So it's time. There has been little to no response from Blue on the Beta forums when it comes to Balance Druid mechanics. I'm not worried that Blue doesn't care, because I think they do. I think they've got some cards up their sleeve that might be changing what we complain about. Slowly but surely, they are trying to patch Eclipse till it "works" for us. Maybe we don't need a patch. Maybe we need a full reset.

Eclipse - The Theory

I think it's important to be very clear about what we want Eclipse to do for our rotation. Currently in Beta, it feels like Eclipse acts as a switch. When Eclipse procs, I switch nukes. I'm offered a damage incentive to perform this switch. I might or might not want to cast DoTs depending on their DPET under their respective Eclipse. Given the latest verbiage from Blizzard, I will likely want to cast Starsurge after I reach the midpoint of the Balance Bar to help proc the next spell switch. This interaction is rather boring and although I can't break it down to a static rotation due to crit, I can come pretty darn close.

While I agree that Eclipse should entice me to change what I am doing, I'd like to see a more involved method for interaction. Why must it be just a flat damage boost? I believe that Eclipse ought to fundamentally change my spells. I think at some level, Blizzard agrees with this as well. They want Eclipse to act as an incentive for changing spells. While their solution (having it buff damage) is clearly the simplest solution, it's also not that interesting. So how do we spice up this mundane ability?

Eclipse the Mechanism

As stated, the current Eclipse mechanism is rather uneventful. Perhaps it could offer DoT extension, drastically increasing the DPET of my DoTs and therefore making them an integral part of my rotation. Perhaps it could add a DoT to my nukes, not unlike what Languish (4T10) already does. Perhaps it could allow me to place charges that will explode when Eclipse expires. Or maybe these charges could trigger some Savage Roar type buff that will increase my damage, thereby making it my goal to swing from Eclipse to Eclipse as rapidly as possible to maintain this buff. There are many options (and I'm not saying any of these is particularly compelling or even the best way to go).

The point here is that there are many options for changing the feel of Eclipse that don't involve a flat damage booster. If anything I would suggest that Moonkin form (if it becomes a cooldown: finger's crossed) be the ability that offers a static damage increase and Eclipse provide the more interesting rotational modification.

What We've Got So Far

So I suppose I should point out that while it seems like some of the Balance community is unhappy with the modifications to Eclipse, we have received some improvements. First and foremost is the extension of the Eclipse buff to 45 seconds, allowing us some leeway for movement without "losing" Eclipse uptime. That alone helps pacify our movement penalty, allowing us to pick up our rotation right where it left off.

Eclipse is now buffable through the mastery stat. I think it should continue to be buffable through that stat as it allows our Eclipse some extra scalability outside of just proccing faster (through crit and haste). This new model seems like it will always have roughly the same uptime for Eclipse due to the fact that it cancels at 0 power. Will that make mastery more valuable than either haste or crit? We'll have to do some number crunching, but it certainly won't hurt. Personally I'd like to see Eclipse last 15 seconds or till the halfway point, whichever comes later.

We've gained a smidgen of control over Eclipse procs thanks to the Balance Bar (although this also adds some rigidity to our rotation). No longer are we waiting for that special Wrath crit or suffering the agonizingly slow Starfire cast praying for that crit at last. We have a visual marker and a set progression rate that will allow us to control when we proc Eclipse. Double-edged sword imo, but it's better than no control at all.

What We Want

We = Me, myself, and I. The three of us have conferred and we'd like to see more dynamic introduced. It's a poorly kept secret that I am a fan of complex rotations with situational aspects. I like the feral rotation. If my guild didn't have a kick-@$$ feral druid, I'd try to slip into that position. But I'm also attached to my Balance Druid for sentimental reasons. We asked ourselves if Eclipse really introduced complication into the rotation. We came to the conclusion that Eclipse is like a switch. Throw it up and you cast Arcane spells. Throw it down and you cast Nature spells. You throw the switch by reaching the end of an Eclipse bar. Lights on . . . lights off. Not very fascinating. Now while I am grateful for the additional control, I still would like more.

A lunar and solar eclipse should feel different. They should do different things. I should want to make a choice between the two. Is it better to push a Lunar Eclipse now, or a Solar? And the situation ought to determine that. Mana efficiency has been mentioned a few times. Perhaps one buff should scale well with haste while the other emphasizes crit. This would allow for a meaningful choice during heroism. Again, possibilities here are only limited by balancing the damage output. Whatever changes come, I seriously hope we promote more spell synergy and less "Cast A till proc, then cast B. Repeat"

So they are here! A return to simplicity and it is simply. . . crappy. The trees come with a caveat: Druids are not as far along as Paladins in development, and as we all know, Paladins get fixed about half-way through an expansion so that's saying something! You can check out the talent trees over at mmo-champion. In lieu of going through talent by talent, I thought I'd just hit the highs and the lows.

Highs

Several talents were removed (obviously) and for the most part I approve. Brambles got the axe, forever putting to ground the Owlkin Frenzy or Brambles debate. Typhoon is baseline now, freeing up some clutter from the tree. I wonder if it will get nerfed to the ground if every druid has access to it. Regardless, I'm glad to see it gone from the tree. I hated wasting a talent point on an ability that I only used on two fights. Nature's Splendor also received the axe, getting built in to the skills by default. Moonkin and imp. Moonkin went missing (and I can't say I'm sad). While I'd rather have Moonkin form as our spec defining ability, I'd also like to see it as an on-use ability. Read more about that below. Celestial Focus got removed and hard coded for choosing the balance tree. Vengeance also got removed and hard coded. That represents a fairly significant change in the school of thought between melee and caster DPS. But then again, melee is now hitting quite a bit harder than they used to. Back in the olden days, the 100% crit was to make up for the fact that melee hit more often but for less while casters did the large damage, large cast time spells.

Lows

Earth and Moon was taken out. I like this talent and will talk more about it below because I put it back in my tree. I feel like tiers 2 and 6 really lack the bang for their relative buck. Tier 2 is just boring and tier 6 probably won't see true use until I decide to PvP or raid. Until then, it's a filler tier. Omen of Clarity is gone too. Granted, it was a mana regen for us. But for ferals, it was a crux and a serious DPS increase. I hope it's trainable, but if it's not, I wonder how ferals will be compensated. However these are just minor annoyances when compared to what I consider the ultimate low. Where are the options Blizzard? I recall a mission statement of sorts that said they wanted to provide us with interesting talent choices. While invariably we will crunch the numbers and determine the best of the best for talents, we ought to have some relatively close choices that we can't pick up. Filling out a tree completely (except for Lunar Justice) and then scrounging a few points in the Restoration tree is a cookie cutter build. Every single raiding Balance Druid will have exactly the same build. Neither interesting or choice.

How Rel Would Do It

This is my redesigned talent tree. I kept Starsurge in the talent tree because I would like to see Moonkin Form as the "defining" ability of the Balance spec. But with an addendum: Make it a usable cooldown like Metamorphosis for warlocks. Clearly it would need to be less powerful (or scale with level).

I fleshed out the top row of Balance to provide more restoration options for advancing in that tree. They can get a full 5 points in the top tier with Genesis and Nature's Majesty, both which benefit them. I moved Lunar Justice to the top tier and reduced it to two points as it is strictly a leveling talent and I believe should be available to a leveling boomkin ASAP.

Given the relative power of Nature's Grace, I moved it to the second tier. I realize it is likely to change, but I still think it should be seated outside of the first tier. Moonglow appears in the second tier still. Both options might be appealing to a Restoration druid, giving them the options they need in the second tier.

In the third tier we find Improved Eclipse, Starsurge, and Balance of Power. Part of me thinks it is too early for Balance of Power, but I don't know what else I would place in this tier if I removed it. I linked Starsurge to Lunar Guidance in the 4^th tier. I also linked Improved Eclipse to Euphoria, just as Blizzard did. I read some people complaining about that link, but it seems not only appropriate, but obvious given the continual buffing of the amount of Eclipse Energy generated by each spell.

The fourth tier brings Lunar Guidance, Solar Beam, and Gale Winds. I actually tied Solar Beam into Lunar Guidance, but I'm 50/50 on untying it. I realize that Solar Beam doesn't seem like a PvE talent in a 25 man raid environment, but given the likely popularity of the 10 man format and all of the preceding dungeons, I think a silence will provide a good chunk of utility to the Balance druid, expanding their possible role.

