I’m going to preface this post with the following general disclaimer: everything that I’m about to post may have already been discovered/mathed-out. It may even be completely incorrect; I’m a fallible human being. However, with that being said, I think an empirical solution for the value of the 2pcT13 set bonus is valuable to a large subset of the raiding elemental shaman population. In addition, I must first disclose that I do not currently raid as an elemental shaman, I’m about to respect to it after 2 years of being enhancement. I do not think I have missed any important elemental considerations, however.
One final note: I make very liberal assumptions throughout this logic based derivation. I don’t think, though, that any of them change the final result by very much.
This also happens to be my first attempt at TC on these forums: Please be gentle.
Start of very liberal assumptions:
Glyphs are Unleashed Lightning, Flame Shock, and Lightning Bolt. I believe the consensus has been formed that for most movement heavy fights (read: any fight but Patchwerk), these are the best dps glyphs. Again, this is one of the more liberal assumptions, and one that I really don’t think changes the value of the set bonus very much, as long as the flame shock glyph is taken.
Normal Rotation with the following caveat: Flame Shock is cast just before Elemental Mastery is activated so this poster doesn’t have to make any crazy assumptions about timing of DoT ticks.
World of Logs - Real Time Raid Analysis
Referenced parse is from WOL Baleroc #1 ranked Elemental Shaman, Frawshok. This means that the numbers quoted for damage done are probably as close as humanly possible to optimal in a real life raiding situation on a Patchwerk-like encounter. **Important note: I haven’t failed to consider that this post has about a 4 minute fight length; I’m just discounting it because the % of damage done by the differing abilities won’t change much with fight length. I’m only using the parse to get his average damage done by overloaded Lightning Bolts and Lava Bursts. Side Note : I wasn’t even aware that WOL could differentiate between overloaded LB’s and normal ones. Kind of nice to know.
Perhaps the most liberal assumption: when I run simcraft on almost any elemental shaman out there, it gives me haste and mastery EP values approximately equal, which has been discussed. If I run my toon through a reforger with equal EP values, I end up with haste and mastery values in the high teens I.E. 1600 – 1800 haste and mastery rating. These values seem reasonable to this poster. If someone wanted to plug and chug their own haste values in, it would be rather easy with the formulae I’m about to lay out down below. Therefore, this model is not attempting to account for every possible value of haste there is, the difference in the final damage calculation is minimal at best.
From a previous post (
https://docs.google.com/spreadsheet/...KPti9oE#gid=15)I found a linked google docs spreadsheet that has the haste values vs. the number of ticks of the Flame Shock DoT. I’m assuming that for the purposes of this discussion the Flame Shock DoT will tick between 13 & 17 times for most posters with the relevant buffs. I’m NOT including goblins or Dark Intent in this calculation.
After getting the number of ticks, I divided the DoT’s duration by the number of ticks, and arrived at the period between ticks at the upper and lower bounds of reasonable haste values found on gear. I then accounted for the fact that for the purposes of this post, only the ticks that happen between t=0 (activating Elemental Mastery) and t=~13.5 (accounting for latency, cast time (no 4pcT12), and human reaction times) actually matter. Therefore, with the above caveats, the flame shock tick will tick 8 or 10 times during the period of EM & and the set bonus being active. I then VERY ROUGHLY assume that (because I don’t want to do the calculations twice) the DoT will tick 9 times for my calculations.
I almost made the mistake of accounting for haste additively, but luckily caught my fail before posting. It’s multiplicative and that matters.
Finally, the only thing I’m attempting to calculate is the relative value of the set bonus, therefore I’m trying to ignore any damage that would already be present without it. This simplifies the calculation slightly.
With 9 ticks during EM, you have 9 x 0.2 (talent) = 1.8 average LVB CD refreshes during an EM CD. (modified slightly (x0.96=(1-(0.2x0.2)) chance to consecutively proc LVB cooldown refresh and be unable to use it w/o 4pcT12) This is negligible. And the math is easier w/o it.
Assumption: 8 minute fight length. (Longer than some, shorter than others.)
Assumption: in an 8 minute fight, you will most likely be able to cast EM 5 times (assuming ~1.5 minute cooldown after talents). It’s also most likely that one of these will be during a herolusteria warp period.
Therefore, you have 4 non-hero’d EM’s and 1 hero’d EM.
Hero’d EM is easy, you haste cap your GCD and you get 15 GCD’s in 15 seconds.
