The New 6.0 Moonkin Rotation

After a conspicuous lack of Balance changes in the 6.0 notes before now, the newest update shows what they’ve been working on: a complete rework of the class.


There is still an Eclipse bar, but it’s changed in two important ways:

  • Its movement is no longer affected by your casts at all; it cycles on its own between the two endpoints, making one complete loop every 30 seconds.
  • Your Arcane and Nature damage bonuses vary continuously based on the Eclipse marker. At the midpoint, you get half of your mastery bonus to either element, and at an endpoint, the full mastery bonus to one element only.

So, some things stay the same. Your Eclipse cycles back and forth regularly, and you cast Starfire half the time and Wrath half the time. You use Starsurge regularly (see below) in either half of the rotation, since it’s buffed by either bonus. You maintain DoTs, and casting them when you have a high Eclipse bonus is still beneficial (even though DoTs will no longer snapshot most effects, they will still snapshot Eclipse).

The reason cited for the change is that Moonkin was daunting to initially learn, but relatively easy to play well once you understood it (which is the reverse of what you usually want in a class). I think this is generally correct, subject to some details of Eclipse management at the highest levels of play. Eclipse is confusing to explain and understand at first, to new players to the class, and the new version hopefully won’t have this problem.

In addition to the stated problem related to the learning curve, the 5.0 Eclipse had some issues that commonly caused frustration among players. I tended to pinpoint the problem as being related to Nature’s Grace–a major DPS buff that lasted a fixed amount of time and whose uptime was determined by your ability to race between Eclipses as quickly as possible. This made the class somewhat reliant on haste to avoid being stuck outside of NG, and severely penalized the class for any time not spent advanced in the Eclipse bar, in particular, movement and AoE. Even before seeing these changes, I’d suggested to the devs that they remove Nature’s Grace in 6.0. With the new constantly-oscillating Eclipse, they are not only removing Nature’s Grace, but making it so that your choice of what to cast doesn’t even effect the uptime of an Eclipse bonus. This should make the performance of the class much more stable.


A downside that remains in the new Eclipse system is that the player has very little control over the timing of the damage highs and lows. To me, it wasted a lot of the potential of having a class where the central conceit is alternating periods of high and low DPS, if you don’t get much gameplay out of choosing when to take the highs (for example of a class that does high/low DPS periods with a very high degree of player control over the timing, look at Windwalker Monks).

This is still a potential issue in the new system–in a sense even worse, because you truly have no control over when Eclipse peaks occur (unless you Astral Communion during combat, which requires channeling and doing 0 DPS). But a few things mitigate this problem now. First, the variance in the Eclipse cycle is much smaller (see graph below). Rather than a high-low cycle, it’s more of an alternation between Nature and Arcane without any true low points, since you always benefit from at least 50% of one of the bonuses. Second, there is a new mechanic that does in fact give you more control over high-DPS periods, in the new Starsurge. While it’s not precisely the same as getting control over Eclipse (what I’ve always thought we needed more of), it’s a controllable mechanic that interacts with Eclipse in an interesting way.

Screen Shot 2014-05-24 at 4.49.24 PM


The new Starsurge is on a charge system (similar to Force of Nature), giving much more flexibility in when to use it. No more mashing it the instant Shooting Stars procs. Instead, you can store a charge or two to use when you need it. The decision about when to use it comes from the fact that it buffs your next 3 Wrath casts or 2 Starfire casts (depending on which side of the cycle you’re on), by 30%. This is, of course, more beneficial near the peak of the cycle, when those spells are doing the most damage. The difference in value is small–30% of half of your mastery bonus on the affected casts–but it will be clear when you want to use it for optimal damage (possibly the bonus could be made larger if needed). However, anytime you need a damage burst at a particular moment, you can use your Starsurges then. There’s always a possible decision of maximum total single-target damage vs. using them to support the needs of the encounter.

