dedmonwakeen

Thanks. I discovered an error in my handling of the Shaman Earth Shock Glyph that reduces the GCD down to 0.5sec. My code was capping it at 1.0sec and it made me worry that the Druid might also be broken...... but I guess Natures Grace is "working as intended"....

http://code.google.com/p/simulationcraft

 

erragal

As far as I'm aware that's the only ability that can do that, and is likely a product of the change for spell interrupts to be off the GCD entirely. I am not certain whether the survival hunter Lock and Load talents removes the GCD either. That ES glyph is very interesting as every other interrupt had its damage component removed (Shaman have Wind Shock as their GCD-less non-damaging interrupt).

 

thedopefishlives

After changing Moonkin Form to be a per-spell mana discount, my prior statement about haste is mostly invalidated. It is interesting to see the inflation in Moonkin Form mp5 values when JoW is added. If you're wondering how that happens, Moonkin Form's crit boost increases Nature's Grace procs, which at certain haste thresholds will increase the rate of JoW procs. I haven't yet tried to find a haste threshold where the jump takes you mana negative, though.

 

Mlkmn

Would there be a dps increase in forcing the full 10 second uptime for Eclipse? When you are at the 40 second internal cooldown you could alternate between wrath and SF and then keep casting the same spell once you see the proc go off.

You could do this once moonfire has a decent buffer time to keep it from falling off. You want it to be just long enough that it stays on since chain casting SF doesn't increase your moonfire's damage unless the extra tics actually land on the target.

Having 45 seconds of moonfire left when the boss dies is pretty worthless.

Assuming an average starfire time of 2.3s and wrath at 1.32 you would only lose about 0.6-0.7 seconds of your moonfire duration every starfire/wrath cycle. Once you get through eclipse spam you can go back to spamming starfire to get the duration back up.

 

• Adoriele

I did some mucking with numbers on this, assuming MF, SFspam under cooldown, swapping once out.
Chance that your Eclipse proc will affect SF =
Seconds added during Eclipse Cooldown =
Seconds added during SF Eclipse =
Seconds lost per swap =

With this, we can calculate the average net MF duration gained/lost per Eclipse proc. If positive, then, on average, we won't have to refresh MF (though, like gambling, this is not exactly true, as the house always wins when you have 0s left). If negative, we will always have to refresh it at some point.

Net MF duration = AddedDuringCooldown + SFEclipseChance*AddedDuringSFEclipse - (1-SFEclipseChance)*LostDuringWEclipse - LostPerSwap*(SFEclipseChance*SFExpectedCasts + (1-SFEclipseChance)*WExpectedCasts)
7.05 + 35.88%*3.01 - 64.12%*10 - .745*(35.88%*2.245+64.12%*4.012) = -.799s

Since it's negative, we know that, for sure, eventually we'll need to refresh MF. On average, once every (12/.799) 15 Eclipse proc rotations. If AddedDuringCooldown ever breaks 12s, we'd be almost assured not to lose MF ever, no matter how we treat the eclipse proc. This happens if SFCastTime = 2.143s. Otherwise, it's a game of increasing the chance that we get the SF-affecting proc (not possible, since any increase in W Crit also increases SF crit), increasing the amount of time we gain during an SF eclipse (which is functionally the same as increasing the time gained during cooldown), or lowering the amount lost during Wrath Eclipse (not possible, it's always 10s).

Note, also, that this doesn't include IS. You're going to lose a lot more MF duration from keeping it up.

 

ranma

About the GCD off Nature's Grace:

Here is the PTR patch notes
Nature's Grace (Balance): Now also reduces the global cooldown of your Wrath spell by 50% while in effect.

That's before the recent NG's change(move up to tier 3 and bloated by 2 points).

If that's true, the real wrath GCD after NG is 0.75s

 

tezi

I'm I totally lost here with eclipse. MMO-Champion talent calculator is claiming following:
When you critically hit with Starfire, you have a 100% chance of increasing damage done by Wrath by 10%. When you critically hit with Wrath, you have 60% chance of increasing your critical strike chance with Starfire by 15%. Effect lasts 15 sec and has a 30 sec cooldown.

This would mean only 15sec downtime, when you spam SF and wait for the eclipse again.