Non-hero’d EM isn’t. With no haste, you get 10 GCD’s in 15 seconds. Liberal assumption: the shaman in question has ~20% haste with all permanent raid buffs besides DI and non-goblin.
Therefore, haste = 1.2 (static haste from gear & buffs) * 1.2 (haste from EM) = 1.44 spell haste during EM. Therefore, you get 14 GCD’s under non-hero’d EM.
4 x 14 = 56. 56 (non hero) + 15 (hero) = 71 available GCD’s during EM over an 8 minute fight.
Assuming you cast FS just before EM, you should only need to cast Earth Shock once during every EM, in order to maximize the number of casts that can proc overload during the buff.
This means that every EM loses one GCD to ES. -5 GCD + 71 GCD = 66 GCD’s left over, 51 not under heroism.
At 1.8 LVB refreshes during EM, you should be able to cast 1.8 * 5 (EM’s) = 9 LVB’s (which are at the GCD at hero and not hero) =
9 GCD’s for LVB under EM during an 8 minute fight.
66 GCDs – 9 GCDs = 57 available GCD’s for LB, 42 not under heroism.
Approximate cast time LB not under hero = 2.0 secs (base talented)/ 1 + (haste) = 2/1.44 = 1.388888888
Time available to cast LB not under hero = 42 GCDs * time per GCD not under hero = 42 * (1.5/1.44) = 43.75 seconds.
Number of LB’s cast during EM not under heroism = time available to cast/ cast time = 43.75/1.388888 = 32 LB’s cast during EM not under heroism.
Approximate cast time LB under EM under heroism = 2.0 (base talented)/1 + (haste) = 2/(1+(1.2*1.2*1.3)) = 2/(1.872) = 1.06837607 seconds.
Time available to cast LB under heroism = 15 GCD’s (haste capped GCD’s) = 15 seconds
Number of LB’s cast during EM under heroism = time available to cast/ cast time = 15/1.06837607 = 14 LB’s cast under EM under heroism.
32 (non hero) + 14 (hero) =
46 LB’s cast during set bonus.
2000 (mastery rating from set bonus) / 179 (mastery rating conversion factor) = 11.1731844 mastery * 2 (% chance elemental overload conversion factor) = 22.3463687 % extra chance to proc from set bonus.
46 (LB’s) x 0.22343687 =
10.2713296 EXTRA Overload LB’s gained from set bonus.
9 (LVB’s) x 0.22343687 =
2.01117318 EXTRA Overload LVB’s gained from set bonus.
From the referenced parse: average Overload LB normal hit: 13.5k.
Average Overload LB critical: 27.7k.
Critical strike chance from parse: 21 crits/59 total = 35.6% crit.
Average Overload LB damage = (13.5k x (1-.356) + 27.7k x (0.356)) = 18555 damage.
Average Overload LVB damage (All Criticals) = 32160.8 damage.
LB Damage gained from set bonus = Avg Dmg x Extra Overloads = 18555 DMG/LB x 10.2713296
LB’s = 190733 damage. DPS gained = Damage/Time =~ 190733 damage/8 mins (480 secs) =
397.4 DPS.
LVB Damage gained from set bonus = Avg Dmg x Extra Overloads = 32160.8 damage/LVB x 2.01117318 LVB’s =~ 64681 Damage. DPS gained = Damage/Time = 64681 Damage/ 480 secs =~
134.8 DPS.
397.4 DPS + 134.8 DPS =~
532.2 DPS.
Frawshok DPS Gain % = (DPS original + DPS gain) / (DPS original) =( 43792.4 DPS + 532.2 DPS) / (43792.4 DPS) =~ 1.0122 =~ +
1.22% DPS GAIN.
Assuming the other math that I’ve read is correct and set bonuses are normally about 2-3 % dps gains, this one seems rather weak in comparison. One of the easiest fixes that I would propose to make this set bonus not so sucky would be to make the LB’s and LVB’s proc’ed from Elemental Overload actually benefit from all the stuff that the original spell does, I.E.
actually do 75% of the damage of the original spell. Also, maybe one day I’ll be even more ambitious and attempt to model the 4 set. Maybe.
Edit : One thing I wanted to point out is that I tried to make assumptions that would increase the relative value of the set bonus, that is, if the assumptions weren't what I thought they were, they would tend to lower the dps gained from the set bonus.
10/26/11, 8:42 PM
← Click Here
Similar Threads