Tier five offers Euphoria, Force of Nature, Owlkin Frenzy, and the return of Earth and Moon. I like Earth and Moon, as a talent. I don't need this type of talent to be unique to Boomkins, but I do like the concept. While it does seem rather dull in its description, it offers a personal buff for buffing the raid's DPS. I like talents that encourage you to be a "team player" and reward you for doing so. Euphoria seems to be our mana regen spell now. With Dreamstate, Omen of Clarity, and Intensity gone, Moonglow is about our only mana regen spell remaining. Technically, it's more of a mana saver than a regen talent too. The fact that Euphoria buffs Eclipse while offering mana return is icing on the cake.

Tiers five and six are as Blizzard designed them. I think if Starfall is going to be our 31 point talent, it needs to stay as strong as it is now on Live. Currently, it seems to have lost its splash damage, which I'm ok with. Not much splash damage going on during boss encounters most times anyway. It's hard to pinpoint Fungal Growth's utility as it is NYI. I think this talent may need tweaking (fuse with brambles, imo) in order to appeal to raiders, but we shall see.

Smo over at his blog posted up our first look at the Alpha testing talent trees for druids. I'm just gonna work my way from the top down in the Balance tree so don't take my first comments as the ones that I feel strongest about. It's just easier to organize that way.

Tier 1

Starlight Wrath
This still decreases Wrath's cast time. I was hoping this would interact with Wrath in a different way, but I suppose they wanted to make sure that Wrath and Starfire don't feel like the same spell. I appreciate that sentiment, but will be very irritated if the haste scaling issue for Wrath continues past WotLK.

Genesis
Boring. Looks like its still going to be a filler for Resto druids to get through the tree. I will say that the verbiage is currently unclear, but it looks like it only affects the periodic portion of our spells. Since it looks like the direct damage of Moonfire will have some significant boosters (including during movement) and IS looks weak, I'm not seeing this talent as useful.

Tier 2

Moonglow
Unchanged. A 9% Reduction in mana cost might seem trivial, but I understand that we might actually have to give a damn about mana. It sure could make a rotation interesting if we had to worry about going OoM.

Nature's Majesty
Unchanged.

Improved Moonfire
Revamped. It now only increases the direct damage of your moonfire spell. This, compounded with a few other talent changes looks like a herald to the return of "Ye Olde Moonfire Spam".

Tier 3

Celestial Focus
Pushback on everything . . . but Wrath and Cyclone? Maybe spells under 1 second won't be able to be pushed back. But this seems like a pretty poor design considering how crucial Wrath and Cyclone are for our PvP arsenal.

Solar Beam
Well, I'm charging my lazer cannon. But let me see if I've got this right. Solar Beam will call down a big, highly visible ray of light. Moonfire shoots down a distinctive arc of silver light. Another talent is going to leave a silver ray over a dead body. Maybe if we're lucky, we'll get a talent that puts a perma-mark over our heads letting people know where we are at all times. We better hope that people fear us in PvP or we'll be a gank magnet.

Ok, that was a rant and said nothing about the actual talent so I suppose I should get back on track. A guaranteed 3 second disorient on an instant cast with at least 1 more second of silence as they run out. Combined with roots, this COULD be very powerful. It can also act as a barrier in doorways and things like that, forcing people to either stay out of los or bite a few seconds silence and run into full view.

Nature's Grace
I'm shocked. Stunned. Flabbergasted. What in the name of Zeus's butthole is this talent doing here? Still in tact. Still causing scaling problems with haste. Maybe they are hoping that with the new changes to DoTs with haste that we won't care. Maybe that's true. Blah.

Nature's Reach
Nothing new here.

Tier 4

Nature's Splendor
Same-ol' same-ol'

Lunar Justice
This is the marker spell I was talking about earlier. I don't really understand it's point outside of leveling unless mana is going to be a huge issue without raid buffs. I sincerely hope that it is not a problem to such a degree that I have to pick this talent up and carry it around in my raiding build. Very nice leveling implications for us though. And AoE grinding will be fun.

Tier 5

Brambles
This talent makes a large jump from tier 3 to tier 5. I find it funny that a talent that we barely consider taking now would be budgeted to a higher tier without any improvements, practically assuring that we won't need to put points into it. Especially considering that this tier contains Vengeance, one of our must have 5 point talents.

Starsurge
Well here it is. This is the "Nature's Torrent" that we heard tell about. It comes with a knockdown so it definitely will help our peeling ability in PvP. I'm actually glad it's not instant. It will make planning a bit more interesting in PvP and maybe even PvE. We see an Eclipse meter interaction from a later talent, but it seems rather minor (especially if the bar is 100 Lunar and 100 Solar).

Vengeance
Still rockin out that increased crit damage. It's no surprise this talent didn't change.

Dreamstate
Could be key again if mana goes down the pooper. I suppose it all depends on how high our int numbers go. It got moved up one tier. It also now leads to Euphoria. More on that later.

Tier 6

Gale Winds
Up a whopping 3 tiers from its original place, Gale winds remains unchanged. I think they just ran out of creative juices. Where's the addition of increased Starfall damage?!? And Wild Mushrooms! Cause that wouldn't be OP at all. C'mon Blizz, let us have the one button press to kill everything ability!

Lunar Guidance
Providing a nice little bonus to the radius of Solar Beam and a small chunk of Eclipse energy from Starsurge. I think this talent is supposed to introduce "complexity" into the rotation because its damage will be so good that we will have to cast it. Things that make you go hmmmm....

Balance of Power
This is our hit talent. Converts spirit to hit and adds an additional 4% on top of that. No surprises here. Actually, the conversion ration of spirit to hit is 1:1. That's higher than I expected, so maybe one surprise.

Tier 7

Moonkin Form
Moonkin Form! Now with added armor and raid wide critical strike! In a bit of a switch of roles, Moonkin Form now offers a raid wide buff, which is nice. I'm glad to see them up this from the leak that had only increased armor.

Improved Moonkin Form
Switching from crit (as leaked in alpha) to haste, with this talent, we will now bring 5% crit and 5% haste to a raid. That's a nice little package. Not to mention our tangibles such as battle rez and innervate. It seems like a good day to be a druid when people start looking at raid composition.

Tier 8

Owlkin Frenzy
It seems like its reverted to the pre mana regen version. Not too awesome.

Wrath of Cenarius
So this is rather odd. It only affects Moonfire, and it only affects it while you are moving. Welcome back, Moonfire spam. How I have missed you! Goodbye Moonfire Glyph, how I have loved you! And that means Starfire glyph is probably going bye-bye too. So then . . . focus, starfall and IS (our least buffed dot) or if we're lucky we'll get a Glyph of Slow Wrath.

Tier 9

Eclipse
I'm having trouble reading this one. It will obviously be necessary, but it also looks like you will have an Eclipse Meter without this talent. Maybe we will see a hybrid resto/balance build with offensive possibilities in PvP, given that critical hits will move the bar. Anyway, speculation aside, this is gonna be a staple talent.

Typhoon
Still the same.

Force of Nature
No changes. I was hoping for one stronger pet, but I suppose the three pet thing has some validity. If we're lucky, blizzard will allow pets to stand at 135 degrees from the front of the mob to avoid dragon tails and poop effects. (Sorry, I seem to have lost my couth filter for the moment)

Tier 10

Earth and Moon
50% reduction on its effectiveness. I don't mind that though. It goes in line with decreasing the raid bonuses (which will really help them balance 10/25 man expected outputs). We retain the personal bonus, so that's a plus.

Fungal Growth
Here's the snare from Wild Mushrooms and ... treant death? Well, in PvP, that will be nice. But whatever happened to Wild Mushrooms being a synergistic piece of the Balance rotation? So far this is the only talent I see that impacts them. And that impact has a very limited PvE scope.

