I don't think the theorycrafting thread is the right place for this until it's had more discussion, but on the topic of totem twisting I have several things I do to make sure my personal DPS stays high.

1) Never miss an SS or Shock. I used to lose a lot of dps missing shocks or waiting for SS timers. Over the course of a fight it becomes really noticeable. Rogues and Warriors will do just fine on dps as long as WF/GoA is down.

2) Use the 5s WF refresh to drop GoA. WF refreshes at the 5s mark, if you're close to another cooldown for the 9s mark, just wait and cast it at 5s. This should not be necessary all the time, but it gives some breathing room for other totems, etc. This also gives you 10 full seconds of WF and GOA, but it's not as big of a deal now that totems are on 1s gcd. DQD (or another mod) is essential for this.

3) On the pull, start with FS(0s), WFT(1.5s), GOA(2.5s), SOE(3.5s), SS(4.5s), ES(6s). The idea here is that you always want SOE and GOA to be down when you SS.

One thing I still need to find out is how much of an impact these actions have on the other dps in my group.

1) As Enh/Resto, it is IMPOSSIBLE to not miss a shock or SS CD, even without twisting. Dueling 6s/10s cycles just don't allow it. If what you're saying is to prioritize shocking or striking OVER refreshing WFT, I agree...but your Arms warrior may not.

2) This is an interesting idea but not quite fleshed out. It may help you get in more shocks, but at a significant GCD/mana cost. Furthermore, you DON'T get 10s of GoA -- in fact, when twisting the most you'll ever get out of GoA is 9s worth, or 90% of the GoA effect. Refreshing GoA twice per 10s, you're getting 80% of the effect.

3) I guess that's one way to do it...except your totems have a 30 ft radius and FS only a 20 ft. If you're running in, you could technically drop SoE / GoA et al sooner than you can FS and still get the benefit from it. But to be honest, I doubt the overall dps gain from any kind of totem ordering is going to be very huge, we're talking about the first 5 seconds of a fight here.

At a certain point, the benefits of twisting to the shaman and his party are offset by various costs of twisting (such as slipping the WFT, shock or SS cycles). I made an attempt at one point to enumerate those costs, but quickly gave up because I didn't have good numbers for the actual benefit of WFT to various specs. My numbers for arms warriors, for example, were way low due to the increased rage gain. I thought this was a minor side effect, but my arms guy tells me it's very noticeable...in one parse, it seemed to boost him by 25% vs similar buffs and a sharpening stone.

<deleted due to earlier mention>

Anyone else have a modeller to test Dual Dragonstrike's? We have two tests, and they both disagree about the end results. Given the price to farm up and craft a second I would like to not jump the gun. It seems to me that a haste proc of 212 haste, stacking, with dual mongoose, would make this combo a monster. Particularly at the higher levels of AP and crit where haste tends to grant you a larger return.

Durigen read the posts above you. Had been discussed already.

Due to a very fortunate set of circumstances, involving Karazhan and T6 raids in place for a newly minted Shaman, I've found myself nearly hit capped with the mix of gear I've attained. Obviously I'll be scooping up any T6 level gear that is offered to me, but in lower instances, should I set loot rank to a zero hit-rating modifier? Has anyone else come across this type of gearing issue that they can remember?


Edit - 28% needed to hit makes this an error on my part. Thanks for the reminder Toots. I wonder where I got 23% in my head...

If I understand the mechanics correctly, wouldn't the EP for any hit past 23% be 0? The sim still offers an EP for hit at 23% and above. There is no use for hit over that cap is there?

I agree I have an odd set of gear that allows me to hit these numbers. Some of the pressure on hit rating will be released when I hit 375 Leatherworking, but much of the T6 gear has hit on it, so I expect to stay near the cap in the short term. I'm actively trying to acquire a DST, Shard and a better chest item in case anyone feels the need to comment on those slots.

Thanks in advance for taking the time to reply.

I'm not surprised the simulator offers EP for values above 23%, considering the hit cap is 28%.

You do have the most eclectic gearing I think I've ever seen. Congrats!

Are you just collecting hit gear for shits and giggles? Because there's no point trying to get hit capped over trying to get more AP/Crit. When we were still doing SSC/TK I had a set of gear that brought me to the hit cap, however trying this out in a raid lowered my DPS by quite a bit, just as espected. Needless to say I'd rather keep my high AP and crit over hit.

