Answered here

Note that haste obviously is calculated at start of cast too.

Shadow Priest simply drop Vampiric Embrace if they're threat capped, especially in the fights you mentioned. Bloodboil is very threat sensitive, and healing Bloodboil is the least of your worries in that fight. It's predictable damage, unlike the rest of his abilities. It is however, an exceptional fight. That one occasionally has healers running OOM, something which is quite rare in TBC.

You thinking stuff doesn't necessarily make it true. If you want to convince anyone here, show us numbers.

Using the tools mentioned in the compendium, us warlock sport a roughly measured 6% damage from WoA, at least 3% from crit depending on how you model ISB, and 3% from hit bonus. Add Draenei +1% if you're alliance. As far as I know, mages benefit less from each of these, correct me if I'm wrong. Now add heroism, which is something mages tend to benefit less from too: good mages run OOM at the moment the boss dies, so getting more spells in benefits them less than warlocks.

Personal rant: I think play skill and latency will be much bigger factors in determining raid dps, drowning out any measurable difference of putting warlocks or mages with resto/elemental shamans. You probably just want your best players in that group. In our raids, I routinely see top dps people at 150% of the average dpser (regardless of class/spec). It might be gear, latency, people being clueless or just slacking. But given this, I find debating 2-3% extra personal dps just rather hypothetical at this time.

Aim for max shadow resist (365) with shadow protection buff up. That was enough for our guild. We didn't use SL specced lock even during the training of the encounter.

Is there a difference between modifiers (=Spellpower / multipliers) on the caster (i.e. trinkets / procs) and modifiers on target (i.e. ISB-Debuff / scorch)??? I mean the moment of "locking" / calculating damage.


Greets Ormus

Actually I'm not trying to convince anybody. I'm looking for some hard numbers myself to min/max raid-setups. We have Illidan on farm for several months now so I have Sunwell in mind where every bit of Raid-DPS could matter. We've invitet just recently an elemental shaman. There will be debates in regard to group setups. We are a caster heavy raid and normally have 3 wls (1 affli-supporter 2 destruction)+ 3 mages (all fire) in our setup with 1 SP to go around for our DD's.

Being the Raid-Leader, I'm trying to decide those matters based on facts and not gut-feeling. That's why I asked.

No.

All stuff is the same regardless of whether they're caster buffs (Shadow Mastery, Demonic Sacrifice) or target debuffs (ISB, CoS).

Generally totally right. If a shaman would boost mage and warlock in percentage the same, more raiddps will be done from the stronger class.

But in reality it depends on the encounter. For Bloodboil e.g. I see no use for me to have either a shadow priest or an elemental shaman since I always scratch the border of overaggro.

However for Illidan a warlock don't have that much aggro problems if you don't start right away with a crit series.

Next factor is the duration of an encounter and in what state you are. For learning a new encounter healers which don't run oom are more useful than dps without mana.

Conclusion: there is no recipe for the best setup as long as you have so many points to bear in mind

I can't profess to know whether shaman totems benefit a mage more than a lock. I'd hazard a guess that you gain more benefit from both crit and spelldamage, given you have a much more friendly coefficient in the form of SnF, Maledicted CoS and ISB. Even though our crit modifier is *2.1 compared to your *2. Hit rating benefits equally I'd think too, though given you output more DPS is more technically valuable to you. It's also more conventional for a mage to hit hitcap before a warlock does in gear terms.

On your other two points, I simply must stop you however. You display lacking knowledge of the mage class: Firstly, no, a good mage will not run OOM at point of boss death. You are perhaps getting confused with 2.2 Arcane spec. When this was possible, it was theorized that the optimum would be to *burn* all your mana on the very costly but very high DPS spell Arcane Blast. Technically, a perfect 2.2 Arcane mage would end with 0 mana but under no circumstances should be OOM before the end.

Either way, since 2.3 Fire is generally considered to be the most raid-effective spec and Arcane has been broken by having it's item-dependency incapacitated. A Firemage should not have any trouble maintaining enough mana throughout an encounter, even without a SP. Given SP however, they can switch to Destro Pots to maxmize Cooldown Synergy (ie. Icy Veins in 2.3.2, Combustion, Trinket, Flamecap, Destropot all together).