Tier 11

Starfall Not awesome. Maybe I'm reading it wrong, but I see almost a 400 base damage reduction with no splash damage. Are you kidding me? I'm not griping because I lost something that was OP. I'm griping because our 51 point talent is so lackluster that I wonder if I'll even want to take it. And Starfall has always been one of my favorite Balance spells since its inception. I'll take it because I love effortlessly aggroing a large number of mobs to Hurricane down. But I'm not gonna take it without saying that it seems like a pretty poor change.

---

Resto Tree

Blessing of the Grove (BotG)
BotG is another talent that is synergistic with Moonfire. I can't imagine that we would pass on this for one of the other talents given that any time we move we will be spamming Moonfire.

---

Summary

I'm pretty disappointed with many of the changes. They left an inherent haste scaling problem in tact. They eliminated most boosters to IS. They boosted the direct damage of Moonfire so much that it's hard to believe we would ever glyph for it and, consequently, Starfire. I feel like many of the spells gained a PvP aspect, which I am glad for. I would love to feel a little more secure in PvP. I dislike that many of our spells will now leave obvious markers of our presence. Moonfire was a tell-tale sign of our presence in a fight. Now we'll be seeing three beams of light! And a couple of them stay for 10 or more seconds!

While I still see the haste problem, especially given that it is both our Mastery and a passive on Improved Moonkin, I feel a lot better about synergy. I find Eclipse less than interesting and really wonder how boring GC's life must be for him to find it "really excit(ing)". Word on the street is that we will see different mechanics for Eclipse popping up. I pray the word is right.

I really wish I could get on and test the synergy right now but I don't have an alpha key so I can't do that. Here's to hoping those of use from the druid community that got a key are giving specific and prolific feedback.

The patch went live today and while the servers have been down I have been thinking about the new rotation. Then I read Macbook's post on the new rotation and it sparked into text some thoughts that had been brewing in my head. Here's the comment that I left on his site:

I would love to make a recommendation for spell twisting. Start by casting Wrath. Since it only has a 60% chance on crit to proc Eclipse, here is what I recommend. We use to have to wait until the actual Eclipse proc triggered (on impact) in order to switch spells. That generally led to one and possible two extra Wrath casts and some time lost off of Eclipse.

Now, I would recommend switching the instant you get a Nature's Grace buff as you are opening your rotation. If you don't proc Eclipse, meh, whatever, looks like you're gonna start with Solar. If you DO proc Eclipse, you will get extra time casting Starfire under the Eclipse effect. You might, in fact, squeeze in one extra Eclipse.

If we keep hold of Squawk and Awe to track the individual internal cooldowns, it might also answer the question of how long to stick with a Lunar Eclipse precasting phase. I.e. if both Eclipses are off cooldown (or will be in ~2 seconds) then it is worthwhile to "anticipate the Eclipse" by doing a premature spell switch on a crit wrath prior to its impact and subsequent possible proc on the boss.

Hell you can even switch back to Wrath after that one Starfire (since you will have used all of the Nature's Grace time anyway) and continue attempting to proc your Lunar Eclipse. Good stuff eh? I need to post this crap on my blog and not in your comments Macbook!

TL:DR

The new rotation eliminates the fear of a "wrong" Eclipse proc. This enables us to anticipate the Eclipse with zero penalty. Previously, if we anticipated a Lunar Eclipse on a Wrath critical, got it wrong, and then crit with the anticipatory Starfire, we would proc a Solar Eclipse and bemoan the loss of DPS due to the incorrect idol and likely some gearing issues. I do think some people were making a mountain out of a molehill with this issue but in terms of max/min theorycrafting, it was a DPS loss.

Enter 3.2. Now, if the exact same situation were to happen, we would not be penalized. In fact, we would be rewarded. We'd have procced a Solar Eclipse with a Nature's Grace influenced Starfire, cutting down on any theoretical precasting phase time. If the Wrath crit does proc an eclipse, we would get to utilize more of the Eclipse time casting the impacted spell. All-in-all, this is a win-win situation for all of us. So don't be late! Anticipate!

Granted, once 3.2 is released, this becomes a bit of a non-factor but I wanted to address the "soft-cap" of 400 haste for rotations. I will open by making the statement that for a Lunar Eclipse, the 400 haste cap is asinine. That's right. It's complete fodder. By definition, the Lunar Eclipse has a 60% chance to proc on a Wrath crit, and while I realize sometimes we have a dry run with many Wrath crits and no Eclipse proc, that is often only our memory latching to the bad instead of the average.

Ok, so why is the haste cap ridiculous for a Lunar cycle? These events are independent (i.e. one Wrath crit does not impact when the next Wrath will crit). Instead of looking at Eclipse as a 60% chance to proc on crit, let's look at it as a 40% chance to not proc. Let us then look at the successive chances that Wrath will not proc Eclipse.
As we can see, having 5 independent crits with no proc is a 1% chance. Now, what does all of this have to do with the 400 haste soft cap? Well, the haste soft cap includes the Nature's Grace haste bonus. If you're not under the influence of Nature's Grace, you can afford more haste.

The current logic states that any rotation that contains Wrath should have a soft cap of 400 haste and I'm afraid I've just seen too many erroneous posts asserting the anything over 400 haste is a waste. It's not. Any rotation that contains many Wrath casts, including crits, with the intention of continually casting Wrath after the crits (read Solar cycle) should aim for a haste rating of around 400. Any rotation that plans to switch spells after a Wrath crit is more than okay with a haste value over 400.

I've been puzzling over how to represent this mathematically since it's very tough to model the probability of some chance encounter when I more or less know the previous events. The problem we encounter is that probabilities are not static once the sequence is in motion. Once I've gotten my first crit and the Eclipse did not proc, I reset the entire sequence. Due to the fluid nature of this problem, I have decided to make a table of likely sequences and attribute each sequence a percentage. That way I could see just how likely I was to be casting under the influence of Nature's Grace. I assumed a worst case scenario (which is that all the crits have 3 seconds of Nature's Grace associated with it). Below are the weighted values of the Nature's Grace time through n = 5.
As I sum this, I find that I get a value of 3.198 seconds of average uptime on Nature's Grace given my worst case scenario of non-overlapping crits. I would be working with for n=1 through infinity. I treated this like a Power series and came to the conclusion that it summed to 3.333 repeating, of course. Now, I did this over the course of several 5 minute spurts throughout a few weeks, so I could have fractured logic (in which case, tell me!!) but I'm forced to come to one conclusion here.

TL:DR

The results seem to indicate that I have an average of 3.3333 seconds of uptime on my worst case scenario in a Lunar precasting phase. That means that for any haste over 400, we clip the GCD for an average of 3.3 seconds. I really don't feel like this is such a bad thing because I showed earlier that the average Lunar cycle takes around 35 seconds. The rest of the rotation (32 seconds) is spent without clipping the GCD. That's 91.4% of the rotation that benefits from haste over 400.

The haste soft cap is exactly that: Soft. Especially for a basic Lunar Rotation. Naturally, math gets more complicated when Starfall is accounted for, although then we're talking 67% of 91.4% or 61.2% of the rotation. I say this because Lunar cycles are using Starfall at most every third Eclipse. If you plan on running a Lunar Eclipse (until 3.2), haste is still good for you. Don't be afraid to go over 400 haste!

So the Balance Druid community is in a buzz about what our new set bonuses will be. Let me start with 2T9.

FINALLY!! we get critting DoTs. Or Dot rather, but hey, that's better than nothing. I'm a little sad that when I look to the future, we stand to lose this ability in a higher tier, but for the moment, it's making me a very happy camper. I haven't seen very much to complain about from anyone in this department. But 4T9 is a different story...

Well . . . despite the beauty of 2T9 it is marred by the placeholder seeming quality of 4T9. A recycled set bonus that is also a downgrade. Many druids have weighed in on this set bonus and have had varying responses. Hamlet seems to think that there's no problem with it because of the relative strength of 2T8. He predicts a buff to 4T9 will be coupled with a nerf to 2T8. Graylo is clearly irritated by the bonus and lobbying for at the very least the same bonus as 4T7. Haven't heard from Macbook or Xiera formally yet. Lissana seems indifferent. And Relevart (hey, that's me!) seems irritated by the lack of originality.

