you link that news thread?

while RF is a big issue, there could be similar (in that they are tied to max res) mechanics for lightning in the future, and additionally, they need to avoid allowing 100% maxres. granted, they don't avoid it completely, but especially since 2 of the hardest bosses in the main game (and maps) are heavily lightning-based (piety+dom, of course), allowing high maxres, especially among builds not specifically geared toward it, can break things.

this is especially the case with gear (and particularly gear in a typically non-critical slot which doesn't currently have a maxres boost like gloves). It's not 100% accurate, but off the top of my head, you have saffels (5% with legacy), purity of (5%@l23+100% from aura+buff nodes=10%), corrupted armor, corrupted ammy, and 1 passive node for +18; this makes it 93 pre-flask. With a flask and the buff nodes, you're looking at +13 from a flask, for a total of 106. This would make it 107 (obviously meaning that you could drop one or more of the above).

keep in mind that even 1% becomes scary big when you're approaching 100% -- from 75%, a 1% increase reduces damage by 4%, from 80% it reduces it by 5%, from 90% it reduces it by 10%, and from 95% it reduces it by 20%.

Granted, even with all of that, just a 1% increase isn't going to be a huge deal. But if you add a 1% here, and a 1% there, they add up fast. So to avoid this, the best policy is just not to add anything with maxres.

Anyways, that's my opinion on the matter, though GGG seems to be more or less taking the same tack based on the last year or so.

-------------------------------------------------------------------------------------

regarding dps increase, it's quite simple. take, for example, a range of 1-2. you have a 50% chance of rolling a 1, and a 50% chance of rolling a 2. This yields an average DPS of 1.5. However, by rolling twice, you get the matrix [1,1|1,2|2,1|2,2] with equal probability. Thus, by selecting the highest roll in each possibility, you get a 75% chance of yielding a 2 (and a corresponding 25% chance of yielding a 1), for a DPS of 7/4 or 1.75. These %s hold true regardless of the range of numbers -- your average for a range of 1-1000 normally would be 500.5, and with the "lucky" method it would be 750.5. This skews a bit because I'm oversimplifying a bit (it's actually [mindmg/2]+[maxdmg*.5] vs [mindmg/2]+[maxdmg*.75]), so for a range of 100-1000 it would be 550 vs 800, but since lightning damage almost always has a proportionally insignificant min, it's for all intents and purposes ~50% higher damage.

arguably this is somewhat overpowerful, but given the fact that you'd see this 50% increase from shock anyway, and the item itself doesn't add any lightning damage, I don't think it's out of hand. maybe a 10% less lightning damage mod, or removing the reflect protection would be more balanced, but I wouldn't say it's absolutely neccesary. especially given that in party play, shock increases EVERYONE's damage, and even for solo play, it increases all sources of damage (if you're not 100% lightning damage, which few builds are these days). Not to mention that it'd be competing against the beast that is Thunderfist...

"

Soepkieken wrote:

"

That is cool news! Can you link that news thread?

http://www.pathofexile.com/forum/view-thread/1232639/page/1

http://webcdn.pathofexile.com/public/news/31-03-2015/GoatPunch.gif

"

Granted, even with all of that, just a 1% increase isn't going to be a huge deal. But if you add a 1% here, and a 1% there, they add up fast. So to avoid this, the best policy is just not to add anything with maxres.

Your entire explanation makes a lot of sense and GGG probably sees it the same way. But instead of tackling elemental reflect directly, I wanted a more elegant and "hidden" way to help with it to some extend while also offering side benefits.

So, what about this instead of the +1% max resist?

=> -40 lightning damage taken

That's not an immense amount (Arctic Armour reaches -161 physical and fire damage at level 20), but it can certainly help against reflect, especially if you have a very fast cast speed. Moreover, it has some use against lightning enemies, but then again, especially against the weak ones and barely against hard-hitting bosses.

Seems like a significant elemental reflect defense without being powerfull in any way, which is exactly what I intended.

Others are welcome to share their thoughts on this!

"

regarding dps increase, it's quite simple.

"

it's for all intents and purposes ~50% higher damage.

I really appreciate the effort you've done of calculating that, but I fear it is incorrect or too hard for me to understand. Please read my calculations:

1) If the damage is 1-2, the average damage is indeed 1.5
If you're lucky, there are indeed 4 options: (1-1),(1-2),(2-1),(2,2)
Which results in 1-2-2-2 because "lucky" means the highest roll is picked.
So indeed that results in a new average of 7/4 or 1.75

So you're going from 1.5 to 1.75 average DPS, which is a +16.7% increase
as 1.5 * 1.167 = 1.75

2) If the damage is 1-3 however, the average damage is 2
There are 9 combinations for "luckiness": (1-1),(1-2),(1-3),(2-1),(2-2),(2-3),(3-1),(3-2),(3-3)
Which results in 1-2-3-2-2-3-3-3-3 when "lucky" is applied, which averages 22/9 or 2.44

So the DPS increase from 2 to 2.44 is +22%
as 2 * 1.22 = 2.44

3) The same calculations for 1-4 damage bring us to the result of a 25% DPS increase due to being lucky.

4) I still did the matrix for 1-5 damage which results in a 26.7% DPS increase.

