Molten Shell - The build

Idea originating from reddit and Lapiz

☑ Memes
☑ Zerphi
☑ Pathfinder
☑ Fixed T-value skill
☑ Legacy rumi


My gear, passive tree, jewels, swaps

Skill tree

Main setup:



Anti-crit swaps:

Summoner swaps (75% (86%) chaos res):

The rest:




Upgrades:
Leg kaoms
Glove Enchant (winter probably)
Boots Enchant (10% pen)
Rings: 6% implicit, better rolls
Staff: 2-stat Spell+Cast Speed, multimod with BCR


Stats

Damage per hit (pvp: after reduction):
33594 tooltip damage with Faster Casting unbuffed (pvp: 4073)
40106 tooltip damage with Faster Casting, and flasks (pvp: 4514)
56148 tooltip damage with Faster Casting, flasks, and Ele Overload (pvp: 5486)
92134 tooltip damage with Faster Casting, flasks, Ele Overload, and RF (pvp: 7312)

54676 tooltip damage with Conc effect unbuffed (pvp: 5403)
65077 tooltip damage with Conc effect, and flasks (pvp: 5977)
91108 tooltip damage with Conc effect, flasks, and Ele Overload (pvp: 7265)
149524 tooltip damage with Conc effect, flasks, Ele Overload, and RF (pvp: 9683)

level 92
7.9k - 8.4k life with reg kaoms
6.2 - 6.4k life with brass
1681 mana
600 mana cost with Faster casting and 2 jewels, 800 with 3 jewels (702, 936 with Conc)
16k Armour (23k with brass)
48% Fire pen (10% on other ele)
3.55 casts per second (full heal each time) (2.54 with Conc)
0.28 cast time (you have to deal 9.5k-ish damage in this time to kill me) (0.39 with Conc)
34% Movement speed unbuffed, 142% with QS


Pros:
  • Tanky (full anti-crit, can facetank Amazon or Shin)
  • High damage (boombooms)
  • Not an offscreener (less cancer)
  • Melee-ish playstyle (FeelsGoodMan)
  • It's not EA mkay? I hope I did something good in my life at least once.
  • Good movement speed
  • Relatively cheap


Cons:
  • Not an offscreener (SRS hurts)
  • Hard to chase down kiters and offscreeners
  • YetAnotherZerphiAbuser

Videos:
1v1 vs Libri (EA): https://youtu.be/PEhEv88MUnk

Mechanics explanation, triggering the explosion, physical mitigation math:
You can read how Zerphi, Scold, and Mind over Matter work and interact with each other here.

In addition to those interactions this build focuses on utilising the physical damage taken from Scold to trigger the explosion of Molten Shell, at every single cast.



Molten Shell counts for all kinds of physical damage mitigation, calculating the difference between the initial hit and the actual physical damage taken, adding up multiple hits' mitigation, and when the threshold is reached it explodes. Recasting MS will not reset the threshold progress, so if you don't mitigate enough damage to reach the threshold in one cast, then it will counts toward the threshold of the next cast, and will eventually explode.
Different forms of physical mitigations that increase the threshold: Armour, Basalt Flask, Endurance Charges, certain passives, Chaos golem buff, Arctic Armour, Immortal Call, Block
The first 5 forms mentioned are additive with each other, and have a 90% cap in reduction (similar to resistance, but the rest of the mitigation forms can go beyond that, the last two can mitigate 100% if procced). Probably Fortify works too, didn't read or test anything about that.

The goal is to mitigate enough damage taken that MS explodes from one cast, triggering at each cast, that is the sweet spot.

There can be a lot of variables when calculating the physical mitigation, more specificly the armour required for it, but the most simple route is:
First choose your mitigation forms you can use and sustain. I use only Armour and AA because they can be sustained the easiest.

Take the threshold as a constant. It depends on the level of the gem, but once you know what level you use, it's a constant. Increasing the level too high is not optimal, because the threshold goes up too much, there are more efficient ways to increase damage instead of levels, at least with Empower. If you have a +1 chest or staff, you can use it.

Mine is level 21, so the threshold is 1664, which means that the physical damage taken from scolds must be mitigated by flat 1664 to trigger MS.

The initial physical hit depends on the mana cost of the skill. The lower you keep it the better, because you will hurt yourself less, which means higher effective HP.