I suggested an alternative bonus and semi-promised some math on it here. That may still happen, but with the chaos of moving and getting ready to go back to work, I find myself lacking the time and the energy to do some theorycrafting. I will only summarize that I feel as though people have skewed the true goal with set bonuses. Set bonuses don't need to be progressive (scaling with each tier) or at some point in time we would be unable to continue to scale without another gear reset. Given the finite nature of gear (we only wear it for so long) I think what people really want are interesting set bonuses. For instance, if the Balance druid 4 piece was ALWAYS 5% crit, I'd gladly move up to the next tier because of the stats on the pieces themselves. Likewise, if it went from 5% to 4% but I gained 1.1% crit from the stats on the gear, I would also happily move to the next tier. The real tricky part of designing this gear is making a 2 piece bonus that is attractive enough to warrant breaking the old 4 piece bonus. In some ways, this makes a 2 piece inherently stronger than a 4 piece (to preserve the continuity of the break). I would not mind seeing 2 piece bonuses continually being stronger than the 4 piece bonus as this would allow us more freedom to pick and choose gear.

I know it sounds like I'm saying nerf the set bonuses, but well . . . I am. A set bonus is nothing more than an incentive to gather and equip a bunch of sub-BiS gear. I would much rather see these pieces of gear act as placeholders until they were replaced by hard-mode gear. Essentially, I'm asking the question of whether the tier set should be the best gear. My personal answer is no.

Back to interesting set bonuses. Looking at the 4T8 bonus. An instant Starfire triggered by a DoT tick of IS. I love this set bonus because it involves the actual use of 2 spells working together to increase DPS. If you don't cast IS, you won't get a proc and if you don't ever cast Starfire, you won't benefit from the proc. There is plenty of room for set bonuses like this one. I suggested a bonus to increase the use of Treants in raid instances. I've seen a suggestion that increases critical strike damage by a percentage. Bonuses that incorporate spell synergy, not straight stats, are much more interesting and, in my opinion, provide the player with more satisfaction. I can see a Wrath of Elune proc and use it, but I can't always notice an extra 5% crit when I'm running a dry spell of Wraths.

That's about all I've got time to say today about this stuff. More to come later!

UPDATE:

They changed the bonus to increase the damage dealt by Starfire and Wrath by 4%. I feel that this is a great change and put to rest some of my worries about the recycling of bonuses. While I still feel that this is a bland bonus, lacking somewhat in creativity, it is much more mathematically sound. As I'm sure I've expressed before, I do not like bonuses or procs that are based on random events because they rely on averages, and as we all know, there are often extreme outliers to these averages (recall the "I cast 20 Wraths with no Eclipse proc!!").

I was enormously surprised to see that Blizzard changed the way our bread and butter DPS talent functioned. It was a very simplistic solution to a few of our problems that also gave Blizzard what they wanted. Blizzard had implied that they wanted Druids to use both spells but that they were unhappy with requiring an addon like Squawk and Awe to manage the internal cooldown. For those of you who don't know what I am talking about, allow me to break it down:

The Starfire and Wrath buffs from this talent are now on separate 30 second. cooldowns. In addition, it is not possible to have both buffs active simultaneously.

Them some powerful words! So with that in mind, let us look at the new Eclipse and why we'll no longer talk about Solar vs. Lunar or even say that Squawk and Awe is necessary for a good rotation.

The New Eclipse

With separate cooldown timers, it will now be possible to spend 15 seconds in a Lunar Eclipse, then continue casting Starfire until you proc a Solar Eclipse and suffer no internal cooldown whatsoever. The new spell mechanic will allow us to drastically increase our time under a given Eclipse, forcing us to switch spells only when the new Eclipse procs. We will no longer have to track an internal cooldown because even if we procced the second Eclipse the instant the first finished, the cooldown on our first Eclipse will have just finished by the time we were ready to proc it once again.

I had mentioned that I thought the Idol was the best thing out of all of this stuff, yet here appears to be one of the greatest changes to the Balance Druid in quite some time. Now you should be asking yourself "What is Rel smoking?" The answer is. . . well I'm pleading the fifth there. But lets take a look at what this change does for gearing.

Previously, someone who ran a Lunar rotation was much less effected by the 400 haste soft cap. The could blissfully stack on haste and not suffer much loss from Wrath clipping. Say goodbye to those days. Now, Wrath will compose a much larger chunk of all Lunar druids arsenal. Therefore, this mechanic change simply reduces the value of haste on all gear once a player reaches 400 haste. Graylo ran his model on the new Eclipse and found that crit rating was better than haste rating after 400 haste.

I really dislike being forced into certain gear because a spell is clipping the GCD. I feel that this indicates poor planning on the designer's part and often results in a reduction in good gearing choices since we are already limited by a hard hit cap. I'm still praying for a new Glyph that modifies the Wrath cast time to allow for more gearing options. Regardless we will most likely determine a BiS anyway and that will guide a good majority of players so perhaps I am being a little overly pessimistic. I guess we shall see.

The second thing I don't like about this change is that it simplifies the druid rotation a great deal. Personally, I like the idea of a priority list instead of a rotation. I feel that this change takes us one step away from a true priority and one step closer to a rotation. I spam one nuke until a little orb appears over my head or MSBT says ECLIPSE!! and then I switch. The only true decision making I have to do is with regards to DoT's and long cooldown abilities. Do I use Starfall with my Lunar or Solar Eclipse. Do I throw down my Treants now? Innervate yet? (Oh PS, all you Lunar folks are gonna find yourselves using more mana than before. Wrath does that to you!).

I guess all in all I am just a little worried that this great change was made to look so great because it was really just a bandaid. In some ways I see this conflicting with Blizzard's stated desire to get away from the rotation and emphasize a priority decision based play-style. I worry that this little piece of cheese dangling in front of me will be snatched away at the last second and I'll be left with only a rumble in my stomach and a heart full of disapointment.

Next up. . . set bonuses. If you thought this was ranty . . . just you wait *eyes glint in the darkness*

I've been trading views with Xiera over at The Moonkin Repository lately over the value of the 4T8 set bonus in relationship to both cycles. The real concern is the 4T8 bonus's synergy with the precasting phase of the Lunar rotation and the Eclipse phase of the Solar rotation. As many of you know, Graylo recently posted on the DPS increase his model shows for 4T8 on both rotations. I think (or hope) that everyone agrees at this point that when the change goes live, 4T8 will be beneficial and deserves a place in any BiS list.

Today, I took a look at the semantics of using a 4T8 proc at the "wrong" time. My initial gut feeling was that it would be worthwhile to use a 4T8 proc under the effects of Solar Eclipse but would result in a decrease if used in a precasting phase of a Lunar rotation and caused the wrong Eclipse to proc. Let's start with the simpler case, the Solar Eclipse proc.

Solar Eclipse

Even though the Starfire from 4T8 is instant, it still evokes a GCD. Since Wrath has a cast time equal to the GCD, I am going to say that for all practical intents and purposes the 4Tb proc and Wrath have the same cast time. I am also assuming that the Idol of Steadfast Renewal is equipped.

Because we have 4T8, we naturally have 2T8 (durrrr statement, I know, but I forgot it in my initial calculations). We expect to see a 45% increase to damage for all Wrath cast under the effects of Eclipse. Since Wrath and the 4T8 proc have the same cast time, we need only show that the average Starfire is greater than or equal to the average Eclipsed Wrath. If that is true, then it would always be a benefit to consume the 4T8 proc during a Solar rotation. So here's a graph of damage calculations at varying levels of Spell Power from 2000 to 4000.

As we can see, the graph shows that Starfire gains an even bigger advantage over Wrath as we gain Spell Power. Thus, it is my recommendation to always use a 4T8 proc during a Solar Eclipse, regardless of time remaining.

Lunar Precasting Phase

In the precasting phase, we are casting Wrath. I will assume that the Idol of the Shooting Star is equipped. I will also be using a druid equipped in Graylo's choices for BiS as he calculated his BiS for a Lunar rotation as I did mine for a Solar rotation.

We only experience a problem when Starfire crits and procs a Solar Eclipse, changing our rotation from the desired Lunar Eclipse. I will be comparing the difference of a crit Starfire vs average Wrath to the DPS gain/loss of proccing the incorrect Eclipse.


