A spell pushback while you're stopcasting has the potential to cancel the spell entirely.

If you can get .05 casting lag for a prolonged period of time, you're just remarkably skilled in that department. My hat is off to you.

Myself, I couldn't conceivably release a button 100 times in a row and not be more than 0.05 seconds early (canceling the spell), let alone manage that in a frequently hectic, sometimes laggy raid environment, especially considering the inherently variable nature of ping.

True, it doesn't matter if you are comparing it to equations that omit the extra variable of client to server lag. Its a matter of personal preference since I can visibly see when raiding my average latency is approx 200 ms. In the end I find the inclusion of casting lag in equations to produce much closer numbers to what can be done in a raid setting. This is tested through wow web stats parsing and analysis.

I will agree to disagree

I completely agree. The above is the very reason why /stopcast macros + spell pushback are a double edged sword for casters. You said it much better than I could.

This crit v damage comparison is kinda interesting.

I first started trying to get a comparison similar to the druid str/agi comparison, but realised that it wasn't going to be practical as I'm looking at the X/Y stats rather than things that affect the X/Y stats.

So in the end I came up with a graph that compares 1% extra crit at a particular crit value (X) with the amount of +dmg we'd need to get the same output overall. The calculation is as follows:

=(((LB+(LBCO*B$1))*(1+$A2+0.01))-((LB+(LBCO*B$1))*(1+$A2)))/((1+$A2)*LBCO)

LB & LBCO are the average LB damage and the LB dmg co-efficent. I build concussion, and overload into these to make my calculations easier. Fury is also accounted for by not dividing crit by 2 to get the extra damage from crits.

B1 is the damage value for the calculation
A2 is the crit value for the calculation

This should work out how much extra damage 1% crit gives, and then divide it by (1+crit) times the LB co-efficient to see how much extra damage that 1% crit is worth.

Well, the DPS part is one we should consider, of course, however, given the difference in item value between +dmg and crit, I naturally assumed crit to be inferior. My calculations have shown that with new LO procs (20% vs. 5%), we still come about 50 dps short (now, LC adds about 72 dps at my 1100 +dmg and 45% crit, after 2.3, it'll add roughly 10 dps more). LC scales better in 2.3, but the difference is too small to cover the gap (If LC in 2.2 is 72 dps for me and it'll be 82 dps in 2.3, at 55%, the difference is 75 dps vs. 86 dps., i.e. not nearly enough to cover the gap).

So, crit scaling doesn't work. Haste scaling, likewise, fails because of our already short cast times. Shorter casts on LB make GCD-capped CL more and more trivial up to the point where we theoretically cast 1,5s LBs and only gain mana efficiency. The only possible answer to bring the dps back to current level is +dmg (because of lower item value). But this means the mana scaling in 2.3 must be good enough to cover for the loss of crit.

My opinion? I don't think there's any chance to get current dps, not even with +dmg. 50 dps would require about 84 +dmg extra in 2.3, assuming we were retaining the same crit. By replacing crit, this number only grows.

Beyond this, I simply don't know where else to turn. Most of our gear is focused on mana efficiency or crit, some has haste, but nothing else. We need Lightning Capacitor mk. II :/

As for the above spell pushback discussion - I think we can safely ignore it for now. The spell cast time stays the same for us, latency will likely stay the same, so we can naturally assume that the percentual effect on damage output is roughly the same. This applies to +hit as well - the percentual reduction is the same, so it actually goes slightly in favour of 2.3, but the difference is marginal at best.

Imo it is utterly wrong that you are basing your dps calculations on gear. "LC makes up for part of that dps loss" shouldn't be an argument at all. Seeing as they only buff our mana efficiency, and our usefulness in raids remains largely the same (MST gains 25 mp5 which affects the whole group) the DPS nerf isn't really justified.