Some theoretical examples:
If 1664 is the result of a 30% physical reduction, the initial hit is 5546.6, the mana cost is 1386, and you eat up 3882 damage, but only requires 23768 armour.
If 1664 is the result of a 50% physical reduction, the initial hit is 3328, the mana cost is 832, and you eat up 1664 damage, but requires 33280 armour.
If 1664 is the result of a 70% physical reduction, the initial hit is 2377, the mana cost is 594, and you eat up 713 damage, but requires 55463 armour.
If 1664 is the result of a 90% physical reduction, the initial hit is 1849, the mana cost is 463, and you eat up 185 damage, but requires 166410 armour. This is the lowest mana cost that can theoretically work, but it needs very high armour.

If you use other forms of mitigation too, then you will need less armour.

Just from these four examples you can easily see that around 70% phys reduction there is a sweet spot, where all numbers are kind of managable. Another optimum is coming from Zerphi, because if you heal more than your maximum life per cast, then the overflowing amount is lost. If you try to cast when below mana cost + damage taken amount, you will instantly kill yourself or not be able to cast at all, so the sum of that is removed from your EHP.

Exact armour needed calculation with my stats

This calculation is made with the old setup seen at the bottom of the thread. New version also has Basalt flask.

My main 6L setup with 2 Fevered Mind jewels has 702 mana cost, which means the initial damage taken is 2808, and to achieve the 1664 flat reduction, I must have at least 1664 / 2808 ~= 59.3% physical mitigation. The AA buff is multiplicative, and added only at the end, it is 13% less damage taken, so it means 1 / 0.87 = 14,94% more phys reduction, so the additive part of the physical damage reduction by Armour is (1664 * 1,1494 / 2808) - 0,1494 = 53.17% which means 10 * 0.5317 * 2808 / (1 - 0.5317) = 31881 Armour.

equations used for AA

(y - yx) * 0.13 = 1664 - yx

0.13y - 0.13yx + yx = 1664

0.87x = 1664 / y - 0.13

x = 1912,64 / y - 0,1494

x = 1664 * 1,1494 / y - 0,1494


(2808 - 2808x) * 0.13 = 1664 - 2808x

365,04 - 365,04x + 2808x = 1664

2442,96x + 365,04 = 1664

2442,96x = 1298,96

x = 0,53171562366964665815240527884206


Maximum effective Zerphi heal

Calculations based on my old setup, but you get the idea.

With my gear I was able to achieve 6.3-6.7k max life, and 1.3k open mana. When casting I take the 702 cost from life as cost, and also take 702 * 4 * 0,4683 = 1315 damage, 15% of that from mana, which means my suicide range is 702 + 1315 * 0.85 ~= 1820 Life

Which means that for healing purposes, a (6500 - 1820) / 12 = 390 mana cost would be enough, everything above that is lost.

With my mana cost, Zerphi heals for 702 * 12 = 8424 which is already above my maximum heal per cast


Old setup, no staff / kaom, full armour based

Skill tree

Main setup:



BCR swaps:

Spell block swaps:

Full tank swap (low dmg, high EHP and heal faster):

Summoner swaps (74% chaos res):

The rest:


Stats:
level 92
6.3 - 6.7k life depending on swaps (8.5k with reg kaoms)
702 mana cost
32k Armour
72139 tooltip damage with Conc effect and Ele Overload
45165 tooltip damage with Inc Aoe and Ele Overload
2.54 casts per second (full heal each time)
0.39 cast time (you have to deal 7.5k damage in this time to kill me)
IGN: Márkusz
My builds: thread/1600072
( •_•)>⌐■-■
(⌐■_■)
Last edited by Márkusz on Jan 15, 2017, 3:18:34 PM
Last bumped on Apr 18, 2017, 12:32:24 PM
Didn't even have to click the skill tree to know it was a pathfinder ^^
This is basically...Shin.

Few subtle differences as in he dual wields Ichimonji. It's a pseudo melee build that literally does nothing melee at all but self casts trigger set ups via Scolds. This is the concept basically everything is going lately be it Mjolnir,Cospri's,Scold's,Hiltless,CoC.
GGG, the ADA of gaming....huuuur i gotz mai skilz.
IGN: MullaXul
🎆🎆 www.youtube.com/c/Ahfack
🎆🎆 NEW #1 LLRF Helm -> 30% MORE|25CON|25BURN|-12fire|352es
🎆🎆 #1 Phys/Impale Leg 70% Multi Amulet + #1 Phys Rings!
🎆🎆 ^^Free 7L Fortify Cyclone!
🎆🎆 https://www.pathofexile.com/forum/view-thread/534134
why not juggernaut to counter bear traps?
"You have great power. You're right to be proud. It's unfortunate you have to die now, but I will honor you with 45% of my strength."
"
IceDeal wrote:
Didn't even have to click the skill tree to know it was a pathfinder ^^


