Well, the "Twink" is lvl 70 of course. It's only my Definition that my Paladin is my "Mainchar".

To your other points:
Hmm, if I would do that...this would be a real fun for my collagues and just to be sure about that: when i melee, i get around 85-100 mana per proc, which does on roughly every second hits. Presuming i have an 1.5 speed dagger, this would mean 20 hits for 1k mana in 30 sec - without dodge/miss/parry. You can't be serious to show the mages an 30 second meleeing OOMkin to regain poor 1k mana. ^^

Thats wrong, Nature's Grace procs when the spell succeeds no matter if the mob dies between wrathbomb leaving your hands and hitting the mob. So you instantly know whether you crit or not BEFORE the wrath hits. Starfire on the opposite is hitting the same splitsecond your cast is successful.
Also you don't have to wait for a 2 in a row proc, that'd mean you already lost 0.5 seconds.
Scenario:
• You cast Wrath (1.5 sec). Normal hit. Cast time = Global Cooldown.
• You cast Wrath (1.5 sec). Critical hit. Cast time = GCD. Nature's Grace procs.
• You cast Wrath (1.0 sec). Normal hit. Cast time = GCD -0.5 sec, Delay till next spellcast 0.5 sec.

Nature's Grace is incompatible with Wrath-spamming as it is now, while Starfire reduced by a previous crit to 2.5 sec gets full benefit.

The perfect dps cycle would be wrath/moonfire on beginning and starfire for the rest of the fight. Wrath is nice because of its "flying time" as an opener, but with specialized +arc dmg gear its soon outdamaged (dps AND dpm) by moonfire.
In Raid setups of course you HAVE TO support the raid with Insect Swarm and eventually Imp. FF.

If you're forced to spend 30 seconds doing absolutely no damage to regain 1k mana, I'm pretty sure that the DPS loss from that is enough to warrant a moonkin being labeled unusable as DPS for that fight.

You need to use a feral weapon to get decent returns on meleeing for mana, since the mana returned is based on a percentage of your attack power (30% I recall reading somewhere, but I'm not sure how accurate that is). The proc rate is somewhere around 75% (again im not sure how accurate that is). I know when I tested it back before TBC, meleeing for mana with the full teir 0.5 set and the aq20 set (lots of hybrid stats, and extra 140AP in moonkin) I was able to jack the proc up to over 300 mana per proc. I don't think gear exists at 70 to recreate that, but switching in a feral weapon certainly boosts the regen.

The best meleekin weapon is The End of Dreams, with a 1.9 attack speed and 305 AP (an extra 90 mana per proc). Sadly there is no equivalent weapon at 70, and it is not easy to obtain.

Looking at the remaing feral weapons (sorted by speed):

http://www.wowhead.com/?items=2&filt...1;crv=0#00kzc:

The End of Dreams - 1.9 speed, 305 AP (an extra 91.5 mana per proc)
Hammer of Bestial Fury - 1.9 speed, 180 AP (an extra 54 mana per proc) matches TEoD on speed, but less AP, but more realistic to get (probablt can 5 man the dragons)
Blessed Qiraji War Hammer - 2.1 speed, 300 AP (an extra 90 mana per proc), slower than TeOD with same regen, same problem of obtainability
Draenic Wildstaff - 2.4 speed, 491 AP (an extra 147 mana per proc)
Dreamer's Dragonstaff - 2.4 speed, 515 AP (an extra 154 mana per proc)
Greatstaff of the Leviathan - 2.4 speed, 491 AP (an extra 147 mana per proc) Also has 360 armor which might be helpful to survive in melee range.
Mace of Unending Life - 2.6 speed, 158 AP (an extra 47 mana per proc)
Atiesh, Greatstaff of the Guardian - 2.9 speed, 420 AP (an extra 126 mana per proc)
Staff of Natural Fury - 3.0 speed, 585 AP (an extra 175.5 mana per proc) This also has a -200 to shapeshifting costs, which I assume affects moonkin. Has 320 Armor.
Feral Staff of Lashing - 3.0 speed, 595 AP (an extra 178.5 mana per proc) Has 300 Armor.
Terestian's Stranglestaff - 3.0 speed, 624 AP (an extra 187 mana per proc)
Wildfury Greatstaff - 3.0 speed, 654 AP (an extra 196 mana per proc) Has 630 Armor.
Gladiator's Maul - 3.6 speed, 718 AP (an extra 215 mana per proc)

Also it is boosted of course by warrior's battle shout, trueshot aura, and windfury. I found this out meleeing the worms before Aran, the mana procs on a windfury are pretty huge.

edit:

I forgot to mention the Savagery enchant, which adds an extra 21 mana per proc.