Surprisingly, I did not notice a very large difference between the Solar Eclipse with the wrong idol and the Lunar Eclipse with the correct idol. I was initially bothered by this until I recalled that all of this was in a perfect Spell Queuing environment and the value of the Solar Eclipse would decrease with every Wrath cast that did not queue properly. So theoretically, my advice as per the numbers is to use the 4T8 proc whenever you can.

TL:DR

The short version of my work here was that no matter when you receive a 4T8 proc, it is appears advantageous to use it.

I wanted to push these results out first because I have the suspicion that my next bit of work might take some time to complete. I want to look into the effects of the spell queuing system and adjust the worth of the Solar Eclipse based on both latency and player reaction time.

Please do not walk away from this post saying that you should always cast 4T8 whenever it procs. This is only true for a perfect environment with zero latency and a 0ms reaction time.

As I mentioned previously, there are some skeptics when it comes to the Solar rotation. Most notably, Graylo continues to issue caveats about the Solar rotation and maintains a Lunar rotation, although a comment (yay!) on my blog explained that his preference of the Lunar rotation was out of both habit and the lack of mathematically overwhelming evidence that Solar is better. I will be taking an investigative look into the drawbacks and benefits of both rotations.

Round 1: Spell Queuing

Spell queuing presents a real problem for Solar rotations because of the low cast time on Wrath. With high enough haste and Nature's Grace buffs, it is possible to get the cast time of Wrath down below the one second hard-capped GCD. This effectively stops Wrath from casting faster than one second.

Blizzard introduced a system to help compensate for latency called Spell Queuing. This allowed you to queue a spell up for casting at the completion of your previous cast any time after 0.3 seconds remaining in the previous cast. The goal of this system was to eliminate any DPS differential based on player latency. After all, a better player shouldn't be better because s/he can afford better internet!

Now I'm sure you're thinking "But, Rel, that sounds great! I can chain cast and not worry that I have 200 ms latency!" And you're right, for Starfire. But for Wrath, we find a wrinkle in the Spell Queuing fabric. It turns out that we can queue any spell we want so long as the cast time of the original spell exceeds one second. As we all have experienced, Wrath often goes below a 1 second cast time. Thus, we are left without an effective Spell Queuing system for fast casts. What does this mean for us? Every 1 second cast of Wrath now has both reaction time AND latency added to the front of the cast time, extending the cast time of Wrath to OVER one second.

As you can imagine, this poses a serious problem for Wrath, particularly in high haste situations (heroism, BL, Hodir, Vezak, etc). There begins to accumulate a significant amount of non-casting time within the Solar rotation.

Chalk one up to the Lunar Rotation here. Starfire is the clear winner with the current Spell Queuing system because it allows for seamless chain casting. The impact on a Solar rotation could be devestating for those players with slow reaction times and/or high latency.

Round 2: Bloodlust & Heroism

Although I mentioned this earlier, I figured I would also cover it here. Casting Wrath during a BL/Heroism, even under the effects of an Eclipse, leads to major clipping and reduced DPS. However, you are using the Wrath Idol and more likely than not, you will proc Eclipse twice in the span of BL/Heroism. So while you could swap Idols, I never recommend swapping. Not only is the lost cast time an issue, but often you could forget to swap it back. It's just an extra thing to manage in an already eventful rotation.

I recommend casting Starfire under the BL/Heroism effects irregardless of your chosen Eclipse cycle. If Eclipse is up and BL just started, I fire off a few Wraths first but if I don't proc Eclipse I don't sweat it. I just Starfire all the way through BL and if things work out well, I will be able to change back to Wrath at the end and get some Eclipse uptime in. Another round to Lunar.

Round 3: RNG Monster - It's Gonna Eat Me Alive!

Allow me to start by quoting Xiera from The Moonkin Repository:

Under closer examination, though, I'm starting to think solar Eclipse is better for 3.1. The main reason is that it removes a LOT of the RNG from our rotation. For starters, we only have one roll for determining whether solar Eclipse procs (crit), versus two rolls for lunar Eclipse (crit and proc chance). Furthermore, lunar Eclipse is ONLY 1.3x damage when Mr. RNG decides to be nice to us; solar Eclipse -- 1.3x damage no matter what.

I think this does a good job of pointing out the issues I want to bring up, only I'm not nearly as polite with "Mr. RNG", my little monster friend . . .

Many of us have come to know the RNG monster quite well. Sometimes this monster is our ally, turning our normal rotation into a super-charged eclipse loaded DPS powerhouse. We've also had this monster turn it's back on us, chain casting 10 or 20 spells before a proc, turning our high DPS hopes into a bleak bottom-of-the-chart reality.

Well heres where the Solar rotation pulls slightly ahead of the Lunar rotation in my opinion. The Solar rotation is a 100% chance on Starfire crit to proc Eclipse. Then the Eclipse bonus is a straight 30% increase to damage. The only RNG involvement here is the crit percentage on Starfire. Everything else is guaranteed! This is different from a Lunar Eclipse where it is possible to see zero benefit from the buff.

On top of this (something Xiera had noticed as well) our 2T8 bonus increases Eclipse effects by 15%. This is again a guarantee for the Solar rotation and an average for the Lunar rotation. During some Lunar Eclipse procs, you could see 100% crit on your Starfire casts. Your DPS will look amazing. On others, you could see 50% crit and you'll wonder "Where's the beef?!". But on average, you'd experience your crit + 45% crit rate.

I'm actually chalking this category up to Solar. The less RNG the better because the sun isn't always shining and sometimes your best really isn't good enough. The less we leave to chance and luck the less we QQ about how disparaged our damage is.

Round 4: Gear - What to Wear, What to Wear. . .

Where as our highest Spellpower and int pieces were coming with oodles of haste in Naxx, we now begin to see gear with less haste. Since Spellpower has always been our bread and butter, we are chasing the high Spellpower pieces and in some ways are controlled by them. Allow me to elaborate.

If I had said "Hey guys, lets run Solar Eclipses with 600 haste!" I deserve to get laughed at. I'd be clipping the GCD so much that I'm bound to incur a DPS loss even if I do have a dramatically low crit for a dramatically low Nature's Grace uptime. Oh, but the tables have turned, my friend. Even Lunar BiS gear lists are hovering around that mystical 400 haste mark! Now, I'm not saying this implies that we should be running a Solar rotation. No, I'm suggesting that this gives the druid the utility and freedom to choose your rotation.

Aside from stats, we also have to look at our set bonuses. Now that we've got the latest final word on 4T8, it looks to be a sizeable DPS increase over just casting Starfire (circa 175 DPS). This set bonus produces a great deal of synergy with the Solar rotation because there are only 15 seconds in the Solar rotation (or ~13, as I showed for the effective buff) that we want to be casting Wrath and only Wrath. The rest of the time a Starfire is either okay (cooldown) or even desired (precasting). Since this insta-Starfire is going to act as a GCD in terms of cast time, we can compare it with a Wrath cast quite easily; it has the same cast time and does more damage. And the choir of Angels raise up their voices in praise of the synergy!

Let me pull out my warning stick again and say that I am not implying that the 4T8 bonus does not synergize well with a Lunar rotation. It does. In fact, it stands to do more damage inside of a Lunar cycle than inside of a Solar cycle because it is possible that some of the free casts could fall within the Eclipse buff time frame. However, a proc during the pre-casting phase could either be wasted or crit and proc the wrong Eclipse type. Both of those situations are bad and perilous to say the least.

I'm also gonna chalk this one up to Solar just because I am a huge fan of the non-RNG based mechanics. I'll take 45% more damage over a 45% increased chance to get 50% more damage any day. Remember, sometimes you take the house and sometimes the house takes you! But if you invest your money in a Fixed Rate Account (and the government doesn't collapse) you'll always know what you're walking away with!

Round 5: Fight Mechanics - The Rise of Positioning

The trend of fights in Ulduar is to force some type of movement from all players. Razorscale has adds and a boss phase. XT-002 requires you to move out of the raid to drop gravity bombs. Kologarn has frickin lazer beams for eyeballs! Freya has little plant bombs, Nature's Fury, mushrooms, Sunbeams (does anyone else hate that fight!?). Thorim had Lighting Arcs and Chain Lightning. Hodir has iciles, moonbeams, fires, storm power buffs. The list goes on and on. Being able to DPS effectively while in movement is a huge problem for us.

