I think that the original post of this thread is drastically undervaluing engineering - I have it as equivalent or better than other professions at high gear levels. This math will all be rough, since I'm using the valuations in the front of the thread, instead of the sim, but:

The current logic is that trading 47 sp and 23 haste for an 'on use: 340 haste' is a loss, because 340 * 12/60 * .9 is less than 47 * 1.4 + 23 * .9; but the logic (and the sim I believe, but correct me if I'm wrong) assumes that 340 haste rating for 12 seconds per minute is roughly equivalent to 68 haste rating. With a modern cycle, this is *far* from the case.

Haste actually has two values - the value during solar and the value during lunar (after the soft cap). The solar value is drastically lower, because any time ng is up haste has zero value. So the net value of haste (ignoring cancelaura macros) is (NG_downtime * x + x) / 2 (x is the value of haste during the lunar eclipse.) The values in this thread put that at .9 (which has been reasonably close to my valuation in the sim most of the time).

Let's assume a 40% crit rate for a lower bound here (raid buffed). NG uptime with 40% crit rate is in the realm of 78% for wrath spam, so x is 1.48; with a 50% crit rate, that gets to 1.57.

Now, if you use hyperspeed only when lunar eclipse procs, you should be hitting it more like every 70 seconds, instead of every 60 on average, so the value of the accelerator at 40% crit is 340 * (12/70) * 1.48 = 86. The value of 47 sp and 23 haste comes out to 86.5 using the values from this thread, so that's obviously comparable; with 50% crit rate, the accelerators are worth 91.2. I think hyperspeed accelerators would need to be added to the simulator to get a *real* value (use only in the first half of lunar eclipse, on cooldown), but I feel comfortable enough on the basis of this napkin math to disagree with the OP's assertion that "Engineering and Skinning are weakest, both giving less benefit than 46 spellpower."

Notes: I've ignored the value of the rocket boots, since some people will doubtless prefer minor run speed enchants over 24 crit rating and 5 minute rockets. I think it's a worthwhile trade for most fights, but I can't easily quantify it. HSA also has high value on aoe, not that anybody cares much. Hitting HSA before spamming hurricane gets two hurricanes hasted, which is longer than most trash lives. And rocket boots are awesome.

On the other hand, I'm actually starting to run into the haste cap on *starfire* in the shorter fights now - there aren't enough separate lunars to use all my cooldowns unless the fight lasts at least 8 cycles (4 minutes), so if it's shorter than that, I have to use a haste pot while my [Fetish of Volatile Power] is up.

It's a good point, I think, it's worth looking into more. One of my next projects was going to be some better analysis of current trinkets, and so I'll probably be playing with the value of haste during Lunar in the spreadsheet. I'll have a better idea of how to quantify Hyperspeeds then, and it's probably more worthwhile to wait until I get that done than to post napkin math on the forums. A problem is that uptime drops signficantly if you sync with Lunar--your value of 70s is quite optimistic; in the sheet now, average cycle length is over 40s (although this is before the phenomenon of Lunar truncation with a macro/addon, which adds an interesting twist).

Hyperspeed for Hurricane is worth mentioning (and you can already get two Nature's Graced Hurricanes from one Starfall), but I don't think AoE winds up being important anywhere in Coliseum (will confirm after Anub'arak tonight).

e: that reminds me, there really is very little AoE in Coliseum. Maybe I'll drop Gale Winds when I respec soon for 2T9.

With the macro, my cycle has averaged ~37 seconds; I could probably do with a bigger sample size though; the variance in my data is higher than I'd like (>2). In any case, your 40s sounds reasonable for a macro-less rotation (and I'm sure you got that from strong data). Either way, my estimate was optimistic, I'll correct it here.

A 40 second rotation with 40-50% crit should get you 75.5-79.8 value points, which loses to the other professions slightly.
A 37 second rotation with 40-50% crit gets 81.6-86.3, which approaches even with jewelcrafting/bs/etc.

So without the rocketboots enchant, engineering doesn't really offer any bonus over the more normal professions, but is reasonably competitive.

I should have pointed out that the margin of error on my numbers is fairly large; it's truly napkin math. I worked backwards from sim-produced values that may or may not match the stats I assumed later. If I have time later I'll try to hack the accelerators into the sim.

That seems reasonable at first glance, although I still wouldn't recommend Engineering over other professions, as it will quickly drop in value anytime you can't maintain a tight rotation.

If you're very mathametically inclined at all, you shouldn't really be using the values listed in the OP, which are just zero-th order estimates based on my gear at the time I wrote it. Try familiarizing yourself with the (attached) spreadsheet. Also, for a more detailed post, look here: Moonkin TTT Article pre-publish draft .

Hey is the T9-2 bonus bugged cause I just got it and I also have the Glyph of Starfire. When i get deeper into my rotation and i refresh my Moonfire it doesn't add the 3sec like it should.

