So I did some heroics. I had 2 x [Plans: Fel Edged Battleaxe]'s and I now got a [Boggspine Knuckles]. I now got [Boggspine Knuckles] in my OH and my [Plans: Fel Edged Battleaxe] in my MH.

From what I read in this thread, I should be better off with my new combo?

So, until I can replace the Malchazeen with the Fool's Bane or something similar, would you say it would be better to put, say... Rockbiter on the offhand?

No. Put Windfury on both of them, and then read the first post in this thread a bit more thoroughly where all of this is explained.

Rockbiter is physicakl damage, so it is also affected by the 50% malus. It also generates a lot of aggro so it is always a bad choice. Dual windfury is always the best choice whatever you are using as weapons.

Rockbiter hasn't been generating extra threat since the pre expansion TBC talent patch...

But other then that, forget there are such things as other weapon enchants, use windfury, end of story.

I have a question. What would be the best combination?
Crystalweave Cape - Items - World of Warcraft + Thundering skyfire Diamond + Dragonspine trophy + Abacus of the violent odds

or

Blood Knight War Cloak - Items - World of Warcraft + Relentless Earthstorm Diamond - Items - World of Warcraft + Bloodlust Brooch - Items - World of Warcraft + Hourglass of the Unraveller - Items - World of Warcraft ?

Hey, amazingly Lootzor and/or Pawn, or hand calculating the worth of those items based on the itemization weights in the first post would tell you everything you wanted to know about those combinations!

Perhaps if you enlarge the font size of the already-perfectly visible and readable red text to say 6 or 7, people might spot it easier and actually bother to read the OP before asking questions.

I feel your pain, Malan.

Noted and implemented.

Just wondering about the -enemyarmor stat AEP calculations. Sorry about any spelling errors or obvious mistakes, my point with the reply is the theory if it would work or not, I'm going home tomorrow after two weeks of 12 hour days + overtime, so bear with me please.

x armor gives y% mitigation, right?

So, could you run a simulation of a mob with 75% mitigation, 50%, 25% and last 0% (perhaps more steps then that too),
then implement the DPS increase graph into this: Image:Armor.JPG - WoWWiki, the Warcraft wiki

Shouldnt in theory the DPS graph be inverse proporsional with the armor graph, and isnt there then some way to figure out the DPS increase in relation to armor from that, thus figuring out how much -1 armor increases our DPS (or how much 1 armor decreases our DPS I dont really know how to mathecraftican it), and then add that value to the x that's on the items, especially when we know that the armor cap (I assume for players) is that value (provided it's correct).

I'm fearing it wont work if perhaps mobs have different armor values and cap% then players, but would it be possible to figure out the value of armor penetration versus players like this? I dont know, just a tired after-16-hours-at-work-thought as I haven't seen many suggestions to how to model this stat.

Thanks for the read.

I spoke to quite a few people about the -Armor effect and the conclusion is that while we could indeed chart some values and say that "for this value X, -Armor is worth Z AEP", but the problem is that it would be worth different AEP for every mob/boss you run into. Because of the varying amounts of armor that bosses have its impossible to assign it a strict conversion value.

Hm. So more testing required but we could possibly value it between 1.5-2.0 depending on accuracy.

That first post is beautiful, Malan. I haven't looked at the first page of this thread since it was a mere child at 12 pages long or so. I'm very impressed with what you've done with it.

Just a public "thank you"

Echoed. I don't even have anything to do with Enhancement Shamans, I'm just a curious observer, and you've done an amazing job of collecting, organizing, and presenting information.

GG, sir.

Due to varying mob armor, it will ofcourse be very difficult to determine a set AEP for the stat, how about a graph saying with x armor, -armor is worth y AEP? Then we could determine at what armor value of the boss it would be preferable to use the -armor items instead of the 'oldskool' items.

Personally I used MobMitigation to check up on mob\boss armor, but I'll try the other addon suggested and see the results, roughly I think I came up with the same as Morelis with 2.5-5k armor.