Pathfinders in pvp??? Unheard of!
Lavender or Leave.
PvPresident, 2016
//
You'd better run.
“EA is fine” -relith
Other variant could be possible for pvp?
Without scold it s possible to use dual heartbound loop if go CI. The on death from skellies will trigger cwdt. If built correctly you can have infinite molten discharge with the minion and I guess if you go necro, you can spread poison when they die.

I guess it could play like hot: first cast skell, they die >boom boom> other cwdt also proc new skelies>boomboom>loop

Seen the skelly version 1 month ago on forum for pve.

Poe Pvp experience
https://youtu.be/Z6eg3aB_V1g?t=302
Last edited by Head_Less on Dec 16, 2016, 12:54:12 PM
Ice: yes, I even mentioned it before the tree :P

Mulla: yes, it's similar to Shin, but also different. My main setup doesn't need CWDT or any trigger support gem to work, so It can be as much similar to a self cast Discharger as to Shin. The concept you talk about is subjective, you just listed most of the things that deal spell damage but not self cast, Imo those are all somewhat different. I'm sure you laughed the same way when I mentioned ahfack's build in comparsion to yours, both are melee, they are the same concept!

Rup: they can easily convert to elemental, most of them are already ele based, but recently even sim gone more towards it. Btw I did meet sim in arena the other day, and I can reliably tank his single traps with brass, but if he swaps to cluster traps It oneshot me every time, that thing does so much bypass on the PvP formula. Also pathfinder is the best class for this, I did a demo on one of my scion before I made the real thing, and since I need to keep up two flasks all the time to operate the build, and not just to buff it, pathfinder is a really big deal.

Plexus: I did see, and even tried a heartbound build myself before, but I don't think it's really viable. It has limited trigger rate, 350 x 2 per 250 ms, so to reach the 1664 threshold they need to do 3 loops at least for molten, which takes twice as long as my cast time. Also you have to build into regen, and then sacrifice most of it just to sustain the damage taken, which the opponents can just remove with frost bomb, and the insane damages taken from opponents cannot be sustained at all. The skeleton loop requires a low level CWDT, and to actually take the damage, to trigger a lvl1 you need 528, so you can barely reduce the 700 taken from the rings to keep up the loow, and since it would be a low level skele gem you wouldn't have any significant base life, or could scale it at all if you want damage on your triggers.


IGN: Márkusz
My builds: thread/1600072
( •_•)>⌐■-■
(⌐■_■)
"
Rup: they can easily convert to elemental, most of them are already ele based, but recently even sim gone more towards it. Btw I did meet sim in arena the other day, and I can reliably tank his single traps with brass, but if he swaps to cluster traps It oneshot me every time, that thing does so much bypass on the PvP formula. Also pathfinder is the best class for this, I did a demo on one of my scion before I made the real thing, and since I need to keep up two flasks all the time to operate the build, and not just to buff it, pathfinder is a really big deal.


bear traps can't stop you as jugg, thats what im talking about, also jugg is good for you i suppose with %armor and movespeed, cannot be stunned etc
"You have great power. You're right to be proud. It's unfortunate you have to die now, but I will honor you with 45% of my strength."
Last edited by Rupenus on Dec 16, 2016, 1:35:22 PM
"
Rupenus wrote:
"
Rup: they can easily convert to elemental, most of them are already ele based, but recently even sim gone more towards it. Btw I did meet sim in arena the other day, and I can reliably tank his single traps with brass, but if he swaps to cluster traps It oneshot me every time, that thing does so much bypass on the PvP formula. Also pathfinder is the best class for this, I did a demo on one of my scion before I made the real thing, and since I need to keep up two flasks all the time to operate the build, and not just to buff it, pathfinder is a really big deal.


bear traps can't stop you as jugg, thats what im talking about, also jugg is good for you i suppose with %armor and movespeed, cannot be stunned etc


Yes, and endurance charge when hit is triggered by scolds, so it would be nice, but without the flask sustain, the build would struggle.
IGN: Márkusz
My builds: thread/1600072
( •_•)>⌐■-■
(⌐■_■)

Report Forum Post

Report Account:

Report Type

Additional Info