That's not 100% uptime then. From our kill I got 22,5k mana from VT but that's with 10% w/o a spriest, 7k MoE and 5,1K from JoW.

This is probably a complex question to answer, but is the Serpent Coil Braid worth using over Flame Caps? The stealth buff the SCB got with the mana gem change makes me wish I didn't pass on it back then (I had Crusade/TLC and was arcane). Now with the Skull not stacking with HSH, is this the right way to go? I will say mana gems are alot easier to farm than flame caps...

Manly, do you think the 2300ish dps Ishaxa put out on your Brutallus kill is the upper limit for mages on that encounter with current gear levels? With best-in-slot Sunwell/BT/MH/ZA gear do you think 2500 is attainable with a reasonable group comp?

After scrolling through some top rdps parses I found Ishaxa's to be somewhat irregular, especially in crit rate. I also found your crit rate to be quite low.

Well, if you check closely the log you see a few details that helped the numbers. If anything, I didnt get anything out of the norm, basically only 1 lust, spriest, elemental shaman, and pretty much did all the scorching. My DPS is lower than it should have been, given I had to cast 3 drums (3 lost GCD) and had to do a semi-evocation (2-3s of lost dps time). Otherwise my DPS is pretty much what I expect any mage to pull off on it. And yes, my crit rate was low.

Lots of mages get 2 bloodlust on most parses I have seen, yet do less damage. I am not sure how. Problem is, next week I don't wear the same gear as I did this week, so the more it progresses the more the numbers will vary.

I think 2500 DPS is doable without any sunwell gear. Give me 2 bloodlust and its a done deal. Or give me a moonkin (why in gods name would you bring a moonkin to brutallus is beyind me, but ignoring this detail) and I expect numbers to be fairly close to that. Or possibly more than yourself as the only drummer in the group. I don't believe the figure is any kind of unrealistic. Although watch out next week since its going to be darkmoon faire, and I expect lots of people posting 2700+ dps parses thanks to the cheese buff.

I managed 2413 DPS with two Heroisms on a 1% wipe last night. My crit, at 48%, was close to the expected value (Chain of the Twilight Owl + ele shaman + ret paladin), and I was scorching, so I don't really understand how I managed to get that high (I didn't go past 2250 on any other attempts). Maybe I got lucky and landed solid crits with cooldowns on the few Nightfall procs we got. That's with not-quite-optimized gear (Archimonde never dropped a single Tempest since our first kill on October 2nd, my OH and wand ar not up-to-par either). If I had that gear and iCoE, I would probably have gotten that 2500.

WWS - Yes, we suck.

That parse shows 36% crit rate, 2 bloodlust, and 2153 dps. I guess you were using recount to get that number ?

(if that is the case, yes, recount and wws dont count the same way. given your 57% scorch crit rate, I guess recount gave you an average crit rate for all your casts, which gave you 48% somehow.)

Well, getting freak crit rates can give absurd numbers. Just ask Koosha to link you that teron parse with 2700 dps and see what I mean.

In any case, the problem of comparing parses has always been the compounding effect of variable raid buffs, and RNG. Hitting your crits during your cooldowns will help a lot. I don't think anyone will ever agree to what is 'common' raid/group buffs. For some people draenai aura is always assumed. For some other its elemental shaman. Some always assume JOW. Some assume moonkin. I don't even get COE consistently. I don't believe the group crit necklace changes numbers a whole lot, nor does drums, unless you have a bunch of drummers in the group. Or the 4 nightfall procs. In any case, I think the double lust is really what makes the big difference.

Edit: to further add to my point, a post-enrage kill/wipe will naturally boost firespec numbers, due to long time spent on execute range.

Given that the second bloodlust is not in mf range, it's giving approximately 1.3*40s / 360s = 3.33% dps increase(napkin math aside). That's not much more than an elemental shaman totem. Most guilds don't use elemental shaman.

First, the 2nd bloodlust stacks with other cooldowns. Second, if you kill him post enrage with people dying left and right, you can actually possibly fit almost 2 bloodlust on execute range.

edit: and not to sound snarky, but even 3% is definately non-trivial.

Not so sure why you are disliking moonkins, they add much more to the raid than an elemental shaman I would say. We have always had a moonkin since day one in TBC raiding and I've never regretted it, 3% hit for all melee is wonderful and I can use the extra 2% crit over 3% hit, especially considering how we get 1% hit from our shamans.

Wow Web Stats The kill. I think the least problem with this fight is actually dps, or it was for us, biggest issue for us was burn healing and the wonderful stomp+mh+oh hits that just annihilated our tanks.

Edit: I think it'll be possible to do quite close to 2500 dps, our top mage in the kill didn't use destruction potions as he had run out and got lucky with crit either way, add destruction potions and less moonkins dying and he'd have been 2500 easily especially if I was scorching.

