-50% proc chance on relic (possibly higher)
--Procs so much it's annoying. It's the same animation as the finishing of a healing spell
--Might be bugged? Seems stupidly high proc
-Moonfire crits are 200%
--Do not proc NG
--Do not regen mana from Moonkin Form
Here's a quick update with the new stuff. 2T9 and 4T9 as listed. Idol lazily implemented as a constant 200 crit; if the proc rate's really that high it blows away all previous Idols. 2T9 raises the value of crit compared to haste, raises the value of Glyphs of SF/MF (irrelevant unless more DPS Glyphs appear), and raises the value of ImpMF to make it quite worthwhile.
Improved moonfire should effect the periodic crits that you get with 2pc T9 but I would still recommend testing it, just in case it was coded such that the crit portion of the talent only effects the initial cast
Make sure you're thinking through what you're doing before you spend a lot of time tinkering with the spreadsheet. I'd assumed you were talking about Vezax hardmode if you were going to all the trouble to think about this fight. If, as you say, you're trying maximize the total damage out of your mana bar, then all that matters is DPM. Cast time, clipping, and any other timing issues are irrelevant. It doesn't matter that Solar "lasts longer"; you can make anything last longer by standing around doing nothing for a bit. It's completely obvious, for example, that you wouldn't cast Moonfire (except on Clearcasts), you don't need to model the rotation to find this out.
Except that the results are different than the expectations; I posted the modified version of the spreadsheet because there seems to be a lot of misinformation going around (I was surprised myself, so I thought I'd share rather than keep it to myself). Moonfire inside of Shadow Crash is 84.63 DPM, which is 55% less mana efficient than the 153.55 DPM of just Starfire. Your expectations believe that because its so mana inefficient, it shouldn't be cast at all or only during an OoC proc, but, the fact is, despite the inefficiencies of the spell, it actually increases the DPM and damage done of the rotation and should be cast regardless of an OoC proc. My original attempts with a calculator made me want to get a better and more concise answer, and it was these inconsistencies of what is expected or should be obvious that made me seek the truth for myself. In the end, DPM of the spell didn't really matter and the DPM of the rotations on the original and the modified spreadsheets showed the same ranking by DPM, but just knowing that much wouldn't have satisfied me.
The only nontrivial issue is Solar vs. Lunar. Eclipsed Starfires give significantly better DPM than Eclipsed Wraths, but uneclipsed Starfires also have better DPM than uneclipsed Wraths. But the issue is the frequent interruption to dodge crashes making the very swingy Lunar rotation impractical. Interruptions due to movement can detract from Eclipse time but not from pre-Eclipse time (this is an irritation with our class on every fight, actually), so frequent movement works heavily against a rotation that requires you to invest in a full Eclipse to get a minor overall benefit.
During Animus, where you might want to trade a bit of efficiency for DPS, I'm not sure anything changes--Solar is more consistent that Lunar in terms of both DPS and DPM. Also, your guild will certainly have Bloodlust up, during which you'd want to just spam Starfire.
I agree that proccing Solar is more consistent (and, actually, I prefer Solar to Lunar), but, once either is procced, its probably likely you won't get a full Eclipse cycle out of either due to moving for Shadow Crash. Even with my meager amount of haste (269 rating), my NG-assisted Starfires clip the GCD just like Wrath which likely means the number of Starfires or Wraths that can or will be casted within Eclipse will be about the same. Bloodlust/Heroism will make almost no difference to us, so the normal best rotation would still apply even during Bloodlust/Heroism. (I didn't notice until now that Starfire was getting GCD clipped as well until I tried changing the "Total haste" field again and saw Starfire's cast time not change due to the 1sec+Latency+Reaction Time calculations).
I'm still going to modify the more recent spreadsheet to at least verify the results I got from the 090519 version; once I'm done, I'll make it available to download (it could be even that all of my reults will change with the new version! ; ). Oh, and my apologies if I misunderstood what you were trying to say; I'm not entirely clear on the point you were trying to make with Lunar vs. Solar, so I tried to respond as best I could.
The problem is 2T8, when you get into solar eclipse you should skip the instant-sf, because a wrath will actually deal more damage. Also you could say to use the proc under solar if it would extend your moonfire, but to get into solar you had to cast starfire, it was ~10dps gain in my simulations.
Current 4T9 seems to be a placeholder, wonder what they come up with, now that with 3.2.0 the 2T8 bonus becomes just uber-awesome
Light the fuse.
For all my homies.
Do not run, we are your friends.
SimulationCraft Druid Guy
Are there any thoughts on spell twisting SF/W when we're not under either Eclipse ICD? Granted that it's a circumstance that only occurs once at the very beginning of each fight, and perhaps after extended movement, but as such a simple change, if it can make even a small dps improvement, it'd be nice to know.
I haven't the knack for modeling it myself, but it seems like an extra GCD casting the correct Eclipsed spell, stacked against the possibility of getting whichever Eclipse proc you don't like once more than the other, if the fight duration and RNG end up that way. I suppose there might be additional benefit in that even a braindead player will be guaranteed to pop their 4T8 procs quickly when twisting, and to extend the initial MF at least once.
An instant SF is better DPET than an Eclipsed Wrath.
Now that you point it out >:C
But 3.2.0 is giving me headaches with BiS gear list for SimulationCraft, as you don't only have to get near 263, but with the change to eclipse you get a new value to watch out where your one scale factors suddenly loses its value: 400 haste.
That's when you reach 1.0-gcd with NG, CF and 5% ToW.
I had one profile, hitcap, 373haste etc. and as plenty of new hardmode items are out, I started bothering about a profile update. So I did a scale factor run in SC and got:
That basically boils down to stat values and socket bonuses. Pt for pt spell power will almost always outweigh haste or crit, however if you're getting a 7-9 spell power bonus from the sockets then yes it'll be worth using a spell power and haste or crit gem depending where you're at.
I'm still under the impression that haste will still outweigh crit until you reach the 400 haste mark.