Is this true what MMO-Champion is telling me? :o

Here is the url to check it out yourself http://talent.mmo-champion.com/?druid

 

• Adoriele

While the wording is correct, the implications are not. Whether they're actually decreasing it to .75s or applying 50% more haste to your GCD than you actually have, the result is the same. You cannot lower the GCD past 1s, and all implementations would if they could. The GCD caused by an NG'd Wrath is always 1.0s exactly.

MMO-Champ has not updated their talent calculator because the Beta itself has not been updated. All of this Eclipse discussion is based off of posts made by GC within the last few days. From what s/he says, Eclipse will last 10s with a 40s cooldown, provide 20% damage to Wrath or 30% crit to Starfire.

I'd also like to revise a statement I wrote earlier. The only true way to never lose MF from Eclipse would be if you can increase MF's duration by 10s during the 30s cooldown, without relying on Nature's Grace. Which means a hasted cast time of 2.25s, 2.05s including 200ms ping. With 11% baseline haste, this would require, at 80, 1159 Haste rating (557 at 70). This isn't reasonable to attain. So, in essence, there will always be the possibility of losing MF, but it will only happen a couple of times per boss fight at most.

 

ranma

Ghostcrawler update his eclipse opinions here.
WotLK Beta (US-English) Forums -> How to fix balance bloat

It seems devs really don't want moonkins to SF brainlessly w/o recast MF for whole boss fight.
However, if moonkins have to recast MF every 45 seconds due to eclipse procs, we lost 1.5/45 = 3.33% cast time for nothing. This simply destroys the whole gain from 3/3 eclipse.

And from the blue post above, didn't he hint that devs might increase the duration to 15s?

 

erragal

If you follow Adoriele's math closely you don't lose moonfire every time eclipse proc, it's simply guarenteed that eventually you will have moonfire fall off unless your Starfire cast time can be lowered to a particular threshold. Her numbers are also intentionally not including NG in order to make the math a lot simpler so we can rightly assume the haste threshold is in fact quite a bit lower (And dependent on crit rate).

When it will actuall fall off is totally down to the RNG. In an ideal average situation based on her initial numbers it would take over 10 minutes for you to need to refresh moonfire at 0% haste or crit, which wouldn't happen in any normal single target fight. Unfortunately the reliance on RNG means you could get screwed early in a fight and be forced to abandon a wrath buff proc to keep Starfire up or let moonfire fall off. If you're rolling a high spell power moonfire letting it fall off will never be the good decision.


The best part of eclipse now is that with only one point you can sometimes have a significant buff to your DPS, but not be so chained to proc watching that our damage output becomes terrible in heavy movement or phase change fights. Additionaly, the skill of the individual player will determine the maximum output that comes from this talent. You can't just set up a macro and get maximum benefit out of it.

 

• Adoriele

The only numbers that don't include RNG are the 1159/557 Haste Rating numbers. Those are a guarantee. Get above that, and you will never see MF drop during Eclipse. Or, really, ever. Everything else is dependent on your Crit rate and Eclipse's proc rate (for the SF-affecting Eclipse anyway).

I actually think I'll really like Eclipse at this point. Like Erragal said, you get out of it what you put into it.

 

Kuruk

There is massive bloat in the Balance Tree and I am finding it very hard to pick a viable Lv70 (not to even mention 80) raiding build so to pick out some talents I am currently questioning:

I still can't see the benefit of Eclipse. It relies on perfect execution and it doesn't seem very useful on fights requiring a lot of movement as Starfire is not really a viable spell then. It will also make maintaining the dots and FF up a nightmare. As Arazan said, far too much space for human error.

What is more, critically hitting a lot with Starfire will mean aggro issues. I am not happy about this big shift to Spell Crit. I'd rather deal steady, even DPS than get lots of sudden aggro spikes. I hope the 30% threat reduction means it won't be that much of a problem.

Master Shapeshifter - 4% damage is very nice but taking it means wasting 3 talent points on Natural Shapeshifter (although that can make sense if we'll be healing much more due to Spellpower changes?) and 2 points on Imp MotW (although at Lv80 this talent's buff might actually become significant).