The shorter cast time of Wrath allows us to occasionally toss a Wrath while in transit, get a cast off before moving, or get a larger percentage of use out of that inevitable and inopportune Eclipse debuff. This one is clearly in the Solar rotations court where you would be buffing Wrath with likely both the idol and possibly the Eclipse proc. I award this round to Solar.

Round 6: Mana - Running on Fumes!

I almost don't need to say it, but the clear winner here is the Lunar cycle. Starfire beats the pants off of Wrath as far as mana efficiency is concerned. No, that's not true, because after Starfire beats the pants off Wrath, it has enough mana left to fill them with insects and drop in some fire before Wrath even has a chance of putting them back on. Yeah, it's that one sided. If you're having mana troubles, turn to Starfire. The Lunar Eclipse proc only increases the damage per mana of Starfire.

Theorycrafting - Well THAT Wasn't Supposed to Happen

As is often the case, theorycrafting happens in ideal environments that we rarely see in game. Even though the raw numbers might say "Wrath is better DPCT than Starfire" we may experience a different result in game because of any of the aforementioned reasons (namely latency). I once knew a warrior in BC that loved his parry to the exclusion of other things. Parry gear, parry gems, parry enchants (do they even make that?), parry coffee mugs, parry t-shirts, parry pajamas. I mean this fellow loved it. Why? Something about Revenge being the best threat per rage or some such. And he was right. It was. But it was such little rage and such little threat that we constantly pulled aggro from him. In the game, threat per rage did not translate into TPS. This was a case of theorycrafting going wrong.

What happened to that warrior, you might ask? Well, after showing him the light and even flashing him a few times, he still refused to abandon his parry suit (under which he wore his parry pajamas, say it, it's fun). So he parried his way right out of the guild! I like to imagine that he is off parrying in a field filled with Panzerkins and Evasion-tank rouges, talking about theorycrafting and days gone by.

TL:DR

Solar won out in the RNG world, being less dependent on chance. Solar also took home the gear round, but just barely, as I feel there is more room for error or bad judgment(both of which I am fond) in that cycle when it comes to gear and set bonuses. Movement went to Solar only because it allows you to squeeze out more casts before, in between, or after a movement, maximizing Eclipse's effective uptime.

Lunar took home the spell queuing, high haste, and mana rounds. It was pretty clear that a Lunar cycle took more advantage of spell queuing, suffered less in high haste situations (of which there are many now) and was more mana efficient. These aspects of the Lunar rotation make it very attractive in its own right.

There is no "best cycle" anymore. You should choose your cycle based on the fight, your gear, your preferred playstyle, and most importantly, whatever I tell you to do. Ok, that last bit was a joke (stand on one foot?), but the rest holds very true. You can be a successful and competitive top-end raider with either a Lunar or a Solar cycle. If you want to be pro spend more time working on good positioning, planing ahead, thinking on your feet, and minimizing non-casting time rather than choosing either Lunar or Solar rotations and mindlessly subscribing to one.

Now go, young fledglings! Boom and be merry, for I have taught you all you need to know (right now)!

So if the early bird gets the worm then I guess the researching moonkin gets the DPS worm! Maybe?

Ok, all bad analogies aside, it's time to digest this information and see if we can come up with any useful tips or thoughts. Do not take these numbers as a comparison of Solar rotations vs. Lunar rotations. Your choice of rotation should be based upon your preferred playstyle and your gear selections.

Flaws:

I ignored a few things that could have a drastic impact on a solar rotation. When a cast time begins to reach 1 sec, the spell queuing system begins to break down. This introduces some lag in cast times and you can't actually cast the spells as quickly as you theoretically can. All of this amounts to getting less wrath casts than you should be getting. Since no spell can queue until the 1 second GCD is over, when your wrath reaches a 1 second cast, your next cast suffers your latency as a casting time penalty on top of the 1 sec hard GCD.

Musings:

The Solar rotation seemed to be based upon your wrath damage. I filled with Wrath and with Starfire and Wrath dealt more damage in the cooldown phase so it looks like we should fill the solar rotation with wrath given perfect spell queuing.

As far as casting DoTs inside of a Solar Eclipse, we came to the conclusion that you should only do it if there are 10 seconds remaining on the buff. Now, I didn't take into consideration that the DoT continues to tick after the Eclipse phase ends, but I think it's misleading to talk about DPCT because technically all you are doing is offsetting your DoT casts. This means that by casting IS under the Eclipse buff, you are pushing back your next IS cast by 14 seconds. Given this shifting, I came to the conclusion that we just wanted to see a raw DPS increase under the Eclipse buff.

The Lunar rotation seemed to carry an excessively high amount of damage caused by Starfire. I can only assume this is because the RNG monster didn't get me because I played an averages game. I think the lunar rotation suffers disproportionately when you get a bad eclipse proc rate. Along the same lines of "shaky results" was the discovery of the fact that sometimes you can cast a Moonfire and suffer zero DPS loss from Starfire. I laid out some guidelines for when you should refresh and when you shouldn't, but it was pointed out to me that these guidelines were pretty useless.

The DoT reapplication is a case by case basis where we would be unable to determine if we had the proper haste percentage until after the eclipse cycle was over. Therefore, we never really know if we will hit a dry spell and have zero NG procs throughout the eclipse cycle. I, personally, am always refreshing Moonfire within my Eclipse cycle so long as I do it within the first 6 seconds. I'm sure that I've been bitten by the RNG monster by doing so, but I also feel fairly confident that it has provided me with a DPS increase.

Long story short, I feel that what I have found here supports what I do in game. I am currently waiting for a second piece of T8.5 (yay for being low on the totem pole) before switching out a great deal of gear, but I am turning out solid DPS using a Solar cycle and refreshing my DoTs within Eclipse if I have 10 seconds or greater left.

I think in the near future I will be doing a post on the good, the bad, and the ugly of a Solar Rotation versus the RNG nature of the Lunar Rotation. I know Graylo holds a Lunar rotation as superior to a Solar rotation and I have seen many great Lunar parses to back up that theory. I have also seen some great Solar parses and once my gear improves to their level, I anticipate comparing my parses to theirs.

On an unrelated note, I apologize for this post taking so long to come out with so little true content. I had my birthday this weekend and was subsequently incapable of coherent thought. You should have seen Friday's version of this post. Quite comical!

to the tune of The Gambler (thanks BBB!)
Well you gotta know when to DoT 'em
know when to nuke 'em
Eclipse is a'tickin
When do you DoT 'em?


Ok ok, sorry. I actually did sing it to the song too, but reading that is probably torture enough. But the focus of this post is to help you decide if/when to refresh your DoTs during the Eclipse. I would say that I will present empirical data and then give you a black and white view of things, but our rotations and therefore our choices are fluid, based upon procs and uptimes. Consider this as my attempt at a guideline.

Before we dive into the math and start dissecting the eclipse bonuses and whatnot, it is important to get a conceptual idea of what our DoTs do and why we would ever want to sacrifice a cast (or part of a cast) for a DoT refreshment. Because of Improved Insect Swarm, our DoTs increase our two main nukes. Thus acting as not only a damage dealing spell but also as a target-based debuff. Naturally, we would want to benefit from this debuff for as long as possible while we are casting the relevant spell. In an infinite amount of time, this boils down to DPS gained by the debuff plus DPS of the spell itself vs. the DPS of what you would have cast instead. However, in a fixed amount of time (say 1 eclipse buff) this gets a little trickier to account for damage done. I will do my best to explain why I chose to work with DoTs and timeframes the way I did while I lay out my math.

My theoretical druid is still rocking out that 3k spell power, 45% crit and 389 haste. I actually ran these calculations with drastically higher haste (~650) and none of the overall results were changed.