In conclusion it's definetly worth having the stat for fights where the mob\boss has alot of armor, while in the typical cloth-mob fights you should go for the 'old' stats?

Back on the topic of twisting...

(I am aware this uses the old non-nerfed version of WF Totem)

I just used the rogue dps spreadsheet (incredibly nice, btw) from the thread in the class mechanics forum, and inputted almost exact values of one of the sword rogues in my guild. Here's his armory:

The Armory

According to the spreadsheet, the following values are true:

Just poisons: 1466
Just GoA + poisons: 1561
Just WF/Poison: 1594
Twisting WF/GoA: 1633
(Assume WF 100%, GoA 70% (to account for lag, imperfect redrops))

In other words, with pretty darn good, but still imperfect, twisting, your combat sword rogues should gain about 40 dps from twisting in the old system. Once the spreadsheet is updated, I'll try this again. Worst case I'd assume 1/2 of the WF dps goes away.

One thing confuses me though - I'd heard that with the WF "fix", WF would still be a better totem than GoA. I might have used the sheet wrong, because that is not at all consistent with the results I gave above (in which I assume not having specials proc WF would be more than a 30% drop in WF damage, which is what it would take to equal the GoA + poisons buff).

Just something to chew on, or for someone more rogue-knowledgeable to investigate.

I guess the big question is, assuming you're in a 5 melee group, and conservatively guess that each person gets about 30dps from twisting, is it worth it in the loss of either global cooldowns for the shaman, and/or in the loss of shock dps? How much utility is 150 raid dps worth...?

The flat dps gains across the party seems to be worth it considering spammed shocks are about 130 dps (off the top of my head, I calculated it earlier somewhere). I tried it myself recently, and though it was a bit of a pain, I usually managed a rotation of Stormstrike -> WF totem (rank 1) -> GoA -> Shock, though I didn't have perfect buff uptime, it seemed to work out alright. I essentially lost 50% to 30% of my Shocks across a fight.

The main, personal benefit was that I had less threat due to shocks not getting the benefit of Spirit Weapons. This was pretty good, especially on trash. I'm usually in a hybrid tank/melee group, so that meant the tank got a bit of extra dodge, crit and armour too. On Hydross I twisted with Nature Resist Totem, which worked out well because I can't use every shock cooldown anyway.

I didn't use it on fights where I had to interrupt after I missed ESing some heals due to GCDs, but other than that it was okay.

Once the talent changes go in, it might not be worth doing anymore. Rough figures are 700 spellpower (~2300 AP), which means the new DPS for Earth Shock and Flame Shock being hit every cooldown, including standard debuffs are:

ES: (((.42 * 700)+675)/12)*1.25 = 100 dps
FS: (((.67 * 700)+797)/12)*1.3 = 137 dps

And with the new talents, I'll try switching into an Elemental minor with 10% boosted damage and 5 second cooldown on shocks.

ES: ((.42 * 700)+675)/10)*1.25*1.1 = 133 dps
FS: (((.67 * 700)+797)/12)*1.3*1.1 = 150 dps (I've kept FS cooldown to 12 seconds for dot ticking)

For a total of 288 dps. Granted, there will be a few dropped shocks due to conflicts with the SS timer, but it seems to be a reasonable estimate. You can probably still fit twisting in between shock cooldowns. Along with the lowered mana costs of shocks, watching WFtotem pulses and WF cooldown timers, we might actually have something close to interactivity DPS-wise.

Edit: Amusingly enough, with the +8% Lightning Shield PvP gloves, the +15% Lightning Shield talent, 700 spellpower, and the Stormstrike debuff, each Lightning Shield charge will go off for ~733 damage.