And as for Bloodlust/Heroism, undoubtably a mage's biggest weapon is Molten Fury which increases their output by 20% when the boss is under 20% health. This makes Bloodlust (and afforementioned cooldown synergy) at 19% boss health a hugely powerful combination for Firemages, arguably better use to them than Destrolocks who have no such fluctuation in output.

Finaly on your point of "getting more spells in benefits them less than warlocks." I don't understand what you're trying to say... A small percentage of your damage comes from CoA/CoD but by and large you're almost entirely dependent on SB Spam to make your numbers, just like we're dependent on Fireball/Scorch spam for 100% of our output. How are you justifying "more spells don't benefit" a mage more than you?

This is new to me. My post was made assuming that mages had ways dumping their mana into more dps, which was the case with 2.2 arcane. If that is no longer the case, and fire is the default pve spec now, I was wrong. Thanks for correcting me.

It seems that this makes it even harder to decide what is better. I'll stick with my previous statement: put your best players with the elemental shaman, be they warlocks or mages. This is assuming they're not threat capped.

Did they fix ignite yet so fire crit modifier is a real 210% now? It used to be 150% + 60% ignite dot that got overwritten by subsequent crits.

You can see people argue for and against arcane day in, day out.
As far as I know, <The Flaming Ruby> is raiding with a couple of very convinced arcane mages, so it's hard to argue for or against it.

* Fire specs just scale vastly better than arcane.
Arcane needs 2/5 T5 and a huge amount of excessive mana to spam AB. Fire scales with all the +hit and +haste on T6 level gear and with 4/5 T6.
Come 2.3.2 and Icy Veins (20% haste for 20sec, 3min CD), fire will be more mana hungry, but 1 shadow priest is about all the mana a mage needs*. There is no real option to burn mana for more damage.
Having more mana means that damage is increases by not using Evocation (cast time lost), using molten armour over mage armour, using Flame Caps instead of mana gems, and Destruction Potions instead of SMPs. Check the mage thread for numbers, it's roughly 0.4-0.5 spell damage equivalent gained per 1mp5 from outside.

* Ignite has been "real 210% crits" for about a year, since 2.0 came and rolling ignites were removed. It can bug on occasion when one player has two crits at the same time, but it has next to no impact on raiding.

* Heroism at 20% on mages is probably the most you can get from a heroism in a caster group. Mages have two 3 minute cooldown talents and usually trinket/flame cap to save when the boss gets to 20% and takes 20% from them.
If you really min-max though, you're probably better off rotating your shaman through the melee group.

We agree on a One Roll system right ?

So the win of 1% hit is only +1% dps without any crit. With a hight crit modifier and a high crit chance the win of 1% hit is alot less. With ruin and 33.33..% crit your personal damage is 50% from crits, so increasing the non crit damage by 1% will only lead to 0.5% overall dps gain. This is why crit is not so bad vs. hit.

Can someone confirm this?

Do not forget about Tranquil Air totem. If you would be threat-capped, that's a good DPS increase. For what it's worth, I'm usually with an Elemental Shaman and a Shadow Priest for Bloodboil and Tranquil Air allows me to make good use of all the other group DPS increases. TA allows for a Soul Shatter around 50% to be all I need on Bloodboil to cast undeterred by threat.

Edit: TA during the parts where threat actually matters anyway. A good Elemental Shaman knows when to switch between Wrath of Air and Tranquil Air for a high damage caster group.

ok so ive been reading about the arguements for and against 30/21/10, and mostly i see "has less crit" and "does not have ruin" so now given this and after making an adjustment has anyone ever tried 19/21/21

basically since most of the highest end gear still promotes casting corruption i tried to keep things like imp corruption, nightfall, and as much empowered corruption as i could, while still having shadowbolt be fairly strong, you only lose out on 3% crit and the 20% more +dmg to shadowbolt from shadow and flame. not saying that this is a great build, as it is mainly just me tinkering, but im mainly looking for input and opinions and if anyone has ever tried this out.