Adoriele - some questions about the builds you have considered:

Lv70

- Why go so deep in Resto when there are a lot of talents in the Balance tree that aren't maxed and that could significantly increase your damage output (Moonfury, Earth and Moon, Improved IS)?
- Improved MotW at Lv70 is a very minor buff and in every raid there will be a Resto Druid that will have it anyhow.
- Why bother with treants that are wiped instantly in all fights where AOE abilities are used?

There are so many hard choices to make here, I am still considering my build for next week and changing my mind all the time. At the moment, I think it will be:
Talent Calculator - World of Warcraft
Mana regen talents taken as Spellpower changes mean I will probably be healing more often. If Mana really won't be an issue, then those can be dropped and put into Furor or trying Eclipse out.

Lv80

The build proposed looks very good, I'd just take out Force of Nature (for reasons mentioned above) and Typhoon (seems a pvp/solo spell, can't see much use for it in raiding) and put those two points into Earth and Moon so ending up with this:
Talent Calculator - World of Warcraft

That build relies on mana not being a problem though and it's supposed to be a siginificant issue in WLK - the build would need a total rethink if it was.

 

Boswell

@Kuruk

This problem has been addressed and fixed through reduction of Moonfury and Earth and Moon to 3 point talents.

 

◊ Lorewanderer

(e) multitasking, beaten. Adding GC quotes (including one new one) to make this useful:

E&M and Moonfury are going to be 3 point talents. They aren't that way on the talent calcs because these are announced but as of yet unimplemented changes.

So it looks like they're still undecided about the duration of the buff.

 

Shuror

It isn't for nothing. The extra Moonfire you need to cast also does instant damage and has a chance to proc Nature's Grace.

Nevertheless, it's odd that Moonkins get such a glyph for only one of their main spells. I'd like to see a similar glyph for Wrath and IS.

 

Soultrigger

We are debating about the SF buff, but what about the other way around?? What I mean is, I am chain casting Starfire with MF rider, and I get the Eclipse proc, is it worth as a dps increase to cast Wrath for 10 secs?? Or is SF buff so superior that I should use Wrath to proc SF buff, because this is the right way for PVE?? With the last GC post I am under the impression that they want us to cast wrath in pve on eclipse, and SF on pvp during the proc.

 

• Vykromond

I'm pretty sure you're just misreading the foregoing discussion. The general tenor is to cast SF and use the Wrath buff, not vice versa.

 

thedopefishlives

My first hack at modeling Eclipse is now in the source version of Rawr.Moonkin. It has some support for non-spam rotations, missing calculations for losing/refreshing DoTs during the proc. It reinforces the conclusions on this thread that the first point is definitely worth taking, at ~1% of your DPS in a non-spam rotation, but subsequent points are worth significantly less.

 

Darian_TruBlade

The Shaman Earth Shock glyph lowers the spell's GCD past 1s. It's technically feasible, it's just a question as to whether Blizzard wants to put in the effort to make Wrath scale better with Haste.

"A man's IQ, yearly income, sexual prowess, ingenuity, physical appearance and generally every other aspect of his character can be condensed down to four digits: his Arena rating." - Zechsy [70 Rogue - Skullcrusher (EU) - 10/23/2007]

 

Soultrigger

The new eclipse seems good enough, but I would like to see slight changes in the %:

Starfire buff = 50% increased starfire crit chance. So if we are at 50% crit chance raid buffed, during the 10 sec we would get:

20% haste (NG proc)
3-4 SF Crits
6-8% Mana return, similar to Thunderstorm in different mechanics, and similar cooldown, 40 sec cd means around 50 sec proc average, for 10 sec, so 1 ppm.

About the Wrath buff, all that I want is that it will give increased dps to MF rider, SF spam rotation for those 10 seconds. I don´t know what the dmg% should be to accomplish that.

This way Eclipse could be a choice between mana regen or increased dps.

PS: Vykromond, yes, I am not a native english speaker, so I do have some trouble fully understanding a comment here and there, and since so many people try to argue, it gets a bit confusing, my bad.

 

Conquistador

I haven't really seen any conclusion about Eclipse's two separate procs. Is it possible to proc Eclipse_SF while spamming wrath while under the effects of Eclipse_Wrath (for a standard IS, SFxN, MF rider rotation)? Do the two different procs operate on a separate cooldown or do they share the same 30s cooldown?