For Alliance:
Elemental Shaman: 3% crit, 4% hit, 101 dmg, ~50 mp5 and Heroism
Moonkin: 5% crit

How can you possibly say the Moonkin is better?

Because there's also the added 3% hit to the entire melee portion of the raid, along with utility in the form of Innervate and Rebirth.

Not saying one or the other is better, but there's more to it than just 5% crit, unless you're only talking from a singular PoV for your damage, not raid damage.

I'd still rather have the Elemental Shaman and Resto shaman in the raid despite the +hit to melee, rather than a Moonkin and a Resto Shaman. This allows another group to have the Resto shaman and still gives a second Heroism to the raid. It seems to get more out of less, but I suppose opinions will vary. I'll concede arguing it because initially I was simply looking at a per group basis.

edit: damn it, people keep editing.

Deleted

Double post

The point is ele shaman single lust vs resto shaman with double lust will produce similar parses on a 6 minute kill given that you can use Two caps and three destro pots either way. A mage with a moonkin in his group would see larger benefit than a mage without a moonkin but two bloodlusts.

Also, like i said it was just napkin math to answer your questioning of double lust parses. To be completely accurate I would probably just compare them in a spreadsheet.

Edit: Of course the flip side is the mages in said parses just aren't reducing cast latency as well. It seems this is the largest factor in mage dps along with being poor at cooldown management.

You're assuming that the elemental shaman gets removed from the raid in favor of a resto shaman and the moonkin is taking a free spot. This is not possible unless you were previously raiding with 24 people in which case adding any class would be a huge benefit... When you bring the moonkin to the raid *something* gotta get out of the raid and I really doubt there's a class worth taking out - and even if there is one there is probably a better class to bring in than a moonkin (yet *another* shaman or a mage/lock/hunter for example).

On a side note 101 damage alone is not much under the moonkin aura and is much better than the 2% difference between moonkin aura and the shaman totem. Not to mention shadow priest would benefit noticeably more from the spell damage than the crit (which is a must-have for either elemental shaman or moonkin as far as I know) and innervate is probably far from making up for those differences especially counting the extra 50 mp5 for the party.


As for using WWS parses, even casting something like 200 fireballs (and you won't even reach close to that in a real fight) will still produce a noticeable variance in DPS just due to crits without even taking other (although also less noticeable) random factors into account, with 40% crit you should expect a standard deviation of ~3.5% in your crit, or when dealing 2200 DPS around 50-60 DPS difference between equally played parses. With just 100 fireballs it goes up to ~5% or ~75 DPS difference. Of course the actual expected variance will be a bit higher if you take other random factors into account like # of procs and wether your crits happened during cooldowns/procs or not.

These factors obviously have no effect on your average theorycrafted DPS but they do have an effect on how uneffective your WWS measurements as a tool to backup your theory are.

http://elitistjerks.com/658230-post3191.html
I should put that into my signature or so.

It's strong point is that it stacks as trinket. It's weak point is that it relies on using mana gems which may not the optimal thing in your situation at hand.

Did Illidan with 3 mages. Swapped staff to lower my hit on Flame adds (Fire/IV), had another fire mage and a frost mage (less than 13% hit, more than 10%, as far as I know).
Total resists of frost attacks on Flames were 0.7%. Total resis rate of my frostbolts was 0.9%.
Not a proof, but at least a decent indication that ghost hit still seems to be in.

Wasn't sure which thread would be the best for this question since it has some overlap but figured this thread would have the most Rawr users.

For the 40/0/21 spec rawr some gear yields higher DPS when socketing with +10 int, +10 spi and/or +5int/2mp5 than +12 dmg or dmg/crit combos. Based on the TC that more mana/regen gives rise to more AB spam time and hence higher dps. I understand the TC but was curious what people have found in game using this gemming scheme in raid fights.

Thank you for the link. Very comprehensive explanation of how you trade damage for mana.

Don't forget that one of the biggest RNG factor which has a great influence on your DPS is your luck with partials resists.

On a previous WWS posted, like this one : Wow Web Stats
the difference between the 2 mages is huge :
1.7% for Rinald
4.6% for Vhad

Considering that the real DPS loss is not correctly logged, and a medium factor of 1.6 to get the true partial resistance value, the real DPS loss for both are :
2319/(100-1.7*1.6)*100 - 2319 ~= 65dps
2186/(100-4.6*1.6)*100 - 2186 ~= 174dps
We get a number as high as a 109 dps difference only due to the luck factor on partial resists.

Note : In case information about the partial resist being incorrectly logged to show the real DPS loss in a raid environnement are not well known, i would be glad to post a topic about it as i made some research concerning this fact.

Arcane Missiles can't proc Timbal's. It's a channeled spell that casts bolts every second. That's fundamentally different from Mind Flay, which is a channeled spell that sustains a DoT. As a rule of thumb, if damage can crit, it's a bolt (direct damage). If it can't crit, it's a DoT (periodic damage). And Arcane Missiles can crit.