In a perfect raid environment, usefulness + dps = constant for every class, thus if you give the raid little or no buff, your dps is higher than it is for a class that buffs the raid a lot. This kind of logic should put rogues/warrs on top of dps, because they don't bring much else besides dps to the raid. Next should be mages and warlocks, mages give int and can CC and do a fire debuff, while warlocks can do a shadow debuff and also CC. etc etc

The hard part is of course getting it all in balance for PvP and PvE at the same time. They nerfed our PvP burst, which was needed and im sure quite a lot of you agree here. They didn't nerf it to oblivion but it's still a substantial nerf. But the downside of this is that they also nerfed our PvE sustained dps which imo wasnt a necessary nerf. We don't need a buff to our gear because of this nerf, we need something to balance out this nerf that isn't gear based.

I thought we were trying to find a way to maximize our DPS after the patch, should we focus on crit, dmg or haste. I think we can all agree this is a nerf, but is it really anything we can do about it? The chances something will change on the PTR are pretty slim.

Also, if you are going to see how much of a nerf this is, you will need to add in average Shaman gear and compare it to other possible gear to get. The mail items may not be that many to choice from in higher instances, but there sure are quite many cloths pieces.

The LC come up as something that would improve after the patch with the new LO, giving it more procs/sec. (Would be wonderfull if it benefited from +spell dmg).


This is also something I have thought about for a while. Just gonna take some values out of nowhere. An average Lightning Bolt is about 2000, 4000 with LO. After patch it will be maybe 1800, 2700 with LO. Do you wanna do 2000 dmg 95% of the time and 4000 dmg 5% of the time, or will you go for 1800 dmg 80% of the time and 2700 20% of the time. I know many prefer not to have LO for PvP. But "burst" can mean different things, atleast to me. Having a smaller average damage, but the top damage being done more often can be a good thing. The 20% chance is something you can rely on in some sort of way, the 5% you can't. To me, you can't simply say that one thing is better than the other. Since a burst you can't control is a kinda useless burst, it will be just as random as before. Sure the overall dmg will be lower, but the "burst" will happend more often.

Shouldn't Warlocks be at the bottom of the DPS-meters together with the Shadow Priests, since Warlocks have +10% Fire and Frost damage to the whole raid. Shadow Priest have some insane group mana regen, Shadow Weaving AND Misery. To me it seems like this should be close, if not even better, than what buff our totems can give.

Of course it should work like this, but unfortunately, it's not even close to that.

*dreams of NS, EM, CL that procs a two LO which leads to LC proc*
aka 6700 dmg in 1 sec. Damn, I need to respec NS and get me an LC now.

Actually Eyonix's post does shed some light on this:

"If my lightning bolt currently does 1076 damage for 1076 threat giving me 3.26 damage per mana, my new lightning bolt should do 1017 damage for 915.3 threat giving me 3.39 damage per mana".

Obviously threat reduction talents are not considered here. So the average damage of a bolt without overload procs would be 915.3, and the average overload damage 1017-915.3=101.7.
101.7/915.3=0.1111111, so whoever thought up that example is assuming that lightning overload gives a 11.11111% damage increase. Now this is exactly what happens when each consecutive proc deals half the damage of a previous proc (yet unlimited chain procs are possible).

After playing as ele in arenas, playing versus ele shamans in arenas and watching quite a few movies, an ele shaman in pvp can do something like 1600+1550+~2850+1200 = 7200 burst (without crits or LO or resists) in a 3 second time window (first bolt + CL land at the same time, and 2.5 secs later NS EM bolt and 0.5 after that a shock.) and then continue doing about 1100-1200 sustained dps if bloodlust is up and he's not getting hammered. Plus you have purge to get rid of any hp buffs so effectively you can do about 65% of a targets hp in burst in 3 (3.5 sec of casting) seconds with your cooldowns before crits. After the nerf it should be around 6300-6700 damage but you have a much higher chance to proc LO and add to that burst (if you have the talent). Currently 14.2625% chance to proc one LO vs 48.8% chance to proc one LO after the patch.

PoM mages.... have you ever tried playing one in highend arena? PoM mages are about as squishy as you can get. They're like charge->execute splat. Sure you can do 4000 pompyro crit + 2200 fireblast crit + about 1500 as ignite but it's only marginally higher than a shaman burst and your survivability is pretty much none at all. Oh and i always love watching a mage doing lance lance, then popping grounding and watching pyro go in that. Works like 60-70% of the times :P

So, under that scenario, did our burst go down or up? 6500 damage plus a very strong chance one of the lightning bolts/chain lightning will proc LO for an additional 800-1400 damage. Sounds like our burst went up (or basically stayed the same with greater inconsistency) a bit when LO is speced.