There is also much better gameplay in Shooting Stars than in the past, as I touched on above. Instead of responding to any SS proc right away for fear of wasting it, you can use the extra charge whenever you want as long as you don’t cap. Since you’ll probably always be planning out your next 1-2 Starsurges based on your Eclipse position and what’s happening in the encounter, a proc gives you the chance to modify your plan.

In a little more detail: since you generate an Starsurge charge every 30 seconds, and you get one Lunar peak every 30 seconds (the moment in the cycle with the highest single-target damage), that’s one place you’ll likely always use SS on a single target. What this means is that you’ll cast SS when you’re around one cast away from the Lunar peak, so that it buffs the two Starfires that hit closest to the peak. Also, regardless of Starsurge/Starfire timing, you want to Moonfire as close to the peak as possible, so snapshot the highest Eclipse buff. Fortunately, if the timing is slightly off, the damage difference will be tiny (a nice feature of the sinusoidal Eclipse behavior). With 30 seconds between Eclipse peaks, and a 32-second Moonfire duration, the goal will be to always have a max Eclipsed Moonfire on a single target.

Additional Starsurge charges from Shooting Stars procs can be used to buff Wraths at the Solar peak (the next-best time for damage), or by using 2 at the Lunar peak (one just before the peak and one just after, to buff 4 Starfires). In addition, they might also be used for AoE.


Even though Starsurge and its new buff are single-target effects only, you actually have a choice of how to spend each charge: you can spend it on Starsurge or on Starfall. Starfall will now be a true AoE spell again, and it won’t be used on a single target, because Starsurge will be a better use on a single target. The AoE system is much richer though, when you have to choose how to pool and spend charges on Starfalls at important AoE moments. This may affect your single-target rotation planning as well, if you hold off on casting Starsurge in order to build up Starfall charges in advance.

Aside from Starfalls, AoE is done using Hurricane and potentially multi-dotting (Wild Mushroom no longer does damage). A lot of math will have to be done on the details of multi-dotting, now that 1) Shooting Stars procs will give extra Starfall casts and 2) the Sunfire DoT has an AoE effect.

The lack of Eclipse control over AoE may still be a problem as it is now, especially since Starfall is only buffed by Lunar. It’s possible the idea is that, since Starfall is buffed by Lunar, while Hurricane and Sunfire are buffed by Solar, AoE damage winds up roughly even regardless of where you were in the cycle. This is something to keep an eye on as we start playing with the class. If AoE damage is roughly balanced between Nature and Arcae (i.e. around 50% of it is Starfall), this should work out about right, but that balance will vary since Starfall use won’t always be totally consistent. At any rate, the situation will be much better than it is now due to the absence of Nature’s Grace.


I’m not going to do a full analysis of talents, especially since it’s too early for power comparisons (and fully developed spreadsheets are needed for that). Suffice to say, every existing talent, ability or proposed Level 100 talent that interacted with Eclipse had to be redesigned. A few preliminary comments based only on the short descriptions:

  • Celestial Alignment: now 15 seconds of maximum Eclipse damage, and freezes your Eclipse bar for the duration. Overall similar to what it does now, but better in two ways: 1) steady-state max Eclipse damage isn’t something you get normally, so it’s a bit more of a punchy cooldown, and 2) you can use it anytime without impairing your Eclipse uptime–Eclipse will pick up where it left off when it’s over.
  • Soul of the Forest: now a passive that increases the Starsurge buff. It remains the basic passive talent choice, and should be simple to math out eventually. It enables some better burst, but presumably still not as much as Incarnation or FoN.
  • Balance of Power: causes Wrath and Starfire to extend DoTs. 4 seconds per Wrath and 6 seconds per Starfire seems like might be enough to extend DoTs indefinitely on a single target (it’s more than twice the cast time of each spell). At first glance I’m not seeing the goal of this talent–all that’s accomplished by infinitely-extended DoTs is saving the GCDs to refresh and getting a better Eclipse snapshot on Sunfire. Even if the damage benefits are there, not very exciting.
  • Euphoria makes Eclipse cycle 50% faster, reaching a Lunar peak every 20 seconds instead of every 30. I’m not entirely sure I see the point of this, since the average is still the same, and it may actually get slightly less convenient to make optimal use of Starsurge and DoTs. Maybe for this reason, there’s a DPS increase component to the talent, where the Starsurge buff adds some haste (very similar to Soul of the Forest). But the big unique part of the talent, changing the Eclipse oscillation frequency, might be a miss. It’s hard to imagine how that will help me utilize Eclipse better during an encounter.