Just as a Reference: Damage coefficients

Starfire - 100% (3.5/3.5)
Wrath - 57,1% (2/3.5)
Moonfire - 14,9% / 52,2% (34,8% * 1.5/3.5 - 65,2% * 12/15)
Entangling Roots - 90,0% (27/15 * 0.5 - immobilizing)
Insect swarm - 76% (12/15 * 0.95)
Hurricane - 107% (10/3.5 * 1/2 * 0.9 * 0.83 - nonsensical derivation)

Without considering CoS, Moonfire becomes worse than IS around +650 dmg by itself and costs over twice as much mana. Even considering the benefits of Crit, MF just isn't worth casting in a serious raid setting unless you can guarantee your mana will last the duration of the fight. And then use it in addition to IS, not as a replacement, IS really is that good.

The ability for moonkin to melee for mana is a pathetic joke given to us by Blizzard because of Panzerkin that PvP'd before BC claiming that Str and high damage weapons (Hand of Rag) were valuable to moonkin because of OOC and Judgement of Wisdom. None of this was supported by math of course, nor will it ever be.

I played around with a feral set in moonkin form. I had around 1600-2000 AP (I can't remember the exact value off-hand) and was regaining 556 mana per proc.

This mechanic, while unique and interesting, is pretty terrible for a raid situation, especially given the fact that we're often threat limited. Running into melee range and going from a 130% threat ceiling down to a 110% threat ceiling seems like a good way to die. Also, even with the feral gear mana regen rate, the mana/5 was slow enough to significantly hurt overall DPS, as was mentioned.

I apologize, I didn't capture any numbers, although I would liken it to drinking Morning Glory Dew.

If the autattack swing wasn't interrupted by casting it'd be interesting (to say the least) as a longevity/damage tool for raids, but without that the benefit only really shows up in extremely long studies where you swarm/moonfire/whack/whack/whack instead of drinking. In those cases, it's better for your total mobs killed per hour to nuke to zero and hit stuff, than to nuke to zero and drink. I think the "mellee for mana" mechanic is simply a soloing/leveling perk and not really intended as a raid/group thing.

My experiences in beta/live with balance suggests that a large portion of the problem with the class is threat (see subtlety change, although I could wish subtlety was tier1/2 resto instead of 3 given the 44/0/17 dilema) and poor internal interactions (moonfire and nature's grace stuff primarily.) Ignite showed us how to make "moonfire" dd then dot type spells work, moonfire's current implementation penalizes crit/hit needlessly by comparison. Meh, it feels a revision or 3 from being interesting to me, particularly without a "niche" that it fills. Shadowpriests became interesting (to me as a standard raid leader) when they started providing something more meaningful/unique than "more damage". Competitive damage is required before you're considered, but when considering what would "help here" it's the utility stuff that comes up. While the moonkin damage bonus to a raid may or may not be competitive, it would take something unique to the moonkin to make me as a raid leader say "you know, we should get a moonkin".

I think that often gets overlooked, enough so that I"ll put it here again.

While damage determines if you are eligible for a raid invite, utility determines if you are desired, and by extension invited.

Two part puzzle, probably requires two part solution.

First off, great thread- I do love these forums- and some great math by Bekah that really helps nail down issues.

It really does look like Starfire might be the best spammable spell, which really does hurt quite a bit- Wrath has that oh-so-juicy casttime modifier which helps, but if it's simply not sustainable...

Speaking of sustainable, what do the numbers start looking like with 1-2 downranks of Wrath? The base damage on the spell doesn't look compelling enough that dropping it a rank or two will really hurt, and it looks like it would vastly increase efficency and sustainability.

While imp FF wasn't proving too much benifit (why doesn't this hit spells again? Baffling to me), the threat/damage increase for tanks/melee specificly is very hard to deny from the base ability- and 3% extra hit does not precisely hurt.

A moderately elegant solution to the above poster's comments might be that your spell criticals restore mana to your party in (smallish) amounts, a SP-lite system that fits the theme set by iLOTP and provides some of the SP benifit without encroaching heavily. In addition, because SPs do nothing for crit nor benifit strongly from it (and rarely seem to have serious mana issues with minimal consumable usage), there would be little incentive to stack both in one group, allowing Balance to help other groups with lesser mana concerns but more crit issues (ele shaman, affliction locks with imp SB, perhaps arcane/fire?). Since oomkin aura does not stack placing them together would also provide minimal benifit, preventing raid stacking further.

Of course, I'm sure I'm not the first person to come up with that idea and there's probably some fundemental flaw I'm not seeing, but it seems like it might give 'kins a potential place at the table.

Woah, I come back and have to read so much.

Well, I did skim alot but from what I read, I think an important factor was missed out when it comes to Wrath or Starfire spamming. I'm from Australia and latency plays a major issue here ever since Fastcast was no more.