Lunar Eclipse and MF

Moonfire directly benefits Starfire by increasing Starfire's crit chance by 3%.
Here's a glimpse at what is happening within the Eclipse cycle with and without Moonfire on my target. Since I decided to use timelines instead of DPS to construct my DPS gains/losses, I chose to look at all of my spells in terms of damage per cast (DPC)
As I looked at my Eclipse phase, I continued to reduce the effective Eclipse time by one Wrath cast, giving me a total Eclipse time of 13.9629. This gave me 6 Starfire casts that benefit from Eclipse with a total casting time of 12.1611. Notice, this leaves 13.9629 - 12.1611 = 1.8255 seconds of spare time in the Eclipse phase. At this point I did a LOT of math, all of which will be rendered irrelevant by my next point, so let's just skip to it. Because I have over 1 GCD of spare time in my Eclipse phase (due to Starfire cast time) it is ALWAYS a benefit to cast Moonfire since I will still receive the same number of Starfire casts.

Naturally, if I change the values of haste and/or crit I change my results. So my guide, as promised, is to calculate the average effective portion of the Eclipse phase and subtract from that the average cumulative cast time of your Starfires. If this is greater than the average GCD during your Eclipse phase, ALWAYS refresh Moonfire. (With the same level of crit, haste between 445 and 725 lead to the loss of a Starfire to cast Moonfire). If the number you get when you subtract is less than the GCD, only refresh if the Moonfire DD plus the Moonfire DoT is greater than the lost Starfire (approximately 10 seconds remaining on eclipse).

Well, wasn't that a messy answer? Sorry about that, but it turns out it's entirely dependent on how much spare time is in an Eclipse phase once all possible Starfires are cast.

Solar Eclipse and IS

Insect Swarm directly benefits Wrath by increasing damage by 3%.

The Solar Eclipse model is simpler. Every Wrath cast is equal to the GCD at that time. Therefore, one IS cast takes the same span of time that one Wrath takes. Thus, to see if we should recast IS during the Eclipse phase, we need to look at the DPS of the wrath cast we lose vs. the DPS of the IS that we gain in addition to the boost it provides the subsequent Wrath casts.


Since the average Wrath will deal 6267.485 we would need that much damage from the IS tick plus the increased damage in wrath. As we can see above, IS adds 160.4392 DPS to wrath. IS itself does an additional 567.4862 DPS, but only does damage every 2 seconds.

Since we see that IS ticks every two seconds, to get a true benefit within the Eclipse phase we would need to refresh with 10 seconds left under Eclipse.

TL:DR

So it turns out that the Lunar Eclipse is much more difficult to work with in regards to DoTs. The answer of whether to refresh or not is entirely dependent on the speed with which you are casting Starfire. The quick answer is

If casting Moonfire causes you to cast one less Starfire, then only do it if there are >10 seconds left on Eclipse. If casting Moonfire does NOT cause you to cast one less Starfire, refresh Moonfire when it runs out.

The Solar cycle was easier to conceptualize because Wrath and IS have the same cast time (I view the GCD penalty as cast time for DoTs). To see a DPS increase within the Eclipse phase you should only refersh if you have 10 or more seconds left on Eclipse. However, the DoT continues after the Eclipse phase has expired, and so that also counts as a DPS increase. The overall net effect should be that it increases your DoT uptime.

Now it's time for my caveat: These increases are dependent upon the time window that we have available to us. Since Eclipse is a fixed time window, we can talk about a pure DPS increase. This is much harder to do in terms of a never-ending or even non-static timeframe.

I will be using all of the assumptions from the first post and all of the same calculations save one: The Idol of the Shooting Star will be equipped. As a result, I will be truncating my math for this post. If you're absolutely dying to see what I did, I can send you my spreadsheet where I calculated all of this out.

Precasting

Eclipse

Cooldown

DoTs

Same rules apply here. I assumed that the DoTs were cast immediately before the Precasting phase and at the conclusion of the Eclipse phase. I included the post-Eclipse DoT damage in the Cooldown portion, as that is where the time spent to cast them was alloted. Here is the damage for the DoT's cast before the Precasting phase.

Totals

TL:DR

Seriously, you didn't read that? It was so tiny in comparison. . . *sigh*. On to the results: The DPS of the Lunar rotation was 4806 DPS. I went ahead and broke down the percentage of damages. Remember we're playing the averages game here, so nothing is precise and if this isn't what your damage looks like (and you have these stats) it's ok.

We've talked about percentage damages and it is very rare to see a Starfire contribution percentage over 60%, even after adjusting for treants and starfall. Movement cuts into our ability to reach these numbers as does our good pal, RNG. And it's always possible that I made a mistake somewhere. But much like proofreading your own paper, your eyes tend to tell you what you meant, not what you wrote.

This is part one in a series on the possible Moonkin Rotations, the two different Eclipse Procs, DoT uptimes, and anything else that creeps into this abysmal place between my ears.

If you couldn't tell already, I am looking extensively into the viability of the Solar rotation (an eclipse procced by Starfire). In general, I find that the Solar rotation is more fun to play. It seems to move at a faster pace during the fights. I find that a Solar rotation is a little more forgiving if you need to move.

This post will rely heavily on Graylo's previous look at the Eclipse mechanism and his calculations of the Lunar rotation. If you haven't seen this post yet, I recommend that you head on over and take a look.

I will be using my same theoretical druid primed for a Solar rotation at 3000 spellpower, 45% crit, and 389 haste. A side note to my theoretical druid: I have reduced the crit percentage from ~50 to 45% to account for the fact that many raids do not have every crit buff, especially in the 10 man world.

There are essentially three portions of the rotation. There is the precast, where we are attempting to proc the eclipse buff. Next is the eclipse portion of the rotation. This includes all of the spells that are cast while under the eclipse buff. Finally, there is the cooldown, while we wait for the internal cooldown of the eclipse buff to expire.

Assumptions: Beneficial DoT is up while corresponding spell is cast. Idol of Steadfast Renewal is used. Precasting period starts with both DoTs up. I chose to use Starfire's Nature's Grace Uptime for all non-Wrath casts since it was smaller and presented a more conservative view.

Precasting

Eclipse

Cooldown (Wrath Filler)

DoTs

I sorta gave these a hand-waving assumption that they would always be up and this is strictly FALSE. So to semi-rectify the problem, I will be including them in the damage and adding two global cooldowns to the front of the Precasting phase. I intentionally included a refresh period for my DoT’s in the Cooldown phase. However, at some point IS fell off during the eclipse proc. I would have to choose when/if to reapply the DoT. I shall be looking into this a little closer when I talk about DoT uptime.

Total Damage/DPS

TL:DR

The Solar cycle is putting out approximately 5185.4602 DPS. This is excluding any starfall or treant damage. I am also not accounting for any BL or any other type of haste buff.

Before you run off and tell your guild that you can do x DPS, keep in mind there are MANY flaws in this calculation. For instance, I calculated average Starfire damage, knowing full well that only 1 (and possibly two) Starfire crit. I assumed that the IS DoT was applied throughout the Eclipse proc without calculating the cast time involved for that. I made assumptions about glyphs and specs that aren’t universally true (but are pretty darn close). This is theorycrafting, folks. Take it with a grain of salt because it really only gives you an idea of what the moonkin arsenal can look like. That said, the following is a breakdown of the percentage contribution of each spell.
Which is not drastically far from what I experience when I am able to "stand and deliver". I'm much happier overall with these results than with the others.

Eclipse Proccing for Lunar Eclipse

Wouldn't this significantly increase the time it takes to proc a lunar eclipse?

Theoretically yes. Although theoretically this would allow you to use higher values of haste with wrath before GCD clipping. With 600 haste, a 2 second cast with Nature's Grace would become:

As you can see, we'd still be hitting the hard cap of the GCD during Nature's Grace. This means that despite the increase in cast time, increasing the proc rate on Eclipse would probably cause Lunar rotations to be easier to proc. (Although there would be a significant increase in Solar eclipse damage at higher values of haste, allowing that to scale as well so maybe that would be a balance). So to answer your question, not really. Especially considering most Lunar Boomkins run with higher haste anyway.

Solar Rotation Impact

As mentioned above, I was worried about what would happen to the damage of a Solar eclipse rotation once the cast time got bumped up and we could fire off wraths with larger benefits from haste. For the following calculations, I will assume that my raid buffed druid has 3000 spell power, 45% crit, and 600 haste. Yes, I realize this set may not exist, but it is entirely for demonstrative purposes.