So back to your earlier statement, why did our burst need to be nerfed? What basis are you using to determine "too much" or "just right" short term damage to balance a shaman?

I'm not going to get into mage vs. shaman balancing. Mages have tools shaman don't and visa versa. POM Mages burst like hell, and it wasn't nerfed, so the absolute ability to burst X damage in Y seconds wasn't diminished in the game. (Also, I've been crit with 6k in a single spell, not 4k as in your example, from mages)

I guess the issue is that some of us prefer that there not be one item that is critical to our success in our chosen spec. Especially one that has a random component to our chance of acquisition.

In other words, it makes sense to swap around a couple of items to optimize our DPS after the patch. It is not so good to have to have an LC in order to continue to do sufficient DPS.

Just thought I'ld mention that I tested LC with chain lightning, and determined that the shaman talents do not affect it. Its crit % is the same as listed on your character sheet.

Unfortunately, due to the nature of shaman dps in 2.3, on-crit/hit procs are much more valuable than they were before. The only one that doesn't have an internal cooldown is TLC. Quag's eye is also more valuable now, as the internal cooldown is lower than most other trinkets.

Post from Drysc that confirms RED like gem for casters.

WoW Forums -> Plans for an alternative metagem for casters?

I was overjoyed to finally have a TLC drop last night, having been long exalted in Kara I had basically given up on ever seeing it. So, joy!

Anyway, in my playing around with it, I noticed something disheartening -- TLC lightning bolts apparently will consume clearcasting charges. Is that something that's been established here, and is it worth complaining about? With the changes coming, that's going to be even more irritating in 2.3 if a free spell is wasting my precious clearcasts.

One thing we'll probably have to test when the 2.3 PTR is released.

Along with how many CC charges we can stack.

It seems to me that a fix that would solve numerous (potential) problems (threat tracking, clearcast consumption) would be to simply rename the spells to something other than "lightning bolt", like just call the spell itself Lightning Overload (Lightning Overload hits for 825!) and Capacitor Charge, or something! That way meters from WWS to Omen could actually track what's happening instead of lumping 3 discrete events incorrectly into one pile.

(Edit: not that I have any idea if this would actually help the backend programming. heh.)

The problem with that approach is that they should then modify LO to also proc LO, whereas now LO is procced off of lightning bolt or chain lightning.

I'm not sure I trust them to keep all that straight across multiple patches.

Well finding LC bolts is possible now. It's the Lightning Bolt hitting for 750ish that wasn't because of a partial resist. But it is very annoying. Finding LO will be about the same in 2.3. It'll be the bolt hitting for half and not because of a partial resist.

So you could sort our a combat log with some simple If statements. The problem for an in game mod would be knowing what your average LB should be hitting for. Which any mod can do, it's just an added layer.

I don't think they should really change the name of a LO proc. By definition it's just granting you a free version of your actual spell. LC procs however should be 'Charged Bolts' imo.

Imo, it's easier to track when the buff "Electrical Charge" fades from you. Keeping track of damage can be quite hard to do. The trinket bolt can crit and some mobs take extra spell dmg/weak to nature critical hits etc. There is also ways to track LO, not failsafe, but there are some ways.

It will also be compounded in 2.3 when LO and LC bolts hit for very close to the same numbers, depends on what your base LB dmg is.. I am not gifted in writing fancy programs to parse data for me, so I'm sort of stuck with what WWS outputs ... nor am I going to be re-writing mods based on my own anticipated damage to try and back out what's actually happening with threat. So, with the LC bolts, sure, I can look at total sums of electrical charge gained and knowing my own crit% work out how much of my damage is coming from it, but how will I be able to track LO when there's no corresponding buff tied to it, and the numbers are so similar to LC numbers? When you say "there is also ways to track LO, not failsafe, but there are some ways" what are you referring to?

these are the questions that keep me up at night