Nemaa mentioned some testing back in post 910. He had better results main handing Tide over two 30-minute segments of combat. Just did two one-hour segments myself

Gear was:
1532 AP
11.89% Haste (125)
6.53% Hit (103)
28.32 % Crit

Other buff stuff: RED meta gem, Tsunami Talisman, Ashtongue trinket, and 2x mongoose on weaps

About the runs: I hit the Servant of Raz till Recount's displayed dps stopped moving more than +/- 1 over the course of about 20 secs. Hit stormstrike as soon as it was off cooldown. Each test had three heal breaks lasting about 10 secs. MH Tide test was about two minutes longer. The SoE totem from the main hand Tide run was the result of a misclick, it was down for two seconds so it's neglibile.

Tide MH:

Syphon MH:

Discussion: MHing Syphon came out on top by a scant 11 dps. Of note - my trinkets activated more frequently in the Tide MH run (as evinced by the higher average white hits and crits from that run and the number of buffs applies in the buff gains table). Even with fewer ap procs to go on, however, the average wf hits and crits were higher while main handing Syphon. This is significant because it implies that the higher dmg range on Syphon more than makes up for the increased % of off hand windfuries. However, I am not sure what % of the windfuries in each run were performed by which hand, so this isn't a reasonable conclusion to draw.

Sadly, even though it seemed like I'd reached stable dps levels in both runs, the crit rates for wf and ss between the runs don't match, meaning they weren't long enough to reach the the mean crit rate on those abilities.

Conclusion: Not really satisfied drawing a conclusion and will perform longer runs later on. For now I'll be main handing Syphon. I'm curious to see if others have performed their own tests and what their results are.

P.S. If anyone could point me to an app/mod that parses combat logs and shows the % and # of MH vs OH wf procs that'd be swell for drilling deeper. Thanks

Interesting results Brum. I've been wondering about this ever since I ran the sim test and it showed MH being slow was more important than OH being slow (to a degree). Since we've sort of migrated over the past couple months to thinking that the most important thing is for the OH to be slower than the MH, I figured my sim had just missed something.

From your tests, it's looking more likely that the "MH needs to be slow" directive still may be the case. Thanks much for the data.

As for the parsing app/mod, during my tests I always just did it 'by hand' in excel, based on the damage of each attack. It's a real pain, to tell the truth, but if you really want the data, you can get it that way.

I got an idear while reading the discussion about the new WF vs GoA. You had a debate up there about FT vs WF on our Offhand and FT might be as good up to a point. But what about the FT-Totem will it scale with Spelldamge too? If thats the case it could be better to put down FT and GoA instead of WF.

My guild currently faces a shortage of shamans.
I have to admit that reading threads like this one made me eager to discover the class, and I started levelling a shaman.

I think I pretty much understood the logic and mechanics of the class described here, but I am still puzzled about itemization, especially about MH/OH speeds.
The quote above, as well as the original post, both go in the "an OH faster than your MH will hurt your DPS", and suggest "for the OH to be slower than the MH" so that most WF procs occur on the MH.
If I am correct, the Runic Hammer and Arena mace (best pre-BT recommended OH) both are faster than the Blacksmithing mace.

Does it mean that, considering that the Blacksmithing mace is the best (easiest to obtain) MH pre-BT, the current itemization just does not allow any OH slower than your MH until the Black Temple ?
This leads me to wonder if the speed actually matters, or if it is just the difference between them that is the most important ?
I have been trying to use the different tools and simulators described here to verify how a 2.7/2.4 combination would compare to a, for instance, 2.0/2.1 combination. But I sure could use some input from experienced theorycrafting shamans, who are most likely to have tested this already.

I would go as far to say that dragonstrike is _THE_ best weapon available for a shaman. With syphon in OH.

On the note of the syphon I was expecting the proc to be alot better, I'm really disapointed that instead of some uesful stats We get the proc.

The weapon proc accounted for 17784 out of 4 million damage and 12371 out of 3.5million damage in the last 2 raids. Less than 0.5% of my damage.

I was trying to think of a way to improve it, I thought maybe having it proc on every melee hit, but maybe that would be overpowered?