=> So the DPS increase certainly doesn't seem to be that simple. First it's not the same for all damage ranges, even if the minimum damage is fixed at 1. Secondly, I think we have to use limits to calculate the DPS increase for example for the case of 1-1000, but I have long left high school and forgotten how that works.

I would be happy if someone can calculate this for me! It may turn out that when approaching the big figures, it will be +50% in the end, which seems possible seeing the way how DPS increases per 1 higher max damage evolve. But I prefer exact figures over an estimation ofcourse... .

"

http://webcdn.pathofexile.com/public/news/31-03-2015/GoatPunch.gif

Well, that looks cool! But it's probably an alternative to facebreaker and in no way as widely applicable as my unique gloves... .

As I said:

This skews a bit because I'm oversimplifying a bit (it's actually [mindmg/2]+[maxdmg*.5] vs [mindmg/2]+[maxdmg*.75])

^use that formula to avoid the matrix ;p

the % that stayed the same was the 50%->75% in the right hand of the equation, not the result.

for very low values, as you mentioned, it will be lower; also, for high min dmges it will be lower. but I would expect that in the vast majority of cases, you're approaching 50%.

"

As I said:

This skews a bit because I'm oversimplifying a bit (it's actually [mindmg/2]+[maxdmg*.5] vs [mindmg/2]+[maxdmg*.75])

^use that formula to avoid the matrix ;p

I don't really fully get how you deducted to get that formula, but I trust it's correct.
The big conclusion is that "lucky" comes close to a 50% more damage when the minimum damage is 1 and drops to a lower percentage when the minimum damage goes up or the difference between min and max damage becomes smaller.

In reality, when it comes to classic lightning damage, we can expect "lucky" to give us over 40% more damage.

"

Ozone:
(strength-based gloves)

-40 lightning damage taken
Your lightning damage is lucky
Lightning damage cannot apply shock anymore

So I edited the initial design because I think this is a more creative and interesting way of partly boosting your elemental reflect problems.

If we take the example of the classic lightning skill Arc, a good build can reach 4 casts per seconds that each chain 6 times. In a crowd that results in 4*7 = 28 lightning hits.
If all of those are reflected, that means if you reduce the damage of each reflect hit by 40, you WON'T take 40*28 = 1120 lightning damage you would normally have taken that second from reflect.

So we can see it is really an efficient boost. Maybe even a bit too efficient?
(Mind though the gloves don't give you ES/life, attack/cast speed, resistances,...).

Against hard-hitting monsters who can kill you in three lightning hits it's absolutely useless on the other hand.

What are your opinions about this?

Is -40 lightning damage A) a good and interesting stat B) not too strong or too weak?

Yes, I think that the -40 is perfectly reasonable. It reduces reflect damage very similarly (to a lesser extent, even) to equivalent existing mechanics (such as the described AA), but isn't high enough to affect enemy damage significantly. Despite resemblance to AA, I think that it's different enough to be considered unique, as well.

For a real-world example, I have an elemental dervish which is fairly endgame (lvl 85, nearly 40k DPS even though I havent gotten my Tamings yet), which is fairly heavily skewed toward lightning damage. It averages 231-2147 per primary hit (averaged between the two weapons it dual-wields, which fluctuate by ~1-75) -- primary because it's static blow, and the tooltip only shows the initial hit, not the static AoE.

Taking the (corrected, see below) formula, we get 231+[2147-231]*.5 or 1189, vs 231+[2147-231]*.75 or 1437. This works out to a ~21% increase.

My (much weaker) silly Arc+6-curse+aura party support, albeit missing some gear/auras/etc scavenged off onto other toons for the moment does 55-1011 per cast (plus chaining but that doesn't affect these calculations). This gives 55+[1011-55]*.5 or 533 vs 772. This works out to a ~45% increase.

As we can see, the minimum damage does indeed affect the % increase significantly -- and furthermore, it allows us to posit that attacks (which stack many more small sources of lightning damage from gear) will generally benefit less than spells (which generally stack fewer, larger sources) since they wind up with higher minimums to their ranges.

Fascinating stuff.

As to the formula, I shouldn't math when I'm tired. The (really, I promise) correct formula is: mindmg+[maxdmg-mindmg]*.5 vs mindmg+[maxdmg-mindmg]*.75. The .5 and .75 are explained above in the original posts. I don't know where I pulled the [mindmg/2] from but what we actually want to do, is remove the mindmg from the range calculation entirely. it's much easier to calculate, for example, the average damage of a range 0-1916 than 231-2147, and we know that the minimum damage is always going to be static. so we take 231-2147, and subtract the 231 from both sides of the range, yielding 0-1916, which is averaged to (1916/2=) 958, then add that 'guaranteed' minimum damage of 231 back in, giving us the 1189. Same thing for the "lucky" version, but *.75 rather than *.5. Hopefully that makes a lot more sense (though I suppose it's not really all that important).

Just for shiggles, we can take the average damage that my good toon from the example above would put out per hit, and see what happens with the -40 damage mod.

1437*.18*.25=~65
1437*.18*.25-40=~25

And yes, flat reduction is applied after resistances, I checked. Keep in mind that this particular build will be about doubling its DPS with Taming and 21ing its auras/heralds. In which case, it goes from 130->90 instead of 65->25. So I think that the balance is just about right on.