This is great, hopefully it is not an oversight (alot of talents and stuff only affected base, some still do I think. Might be a similar issue here).

This would explain why blizz claimed the changes to be a buff and not a nerf though.

put some PTR updates of my calc sheets up...

also:
Thats alot of spreadsheets o.O

I think it has to be considered as a possibility that this is not the exact intention of the change. It's a similar situation to the Healing Way bug. Where a % reduction is interacting badly with +dmg/heal.

If they just did the basic coding change and tested it to find a slight dps increase they probably just said "woot" thats fine and didn't investigate close enough to see the real mechanics.

Once they seen you can get an infinite chain of LO procs all of which recieve normal +dmg they will do something. 20% procs that can proc of themselves are very dangerous.

Given a low damage LO proc of 1000 damage
1 LO proc 1000 damage at 20% chance or 1 in 5
2 LO procs 2000 damage at 4% chance or 1 in 25
3 LO procs 3000 damage at 0.8% chance or 1 in 125
4 LO procs 4000 damage at 0.016% chance or 1 in 625

That is not balanced. If the stars allign you can kill anyone not nature immune with one spell. There were plenty of 4 sword spec chain procs that got screenshoted and cried about, LO is going to be so much worse. Having a really low chance for an instant kill doesn't mean its not going to happen and certainly is not good from a design perspective.

If they're going to remove the chance for LO to proc itself, they'll have to increase the proc chance. Keeps the dps buff, and removes the one-shot chance.

The fact that LO is half base + full +dmg may be unintentional, but it most likely won't be changed (I hope)

Having the procs scale down in damage by half for each sucessive one would easily remove the one-shot chance and have less impact on the dps.

If your lucky it will be like when hunters were fixed to have ranged dodged and parried(maybe?) but the change was fix was reversed because it nerfed hunters too much.

Just thinking out loud here, with 2.3 are we looking at an increase in mana regen of upwards of 145 Mp5 ? 20 from the mana spring change (8mana/2s * 2.5), 50 from water shield (600mana/60s / 12) and another 75 from the mana cost reduction of lightning bolt (30mana/2s * 2.5).

Water shield isn't really worth it. You will have plenty of extra mana, and losing 2.5% of your damage isn't really worth more. You'ld need to cast 3 CLs to make up for that... you're basically coming out even if you time the CLs with a clearcasting and coming out under if you don't. That's not rigid math though, that's just my intuition.

As a warlock, I can tell you that most fights have you running around enough that you'll be glad for use for your GCDs. You won't be at 100% WS uptime, for sure, but even at 30 seconds average (pulling a number out of my ass) between going down and getting a chance to refresh it's around 30mp/5 without affecting your dps in any way. It's like darkpact on a 60s cooldown, less if you're getting hit (AoE DD spells proc it, right?).

Clearcast not proccing off a LO crit. I talked to some other shaman on the server, they claimed, anecdotally, that they had seen it proccing, and I had thought I had as well, but everytime I seem to check the combat log, I'm not seeing Clearcast proccing on LO.


Also, I saw some debate in the other thread about the possibility of it proccing off itself.



LO proccing off itself, proccing off Chain Lightning.

Its interesting to note the "chain of procs, " the largest step of the 2nd CL was on the same mob as the smallest step of the first, then the reverse again. Way too early to say if that is design, or luck, but interesting just the same. Begs the question of whether or not LO procs off only the initial pulse of the CL, or if each pulse can proc it, which lead lead to the conclusion that CL has a higher than normal chance to proc LO.

Also, can all three jumps proc it individually....guess back to more AE pulling to test it.

Don't you mean in the worst case?

That is the sexist picture I've seen for awhile. Almost makes me want to switch to Elemental.

That can currently happen in the game already, it's just a pretty low chance. What's nice is that seeing this on the first spell of an aoe pull or something won't mean instant death for you any longer cause of the threat changes.

It's on spell cast, not on spell hit, so I'm guessing the answer to 1 is no. Which should also answer b, you won't be able to get 3 CLs from the first. You'll just get a second CL on the same target, or maybe three CLs on the first. Or if you hit the jackpot 4.

C has been answered with LB already, its based off the first. I don't see why the mechanics would work differently with CL. So first CL is full, any LO procs are half base dmg plus full spell damage effects.

2. CL only ever hits three mobs. Why would it hit more than 3? The thing is that CL is a proximity spell to who it hits. So I'm thinking that with CL on a pull with more than 3 mobs, LO CL will be tied to your primary target. When you CL your target the first time, it jumps to the next 2 closest. If LO procs, just like now, the CL LO will hit the next two closest target to your main target at the time your LO CL hits.

Not sure how most of your questions should have answers any different than how LO currently works in the game. If a CL could give you 3 procs at LO instead of 1, you'd have seen 3 chances of a 5% CL LO proc on hitting a group with CL. This obviously isn't happening. We've always gotten just 1 5% LO chance on the target we're casting on. The only thing I'm not sure on is LO CL procs currently. Usually due to the .5 seconds between the first CL hitting and the second the mobs haven't moved. So its pretty tough for it to hit a different 2nd and 3rd targets. Been looking through Kael to try to find it, but that didn't work. Perhaps looking through my MH trash logs will show me that. I'm doing MH tomorrow, so I'll try to keep track of it there.