Personally I run with Typhoon unglyphed for the likes of bombots on Mimiron, adds on IC "Medium" Mode p3, Scattering exploding flowers on Freya - etc etc. I rarely use it as part of a DPS rotation (sometimes pop it if I'm moving or when in the Yogg brain illusion) but not enough to warrent removing the knockback - used properly it can be a life saver.

Entangling Roots is a big waste of a Major glyph in a PvE environment. The very first post of this thread explains what majors to run with:


Major Glyphs: the best three for single-target DPS are Insect Swarm, Moonfire, and Starfire.

Some people like to use Glyph of Starfall, as it stlll gives some single-target DPS contribution, and it's a very strong AoE spell to have available more often. If you want to use it, replace Moonfire (if you have 2T9, replace Insect Swarm instead).

Whats better to have 258 item level t9 4p or stay with 2t8+2t9, for me it looks its worth going for full t9 as 4p bonus is realy nice and you will go over 100% crit. for lunar eclipse anyway.

Ok, I'm adding a "haste during Lunar" stat to the spreadsheet (will upload as soon as I finish this post). First conclusion: Engineering is still weaker than the others (but not by much), even when I model Hyperspeeds generously as 170 Lunar-haste.

Now, trinkets:
I did a really quick analysis for now. I think the on-use haste effects ([Scale of Fates] and [Talisman of Volatile Power]) are rather weak. One of the issues is that they're both 20second effects, so their uptime is somewhat wasted in a ~13s Eclipse. The spreadsheet model does assume that the haste effect applies only during the Eclipse itself, but with truncation, you're going from a Lunar into casting Wrath nearly every time, so this is pretty accurate anyway. This effect is even worse for Talisman, since you waste the good part of the proc.

I put in a simple model of [Reign of the Unliving] and it seems to be ridiculously good. I'll try to refine it somewhat, but the two versions of this trinket are likely to be BiS by a substantial margin.

Just inputting their stats manually into the sheet with my current gear, they ranked roughly in this order:
[Reign of the Unliving]
[Flare of the Heavens]
[Illustration of the Dragon Soul]/[Eye of the Broodmother] (IDS is a touch better on paper with 100% uptime, but it stacks slower than Eye in practice. I'd probably choose Eye).
[Sundial of the Exiled]/[Abyssal Rune] (depending on how much haste you need)
[Scale of Fates]
[Talisman of Volatile Power]

Was this discovered on the heroic training dummy? I've had some issues with moonfire not being refreshed on the dummy in Orgrimmar, but never on a boss fight. If it was on a boss fight, I'm still not sure if you should relate it directly to the T9-2 bonus though.

Have you tried to reproduce it on something else?

I'm curious to know the interaction between the two ilvl trinkets while both are equipped, whether they share the same proc buff and proc ICD, proc seperately off the same crits etc etc It's a weird precedent that could have interesting and powerful results.

For those wondering if they can both be equiped, here's a shaman with 2 [Stonecloth Bindings] equipped(at the time of this post, gathered from one of the Enh shaman thread); The World of Warcraft Armory

For now I'm assuming they work perfectly well independently, since the buffs they use have different names.

On the first post of this thread there is a couple addons listed, there are other options but I think Squawk and Awe takes the cake.

oh ic never saw it sorry about this guys thanks! im loving this add-on helps me alot
edit: what are two good blue gems to use to activate the meta gem?

the purified one, sp/spirit

also get wiseeclipse (link in the moonkin ttt article). its just an install&forget mod, but it helps quite a bit

I have just recently started gearing up my moonkin. I'm the type of person who stockpiles all potential specs of gear as I level/instance/raid and then when I get time/money, go crazy enchanting/gemming and getting all specs I'm willing to play up to speed (always bare minimum hit/uncritt/hp/def/etc).

I know trash in general doesnt matter, but does anyone have a happy easy solution for just holding my own on trash? I picked up gale winds, which helped a bunch, but what if it's a smaller group of mobs? (I'd link my armory except I switch between bear, cat, and boomkin almost constantly) IS-W, MF spam? Just wrath?

My haste is also much below suggestion (I'm at ~2k sp, ~210 haste). Is it worth gemming for haste? Or will haste just come as I gear up?

When I look at rawr, it heavily pushes int (or int/hybrid) gems (I have rawr left on defaults, dont know how to get it to suggest minimum numbers). I ignored it and went spell power for now. Just wanting reassurance.

If you tinker with gemming templates in rawr you can customise how you want to gem and make any possible combo of gemming you wish to have. Also try and list all buffs you know you will have to get an accurate gearing model, I'd advise you to play with gear options that are realistic for you rather than go with what rawr tells you to be bis.

For trash generally the rule I follow is - minimum of 3 mobs then hurricane and then switch to single spell spamming -usually wrath - with at least moonfire up (to get the crit from idol) when there's 2 mobs left.