Here are some numbers that I'm gonna be using throughout to calculate this. Now on to the actual details. I am just focusing on the 30 seconds of the fight that we are either in Eclipse or on an Eclipse cooldown. Given the increased cast time, I'm going to assume that Solar rotations are now filling the cooldown time with Wrath spam. For sake of comparison, I will assume that we manage to predict when Starfire was going to crit and did not waste a cast in Eclipse on Starfire. We open by refreshing IS. Then we refresh IS and MF after the Eclipse proc expires.

Normal Wrath Eclipse Glyphed Wrath Eclipse

TL:DR

So it seems to me that the glyphed version of the solar eclipse cycle would allow for an increase of 4772.93 - 4601.47 = 171.46 DPS which is clearly a nice increase. This "extra" dps came from the glyphs ability to allow wrath to scale with haste. This allows the solar rotation to continue to scale with stats just as well as the lunar rotation does, where currently the lunar rotation begins to scale better after 400 haste when wrath casts begin to clip the GCD.

A side note, since editing: The DPS of the glyphed Wrath cycle suffered by essentially one wrath cast in the eclipse cooldown portion. I had .91 of a cast which is easy to imagine that with a good RNG (or even decent RNG) you would garner that cast instead of switching to avoid proccing the wrong eclipse.

Oh Mumurs. I just talked and mathed myself into this as a fix for just about everything excluding our movement issue! How elegant and easy of a solution! Introduce ONE minor glyph that doesn't change the DPS of Wrath and solve a large chunk of our concerns! Well played sir. You have my vote in 2012!

So with yet another tip of my horns to Murmurs, I'm gonna take a look at eclipse, the RNG, and some of the proposed changes. The original discussion can be found here.

Current Eclipse

The current eclipse mechanism is very RNG dependent. For a lunar cycle, there is a 60% chance that a critical strike Wrath will proc eclipse. Right there are two instances of RNG. Even a solar cycle, which is 100% chance on a Starfire crit, still requires an RNG dependent event.

It is very easy to appear anti-RNG when talking about Eclipse and its current mechanics. Let me be clear that I am NOT anti-RNG. I think that one of the only ways to prove skill within WoW is to show how you react to dynamic situations and RNG dependent buffs/debuffs provide an excellent way to separate the great from the good.

Now that we have that little soapbox behind us, let's continue looking at eclipse. A 30% increase in critical strike chance to Starfire spells or a 30% increase to Wrath damage is the net result of the proc. Now assuming that Starfire and Wrath were doing comparable DPS, then 30% crit ~ 30% damage (with the point going to crit if the RNG loves you or increased damage if the RNG doesn't). So what is wrong with eclipse?

Eclipse provides 15 seconds of this bonus. In a "Stand and Deliver" fashion, this is pretty solid, but it breaks down in high movement situations because the buff doesn't stop ticking down just because we aren't casting. It's not unusual to spam away and once eclipse procs be forced to move in some manner.

So gandering, yes sir, gandering at Murmurs post on the WoW forums, we have the following recommendations:

- Every Starfire you cast has a 20% cumulative chance of making your next Wrath do 30% more damage. So after SF1, the chance is 20%, but after SF5 the chance is 100%.
-Allows for a higher opportunity to benefit from Eclipse procs
-Still completely RNG based and does not promote any true form of skill

Mathematically speaking, this eclipse would be centered around finding the highest probable DPS without doing a 5xStarfire>5xWrath and repeat ad nauseum. This becomes a very standard rotation and I can see the SquawkHard showing a big starfire picture with a number inside of it to count the number of Starfires cast. My brain begins to freeze over a little bit and a small puddle of drool collects around my keyboard as I raid and continually count from 1 to 5 and occasionally refresh my DoTs.

- Every Starfire cast increases Crit chance by 1%, every Wrath cast increase Haste by 1%. Both effects stack and have a duration of 10 to 15 seconds.
-Very simplistic and removes all concept of RNG and gives a flat, predictable bonus instead
-Does not create a very interesting or skilled rotation. Only one 'off choice' spell needs to be cast every 10 to 15 seconds.

This idea turns eclipse from a proc into a ramping DPS talent. First you lay on your haste effect and then you lay on your crit effect and you cast your spell of choice (likely Starfire since you'll reap both a haste benefit and a crit benefit while wrath might do some GCD clipping) with one of the other spell every 8 or so seconds to refresh your stack. I agree that the idea is a solid and dependable solution to the perceived RNG problem but it removes all need for skill. Thus is another place where an addon like SquawkHard would rule your rotation. I find this to be vastly different than the feral druid priority that requires keeping up buffs and debuffs that all have different cooldowns/requirements. Their rotation is a balancing act and the good players are easily separated from the bad players. I think a change of this style would only create a plateau of Balance druids.

The only hope that I see for this change is making it impossible to maintain both stacks completely thereby forcing people to choose between running sub optimal stacks or picking one stack over the other.

- Change Eclipse to be based on 'charges' instead of a timed debuff allotting a specific number of casts to be effected by the proc.
-Increases the opportunity to benefit from Eclipse
-Does little to remove the major RNG factor of Eclipse, though it does lessen it -Removes Haste scaling from Eclipse
-Would still have to have some form of time limit to prevent 'chaining' Eclipse or buffing pre-fight

Believe it or not, I really like the idea because it gives the player control. Gone would be the days of a movement wrecked Eclipse proc. It also preserves the unpredictability of the moonkin rotation. I agree with the thought that there needs to be a time limit on this buff and that it should be in excess of the time needed to cast the spell. There is always the option of starting the cooldown when the buff is either used up or expires. This will encourage druids to use it immediately but not penalize them for having to succumb to fight mechanics.

- Change Wrath's portion of Eclipse to be 100% chance on Crit instead of 60%.
-Removes the larger portion of Eclipse's RNG
-Still does little to help with mobility to making use of Eclipse

I hate to be the naysayer here, but the reason that the Wrath chance is 60% is because a Wrath cast time is less than a Starfire cast time. Now, ignoring the first cast proc, this means that you would have more chances to crit with Wrath than you would with Starfire. This would tip the scales back in the direction of the Lunar cycle and you'd see every Moonkin doing exactly the same thing. Blizzard has made it clear that they don't want to encourage this type of blind dopplegangering within a class/spec.

- Change Eclipse to be two separate 'on use' abilities that share a 30 second cooldown with each other
-Removes the concept of 'spell twisting'
-Allows for the best use of Eclipse buffs

So while this provides the most "user control" it again eliminates any randomness from the druid rotation. In fact, we become uni-dimensional. Pick your spell and subsequent DoT(s). Cast said spell repeatedly. Pop eclipse when most approriate. Replace keyboard from having hit 3 or 4 so many times that they no longer work.

-Change Eclipse to work like Arcane Blast and/or Missle Barage.

Ok, this is backed by Foofy (if I'm reading that correctly). Before I do a review of this I am going to need to see a little more information on it.

- Change Eclipse to add a stackable buff to Starfire and Wrath - such as each Wrath Crit increases Starfire's chance to Crit by X%. The effects don't have a shared cooldown, thus the two are theoretically chained together
- Allows for better use of Eclipse procs
- Less RNG dependent
- Promotes spell twisting
- Has the possibility to be abused in un-intended ways (although I don't think this is a major concern)

Ok, so this was my baby and I'm pretty biased to it, but lets just pretend I'm not and that I am taking as objective a stance as I can. I suggested this adaptation because it still contains the RNG factor based on your crit. It would be 100% on crit for application and in this case, you needn't worry that wrath casts faster than starfire because you'll want to have both buffs rolling. Here's the catch: You may not be able to do that all the time. In terms of gearing, I feel that this also encourages the best balance of crit/haste because you will be using both spells copiously. I agree with the un-intended abuse as it might be possible to stack crit high enough to run both buffs all the time. I feel that doing so would cause you to sacrifice so many other stats that even with both buffs rolling, your dps would be sub-par.

In the end, I feel like we've had some good ideas thrown forth by the community and I hope that GC continues reading past the silly banter people feel the need to throw into constructive threads. If you have an idea but are afraid of GC and the megalithic WoW forums, feel free to leave a comment and I'll post it over there for you.

On an unrelated gear note, I have updated my BiS spreadsheet with some new stuff and a nifty new set of dropdowns on the BiS page to make gear changes easier!