 

◊ Lord BEEF

No, it won't.

The amount of threat that tanks generate now cannot be understated. If you go balls to the wall with damage and get a lucky crit streak you might end up at 60% tank threat, but probably a lot less. If by bizarre luck the tank misses like crazy and you crit like crazy, a hunter or rogue may have to help out on threat with misdirect or tricks of the trade, but I haven't seen this happen.

Maybe it might happen a few tiers from now if tank threat doesn't quite scale as well as dps, but level 80 tanks are holding aggro over bugged/overpowered specs doing 5000 dps easily.

 

Kuruk

Thank you for the clarification Boswell and Lorewanderer, I have missed that in the preceding discussion.

Lord BEEF - yes, I forgot to take buffs to all the tank classes into consideration. Considering that more emphasis in being put on spell crit for all caster DPS classes, that does make sense.

I still can't see the benefit of taking Master Shapeshifter at Lv70. You are spending 12 talent points on a 4% damage increase plus a further minor increase from Furor combined with Lunar Guidance.

For 11 points, you can get Improved IS, Improved FF and Furor which will together give you more than 4% extra spell damage (plus iFF will benefit the raid as a whole too) and the leftover point can be put into Dreamstate for a bit more mana regen.

 

• Adoriele

iFF only helps the raid if you have no Spriest, which is unlikely at 70 due to the lack of 10-mans. Further, once you reach tier 9 for talents, you have to fill in one of the following to progress to E&M: iIS, iFF, Eclipse, Brambles (you can do it without, but that wouldn't be optimal, which seems to be what you're trying to do). Eclipse is really only great with one point, Brambles can be passed over easily, so you're already filling in either iIS or iFF, and iIS is the better of the two until we can confirm that all FFs provide the self-crit bonus.

So now we're at 3/3 E&M, 5/5 Furor. You have 8 points left to spend. 7 can go into Resto with any degree of usefulness, for 4% extra damage, more raid utility (marginal, though, as it depends on having no other Druids with iMark), and some personal utility for when you need to Battle Res, with the last likely in Eclipse, ala Talent Calculator - World of Warcraft. Or, you can fill out iFF, Brambles, and put the last point in Splendor, ala Talent Calculator - World of Warcraft. Given my last WWS, each Treant was putting out about 1650 damage, or ~5k per cast. Over 2 min cooldown, that's ~41.7 DPS. Brambles would add 15%, or 6.25 DPS. With us doing ~2400 DPS in 3.0 with the same gear not including trinket/etc. procs (those 70 numbers in my sheet are pulled from my actual gear as of about a week ago), that's a .26% increase. 3.74% to go, and iFF won't come close to covering it, since 3% on top of the 45% we'll be sporting for Starfire is about a 2.01% increase. 1.74% to go. Nature's Splendor adds about 33 DPS, about 5 times that of Brambles, so 1.25%. So yes, I think Master Shapeshifter is worth spending those talents in Resto.

tl;dr: Going for Master SS, in my gear, gives me ~2400 DPS. Going without, and picking what should be the optimal talents, gives me ~2300 DPS.

 

Arazan

Here's the thing about Master Shapeshifter... it's not that you spend 16 points just to get 4% damage from MS, but that if you invest 16 points in resto, you get +10% int (good mana and extra crit), intensity, OoC, imp Mark, *AND* Master Shapeshifter. Imo the true prize is intensity, but I think in general it's just a good investment of talent points to go 16 resto rather than 14 resto simply to ignore Master Shapeshifter but still pick up OoC and intensity.

That being said, I'm still really confused as to what people are spending talents in that's making their 55/0/16 builds leave out so many crucial talents. With the changes they stated they'd make to Moonfury (or was it Vengeance?) and EnM to make them 3 point talents, you can do this:

http://talent.mmo-champion.com/?drui...03213335001031

and still have 3 points to fill out whatever you want. Gale winds and 1 point in eclipse, 3 points in dreamstate, 2 points in imp moonfire and 1 point in eclipse, 3% haste from celestial focus, whatever. In any regard, no 2 talent points that aren't already taken in that build would be more beneficial than Master Shapeshifter.