Fewer than three then I just dot both up and get e&m on both targets and just do a normal dps cycle but taking care to switch to the other mob to refresh debuffs. If It's a high dps group - ie the mob isn't going to stay alive long enough for decent dot ticking I just spam wrath, with a moonfire for idol proc, till lunar then starfire chain till the mob's dead.

As for gemming haste, I personally wouldn't do it unless I was close to the soft cap. My sockets would pretty much be spellpower unless I lack a few points of hit - that's the only non sp gem I'd gem in my gear, instead I'd favour haste items over crit items.

I did a little number crunching about whether it's better to hit a spell power trinket during Lunar Eclipse or Solar Eclipse. Using coefficients from the TTT draft my math looks like this:

Solar Eclipse
1.2078 = SP coefficient on Wrath during Eclipse (accounting for moonfury/eclipse being additive)
11 = Wraths affected (I could particularly use confirmation on an expected number of Wraths that benefit from Eclipse)
(paper doll crit or PDC)+.2 = effective crit rate (ECR for next formulas)
ECR*2.09+(1-ECR) = 1.09*ECR + 1 = multiplier on damage due to crits

Total SP coefficient for the duration of Solar Eclipse:
1.2078*11*(1.09*ECR+1) = 14.4817*ECR+13.2858 = 14.4817*PDC + 16.1821

Lunar Eclipse
1.646 = SP coefficient on Starfire
6 = Starfires affected (this is dependant on haste and Nature's Grace)
PDC+.23+.3 = ECR during Lunar Eclipse
1.09*ECR + 1 = multiplier on damage due to crits
.6 = Unaffected Starfires
PDC+.23 = ECR for unaffected Starfires

Total SP coefficient for the duration of Lunar Eclipse:
1.646*6*(1.09*ECR + 1)= 10.7648*ECR+9.876 = 10.7648*PDC + 15.5813
Partial Starfire correction = 1.646*0.6*(1.09*ECR+1) = 1.0765*ECR+0.9876 = 1.0765*PDC + 1.2352
Final = 11.8413*PDC + 16.8165

Break even point is found at:
11.8413*PDC + 16.8165 = 14.4817*PDC + 16.1821
0.6344 = 2.6404*PDC
PDC = .2403

If my above math is correct, then it is better to use SP trinkets during Solar Eclipse if your paper doll crit-rate is 24% or higher. If there are any flaws with my assumptions or my algebra, please let me know (PMs preferred to keep the thread clean).

Draugdae,

Even assuming most of your math is correct:

1) Delaying the use of a trinket for 15s means losing roughly 15% of its benefit (15/90 or 15/120).
2) Using timings from a recent Hamlet Wrathcalcs, in the time it takes to do 11 Wraths, you can cast 6.6 Eclipsed Starfires. That "wasted 0.6 SF-cast time" means the trinket SP gets some use on something else (a DoT, or the first nuke after Lunar Eclipse).
3) Assuming wiseeclipse works well, you've got more eclipse downtown after Solar, then after Lunar. If you pop a 15s or 20s trinket a few seconds into Solar, some of its SP will be "wasted" on non-eclipsed Wraths.

On the math front:
1) Comparing PDC+.2 to PDC+.26, I think the SF crit rate (before eclipse) should only be 3% higher (IIS) than the Wrath crit rate.
2) I'm pretty sure that the Solar eclipse bonus is additive with Moonfury. I think Wrathcalcs is also using an additive 4t9. If those are correct, your final Solar coefficient is a bit too optimistic.

1) Delaying the use of a trinket only reduces its benefit if you end up getting less uses during the entire fight. In a 5 minute fight you can use a 2 min. trinket 3 times. If you wait 15 seconds before each use you still get 3 uses of it. In a 4:30 fight the same is not true.
2) I should include .6 of a non-eclipsed SF to accurately pair up up-times. This is a good point.
3) If you're able to reliably go straight from a Lunar into a Solar that would be ideal for a 20s trinket. If you cannot, the SP that would be 'wasted' on non-eclipsed Wraths will likely be very similar to the SP that would be 'wasted' on non-eclipsed Starfires.

1) You are correct. I misread in haste and .23 is correct.
2) Further reading confirms the additive nature of eclipse with Moonfury. I will update my above math.

Edit: Above math has been corrected. SP on Solar is only better at 24% paper doll crit or higher.

Has someone made Moonkin BiS list? I didnt find any with search and im just wondering if someone has made it.

Anyone notice that Starfire glyph [sometimes] doesn't increase the duration of Moonfire? I noticed it once on (Heroic) Lord Jaraxxus and now on training dummies. Not sure what's causing this X_X.

Experiencing the same thing. I think i heard something about this having something to do with the T9 2P-bonus? Does anyone else have the same issue?