I did try to keep up the IS, SFx3, Wrath rotation but I always found out my IS finished 1-2 seconds earlier than my cycle and according to my UI, I was chain casting effectively. Thus I had to somehow modify my dps cycle to accomodate for the .3 to .4 latency.

WRATH vs STARFIRE spam - One thing people have to remember is that in practice with latency wrath actually becomes 1.5 sec + 0.3 sec cast whereas Starfire becomes 3.0 sec + 0.3 sec cast, not including NG procs. I'll leave you to decide which is better.

As for the mana/threat issue, I'm still fairly convinced if grouped with a shadow priest, my primary concern will be threat. That Gruul encounter left me with alot of mana (~40% and sorry if I don't have a screenshot, guess you just gotta believe me) and I was not flasked/potted or anything. Perhaps when my guild becomes more relaxed in SSC/TK, I'll try Moonkin on a much longer fight like Karatherass or even Void Reaver.

One thing I'd like to state is that, for my situation (perhaps not all situations), my problem was not my DPS cycle. If anything my DPS cycle, was enough to stay on par with agro with the MT so I was not lacking in that department. What was the problem was threat and the ability NOT to push out more DPS.

IS is a very godly spell, 80% damage coefficient and at a extremely high DPM. I'd use it in almost every encounter kinda like on my lock moving with dots and static with SB's. IS and perhaps MF while moving out of Void Reaver lightning bolts and SF while static.

As I stated above, this thread was not meant to prove the viability of moonkins in raids (read end of first post). However, since you did bring it up as inevitable as it was, I'll shed some light on it from my point of view.

Main issue with raid viability in a raiding guild comes down to 30 people online, and you're gonna run SSC. You get the basic class make up (tanks for Hydross, Healers, standard awesome DPS'ers and CC). Then you're left with 6 people who happen to be a Moonkin, Mage, Warlock, etc. whatever.

Now what determines whether or not you actually are desired is what you can bring to a raid over someone else.
People dying can be prevented (Battle res), People running out of mana can be prevented (Innervate). People losing DPS can be helped via consumeables (Crit Aura / Hit Buff). I'm saying what Moonkins bring to a raid can otherwise be prevented. In other words, Moonkins bring nothing but treatment rather than preventive measures.

What CANNOT be prevented are things like Infernals running all over you, Murlocs coming at you every minute, 4 Spawns spawning every Hydross switch, etc. This is where mages, locks, classes that can provide AOE and CC can bring preventive measures. You cannot expect to do Tidewalker without AoE classes and the more the better.

I completely understand this and I'm not pushing for moonkin viability, I'll let others do that. To me atm, the best comparison I can give a moonkin to a "pure DPS class" is to a rogue somewhat and that's being generous. Rogues provide much needed interrupts on alot of cases. The similarity lies in raid slotting, and atm with the not-melee friendly encounters, rogues do get shafted alot more than any other class despite our rogue being able to do incredible DPS when left in. This will all change in 2.1, and who knows 2.1 might still give moonkins more utility.

There's regen time on Gruul during the tosses and reverbs, you're not in the 5sr the whole time and thus you're getting in quite a few spirit ticks to help regen- which makes it a little more friendly for mana limited DD classes- ESPECIALLY if you're grouped with a shadow priest who's dots are ticking (and generating mana for you) even if the priest is silenced or tossed and not actively casting.

The problem is that your dps goes down as well because the bulk of your dps is DD. Just a quirk of the fight. If you have fairly high latency, looking into a /stopcasting macro can really help close timed spell rotations- again, Latency and You is a great thread as are the threads about haste effects- although those are fairly scattered.

The important thing about theorycrafting out arguments for acceptance in raids is proving capacity in a pure setting (impossible to replicate in game), then applying that to real situations and arguing your advantages/disadvantages in real world settings. If you want to compete with a mage for a raid slot, you have to be able to discuss your advantages and disadvantages in the same language- on completely equal ground- so you have a real basis for direct comparison.

The next logical step after modeling "spam" is to model a basic ideal rotation and apply variables to those calcs. This is usually where a good spreadsheet comes into play that allows you to change the +dmg/crit/int/spi gear variables against time variables for the fight- all being churned through the modifiers.

Once you've got that kind of spreadsheet (it's MUCH easier for moonkins than for mages because you don't have to model 5-10 different spec variations in addition) you can present the findings compared to other dps classes with similar gearing and THAT'S when you start arguing the "worth" of all the non-dps stuff you bring to the fight... and proving that the spreadsheet is reliable by real world data examples. =)

That's true only if the druid is actually present. Stacking up on gear for whether or not the druid is there is probably not a desired course of action.

As long as the moonkin can pull his own weight as dps it should be no problem to find a place for him in the raid, even when min/maxing. If you don't need the healing, but could do with more dps with the added bonus of utility, a moonkin would be viable addition.