Edit: I think the easiest thing to look at here is the LO description.

"Gives your LB and CL spells a % chance to cast a second, identical spell on the same target."

So I think the easiest way to think of the spell is to really ignore the jumping effects of the spell. Just think of CL as a damage spell that has some area of effect damage to it. But for the purposes of LO, it just matters what your target is. The jumps of the spells shouldn't count as extra casts, or extra chances to proc the spell. Only the primary target matters. So atm we have a 5% chance to get a LO proc, a 0.25% chance to get a second LO proc, and so on. Of course now the numbers for procs will be higher, but the damage on the spells lower. But it still matters the same thing. The target you're hitting. Which also likely means if your target dies prematurely, it ends your chain.

Right, I am looking for data to prove it one way or the other. Most of these answers are tied to assumptions. The primary assumption is that each Chain Lightning jump is not a separate cast. For the purposes of Clearcast procs, they are, the 2nd and 3rd jump proc Clearcast. For the purposes of TLC, they are separate as well. It is also complicating in that the LO isn't hitting the targets in the same order as the first CL, which lends credence to either there being some sort of AI behind it (intentionally averaging the damage to prevent all three "first" jumps from plowing a single person in an arena), or it proccing off something other than the first jump.

We can currently test it on the live realms with a 5% proc rate, sure, but the point is we've never done in-depth testing on it due to the low chance to proc anyhow.

The other problem I've had is finding pulls with unique named mobs I can reliably pull and Chain Lightning on to test separate targets. I may have to jump into AV and take the liability of getting ganked each time I try and cast into a crowd, but at the very least, it should be easy to get definitive answers as to the jump order on LO procs.

I think there's a 3-pull of different elf mob types at the beginning of the mechinar, within CL range, and within running range of the door. Zoning back and forth on the test realm is always a gamble, but the mobs themselves are nice and instanced. As a bonus, you can use the patrols to test 4+ targets.

I may try that. I really wanted a bit less "dangerous" to test, so I could use it every 6 seconds, with 5+ mobs. But that should give me data anyhow, and with the proc chance theoretically high, I should be able to get numbers confirming rather easily.

However...if it doesn't work, it would take a much more comprehensive study to disprove possibility.

Seems like there is a bug with LO at the moment as well. If LO proc shoots off at the same time as a regular LB cast the LB cast is affected by the LO and it does reduced damage.

This video shows what I'm talking about. Not my own test but you can see the average lightning bolt is 1400-1500 or so but when LO procs it changes to 1100-1200 and affects the casts.
Stage6 · ptrtestÂ*-Â*Video and DownloadÂ*·Â*Evovi

Basically its a slight dps increase for all ele shaman. Not to mention increased sustainability and reduced threat. its not an amazing buff but overall pretty good and I've been quite happy both from the theorycraft and PTR testing.

That's probably because they are still calling it LB. If they just call it something else, it would be so much easier.

Thanks for this thread. I think it'd be valuable to see an elemental shaman gem assessment section, similar to the enhancement shaman thread.

I'm not sure how much assessment needs to be put into gems. I guess you can give a DEP for each gem slot and add the bonus to it and compare that to straight dmg gem DEP. Whichever one is greater, that's the direction you go in.

For the most part, sockets should be filled with straight damage gems. The only real off shoots of that are like Stormsong Kilt. With the +5dmg and a few of the epic gems, it ends up being worth it. But for like the Netherstrike set, the chest sucks with 2B and 1Y socket.

deleted due to innability to do math this morning.

So it's currently only increasing the +spelldamage coefficient in some odd way as you currently do nearly the same DPS with and without the talent with zero +spelldamage (it's a little increase due to doubleprocs however). With the half base damage for the proccing and the procced Lightning Bolt you can currently calculate the average damage as:
Dmg(base) + Dmg(bonus) + 0.2 * Dmg(bonus) + 0.04 * Dmg(bonus) + 0.04 * 1/2 * Dmg(base) + 0.008 * Dmg(bonus) + 0.008 * 1/2 * Dmg(base) . . .
Simplified you can say:
Dmg(base) * 1.025 + Dmg(bonus) * 1.25

While Dmg(bonus) is your +spelldamage * 2.5 / 3.5.

So your average +spelldamage coefficient increases from 2.5/3.5 to 3.125/3.5 through the talent which is nearly the same it was before the patch (3.15/3.5) while the average base damage factor decreases from 1.05 to 1.025. So the patch is a little nerf to coefficient and base damage of Lightning Bolt currently, however no one knows if the current weird version of the talent is intended or just a bug.

I found two errors:
1. You assume 64 LO procs in 2.3, how did you get this number? Your number would equal a 21.3% proc chance even though with multi procs it's more like 25%.
2. You rounded the spell coefficients resulting in a lower coefficient for 2.3 and a higher coefficient for 2.2 than it really is.