I'm willing to argue against rotating Bloodlust through melee. Any melee class never has more than 60% of their damage from white attacks and as such can never gain as much as any caster which has 100% damage from casts (Edit: which can be hasted, of course. So AB spam and Aff locks are out). Rogues may have a dozen cooldowns, but more than half their damage is from yellows. Shamans will hit the WF CD cap and gain a maximum of 0.3 WF procs per second, yellow attacks also are CD limited. Fury warriors? Arguably of the melee they gain the most, as the increased attack rate will cause an increase in WF-totem proc over time, as well as more crit-per-second making more Rage. Rage of course, translates into more yellow attacks.

Even so, unless we're talking a seriously melee-heavy fight, or some kind of melee gimmick (Twin Azinoth vs. Demon spring to mind, though I understand that has a nerf inc at last) I'd definitely not think it reasonable to rotate Bloodlust through melee.

What exactly are you trying to achieve with that spec?

well on the destro side at least having devastation and ruin since everyone seems to complain about the crit and lack of ruin, getting at least DS on the demonology side to sac the succ, and still having a decent corruption that can proc nightfall on the affliction side since a lot of the later gear procs from corruption, ie the 2 piece tier six or the ashtongue trinket, or benefits from having it up ie the tier 5 4 piece.

yeah i know you lose siphon life and shadow mastery but people seemed to think the crit and such was more important so i swapped it out

just as 0/21/40 it seems like you would never cast corruption at all, given that it is no longer instant cast

Your 19/21/21 build is bad, the 30/21/10 is better. First off, 2/5 Tier 6 will overheal nearly all the time, no matter what spells you cast (note it is a great soloing bonus) and the Ashtongue trinket is one of the worst BT trinkets mostly due to the short buff uptime (again, it isn't so bad if you are soloing via DoTing a lot of mobs).

While criting more often is a nice goal, you want to help kill the boss with a decent dps spec first.

yeah thats what i was thinking but everyone was clamoring about the lost crit, but yeah im one to think that 30/21/10 isnt such a bad build since it will double percentage buff just about everything you are casting, and as you get the benefit of more % buffs the benefit becomes bigger. with a lock casting CoS with malediction, full shadow weaving, misery and a ISB debuff up your shadow damage % goes to 198% of normal whereas without the shadow mastery its just, 180% with everything i mentioned up

13% buffed CoS (1.13)
10% shadow weaving (1.1)
5% misery (1.05)
20% ISB (1.2)
15% sacced succ (1.15)
10% shadow mastery (1.1)

1.13*1.1*1.05*1.2*1.15= 1.80 without SM
1.13*1.1*1.05*1.2*1.15*1.1 = 1.98 with SM

this is also assuming one gets calculated after another and that it doesnt calculate as pre percentage buffed damage for each percentage buff then adds them all together and adds that to your damage, but im pretty sure it does what my calculations are doing and are actually multiplicitive. but please tell me if my math is wrong or whatnot cause im really just interested in making sure i calculate this all right

If you are comparing the crappy 30/21/10 to the better 21/40, you forget the 20% more spell damage to Shadow bolt, which is a lot more than 10% damage (13-18% is head math, it scales with gear), and the first bad build uses corruption, so has a lower SB uptime.

well i was mainly listing the buffs that affect everything shadow and affect the entire spell rather than granting bonus of +spell dmg, also other than the lower crit and global cooldown incured by casting it i guess what about corruption makes the uptime of ISB less? its not like it eats charges.

+spell dmg is just a DPS increase as +% dmg. The only difference is that one depends on how much spell dmg you have and the other on how much damage you do, but both are DPS increases and you cannot just ignore one. Basically I don't see what all the arguments are about. You have a spreadsheet - use it. It even lets you take just about any possible assumption you could think might tip the scales into account.

I think this is a critical part of the equation. Warlocks in mid BT/Hyjal can assemble a 12% hit set to take full advantage of the totem. Even in endgame gear, theres some hypothetical loot sets floating around here that have 12% hit in full endgame gear. I'll try to find them.

Sure, mages can dump the 3 talents in the frost tree if you are guaranteed an elemental shaman. But where would you put them?