Overall, I’m happy with where these are going. I’m looking forward to getting a full-blown model in place (which may take some time given the new, unique mechanics). I think they address most of the problems that Moonkin had with the current Eclipse of 4.0 and 5.0, while keeping the flavor of Eclipse. The new Starsurge mechanic is key–without that, I would have been very concerned about the gameplay of an Eclipse that you have almost no control over in combat. The biggest thing I’ll be focusing on once we have an alpha to play is that Starsurge (with its new counterpart Starfall) bears the burden of providing nearly all of the decisionmaking in the new system. Starsurge charges are the new class resource, and effectively the only one, with mana and Eclipse energy not functioning as resources. Making sure that decisions of how to spend them are interesting and meaningful, and have no undesired incentives, will be a critical point for feedback between now and 6.0 launch.

14 thoughts on “The New 6.0 Moonkin Rotation

  1. I always like your writeups, very thoughtful.

    In this case, I think Balance of Power is for people like me who like the theme of Balance and wouldn’t mind playing it, but hate having to always manage DoTs. For me, the idea of playing Balance with basically infinite DoTs (at least on single target) is actually really exciting. I doubt I would ever be moonkining at any competitive level, so the talent would be fine with me even if it wasn’t a straight up DPS increase. It also probably serves a goal of helping new players get used to the class without being overwhelmed. Just my thoughts on it.

  2. Very nice writeup.
    I got confused by the red square wave on your chart. Are you just saying that the current eclipse is “old” while its power is dropping? I read “old eclipse” as “previous eclipse.” Maybe “eclipse is old” would be better.
    Am I correct that the Lunar buff will be a full sine wave (going between 0% and 100%) and the Solar will also, but they are just out-of-phase, so that one peaks when the other hits bottom?
    Its worth mentioning that Moonfire/Sunfire is a single auto-switch button again. I assume the switch occurs at the 50% power point. I wonder if the Dots will be strong enough that you want to start fights just before the 50% point, so that you can get both of them on the target quickly.

    • Thanks!
      “Old Eclipse” is 5.0 Eclipse; “New Eclipse” is 6.0 Eclipse.

      Right, Lunar/Solar are both sine waves, and 180 degrees out of phase.

      Yeah, I’ll have to think about DoTs more. It seems clear enough that you’ll MF (32s duration) once at each Lunar peak, and SuF (16s duration) twice each Solar.

      So far it doesn’t seem like we’ll be able to choose where we start, but that might be worth revisiting, since that was a handy part of the old AC function.

      • I feel like your graph is off by one phase on the “old” Eclipse. Isn’t when the new Eclipse at 100 when the old reaches 100? Then when new reaches 50 (leaves old Eclipse) is when the old is at 0? Or are you accounting for something different with time in the old one?

        • I actually didn’t consider the relative phase when I made that–you’re right, that might have been a little clearer. Although the two can’t exactly be compared. In old, you Starfire from 100 at one end to 100 at the other end. In new, you Starfire from midpoint to midpoint. So the graph is sort of comparable in that t=0 for both is when you which from Wrath to Starfire.

  3. I wrote up a long-ish thread on the EU site, but nobody is reading it :)
    Anyway, here it goes:

    I share your concern over the lack of control over eclipse (even less than before…), as it kind of takes the fun out of the class. On top of that, we have a few HUGE issues with the new system: not having both dots as separate spells make us vulnerable to dispels, hurt us when we need to multidot adds and makes movement REALLY hard (you cannot keep the “buffed” dot on the target and spam the other *fire…)

    I love being a boomkin, nearly everything about the class is amazing (flight form, stealth, amazing offheals), but I always played it DESPITE its rotation. I am super happy that they are changing things — but the net effect, at this point, feels like for the worse :(

    • I’m a concerned about movement right now.

      Not as much about Eclipse–even though we have no control over Eclipse, we do over Starsurge, and that’s a significant DPS bonus. We didn’t have too much great control over Eclipse before, since you just cycled as fast as possible anyway. Now the cycling is automatic, which frees up some restrictions, and a new controllable mechanic is added that you care about.

      • True that — just looking at it from a PVP perspective. The increased vulnerability to dispels scares me.

        Also, I am not sure what to do about dots. I see that in WC you cast SF twice and MF once. As they both only last 16 seconds, if you don’t have the Balance of Power, this means you are not going for 100% uptime.As you cast MF at full lunar, you go basically 47% of the time without the MF dot, is that optimal? Similarly, with a 30 sec eclipse length and 16 sec dot duration, 9.2 secs between SF refreshes leave you 4.4 secs without SF, no? Or I misunderstand something about how these dots work?

        If I am not off, one’s choice is either to have 100% dot uptime on both dots (more dot dmg, and more procs), or ~50% uptime of a powerful dot plus 2 extra GCD worth of casting. That is about 1 wrath in a 30sec cycle — but NOT at the peak of the power, as the 2-refresh rotation refreshes close to 0 energy, while the 1-dot refresh refreshes at the peak.

        • You seem to have missed reading some of tha patchnotes. Take a look at Astral Showers ( and a pandemic-like effect.

          Astral Showers increases the duration of Moonfire by 100% (among other things), while the pandemic-like effect allow you to keep 30% of the previous duration of a periodic effect and add it to the new. As it happens, 30% of 16 sec is 4.8 sec, which as you noted would let it have 100% uptime.

          • So, if I understand correctly (and sorry for sounding so bad at this, haven’t gotten much sleep the past few weeks), boomkin dot-refresh gameplay (even in the best of situations) will come down to impeccable timing. MF is kind of trivial with 32 sec duration and 30 sec cycles, but SF needs to be refreshed right at entering solar, and then 16*1.3=20.8 seconds before exiting lunar, so 5.8 seconds before leaving solar (to guarantee the most powerful SF to last across lunar).

            Do you believe this is something that is feasible by hand? Or will we just write/download a “SunFireRefresh” addon that does the thinking for us?

          • Well, as I understood it, you want a 130% duration solar that would end just as you enter solar again, to get maximum benefit for as long as possible, then you refresh it upon entering, and again at the right time.

            Granted the optimal time is probably slightly shifted to be more centered around solar. I’m sure we will be enlightened on what the math say when things solidify.

            Greatly looking forward to play my moonkin again. It was my main for many years!

          • You guys are generally right about how Sunfire will work. I discussed this more in a followup I just made today:

            I wouldn’t say it comes down to impeccable timing, as the DPS loss for deviating slightly will be small. If you delay both, you’ll get a slightly better bonus on the short Sunfire, and a very slightly worse bonus on the long Sunfire–the net loss is tiny. That said, as I mentioned in my post today, it’s unideal and be generally irritating to always that that timing sensitivity (even if it’s mostly psychological). Moonfire has a similar issue but was fixed by adding the flat portion at the top of each Eclipse peak. I’d like to come up with a similar solution we could suggest for Sunfire.

  4. Pingback: The 6.0 Moonkin, Second Look | It's Dangerous to Go Alone

Leave a Reply

Your email address will not be published. Required fields are marked *