No, I am not actively chasing hit gear. I'm taking the highest EP valued items I've been able to get my hands on. Reviewing the upcoming loot, however, shows that I won't have anything to worry about. The highest I'll attain is 27% and it will end up lower as I pick up some upgrades without hit on them.

I must have gotten the 23% from the hit cap for level 70 mobs.

If I'm not mistaken, the miss rate for DW against 70 mobs is 5% base + 19% DW penalty, or 24%. I'm really curious how that number popped into your head.

Wouldn't 23% with a 1% chance to miss come out to the accepted 24%? Or am I still struggling to wrap my head around 'accepted' values? I may be mixing and matching spell casting here, I've spent two years TC'ing a warlock to flop into the melee world a few months ago.

Ok, 24% chance to miss when DW a level 70 mob, and I can reduce that to 0% chance to miss with Hit rating? I have to hit the books again!

That's exactly it. Spell casting has a 1% chance to miss that cannot be avoided. The same penalty doesn't apply to physical DPS. You can remove miss from the attack table entirely.

Against a boss, of course. I was discussing an equal level mob.

It's not much more here, 14 DPS upgrade for vanir's mh.

About speed difference, the weapon speed is factored into the windfury dmg, two weapons of the same speed get 50% base wf dmg, since they should get the same amount of procs in the long run, otherwise it's 1-wspeed/(mhs+ohs).

Unless knowing what Yo did source code wise for his modelling of two identical weapons, i would be very careful about such decisions. I did several experiments with dual syphons and never found any "unnormal" behaviour for two weapons with the same speed, neither with wf, nor with flurry.

We're talking of close to 6 dps more (which atleast should result in ~10-12 dps increase) and more stats than the dragonstrike proc could ever deliver.

Two weapons of the same speed won't have the same amount of procs -- the main hand gets a slight advantage because storm strike always evaluates the MH before the OH.

That wasn't what I was asking, though; I don't think there is a significant benefit to having two weapons of the same speed. I was referring to the fact that a 2.7s weapon will have 8% more damage from each SS and WF proc than a 2.5s weapon of the same DPS simply due to its speed, yet the number of procs of each cannot increase by 8% to counteract this. I was wondering how you figured that into your equations, since I have never seen a clean way to state it, and frankly I've been looking for one.

Yes, I am a bit wary of making the jump to dual DS, but 1) It's just money, which is easy to come by lately 2) It really would make for a bad ass around town set, even if the DPS is lacking.

A quick question about the simulator.

It shows 1 AP, 0.3159 DPS on one of my AEP value simulations. Does this mean that ~3.17 AP = 1 DPS? I'm quite confused since that seems like a silly number compared to the 14AP=1DPS saying.

The 14 AP/1 DPS ratio isn't for your total DPS, it's for your auto attack swing damage only, I believe. Taking into account specials and what not, the AP:Total DPS conversion is much more favorable for shaman.

SS is simply a haste effect for me (with yellow dmg correction, e.g. no glancing). so basically for a 2.5s weapon we get a 25% haste effect (a tad more tbh, to fix the yellow dmg stuff), SS can proc everything so haste is a fine replacement for it. For a 2.7s weapon you have the 27% haste effect.

About Windfury, you can calculate the expected time for a wf proc for each hand (weighted percentage of avgspeed/wfprocrate), let's say it's 5s for both hands, and ~9s for MH and ~10s for OH (not necessarily realistic values).

So if your wf proc does on average 3600 dmg the dps for it is 400.

The results of both are fairly exact.

List of maximum buffs. I hope I got everything. I am basing numbers off of myself.

1632 (342 str)
382 atk from Battleshout (talented)
264 atk from Imp BoM
230 atk from Shard of Contempt
160 atk from Band of the Eternal Champion (don't use it anymore but let's pretend I do)
120 atk from Flask
110 atk from Stonebreaker's Totem
97 str from Strength of Earth Totem (talented)
70 atk T6 4 piece
20 str from Scroll of Strength V
20 str from Roasted Clefthoof
19 str from Mark of the Wild
50 str from BoK with above listed buffs
___________________________________________
3380 atk

Unbuffed:
1632 atk x 30% = 490 spell dmg
490 x .08 = 39 (searing totem gains 8% of spell dmg)
39 x 115% = 45 (call of flames adding 15%)

Buffed:
3380 atk x 30% = 1014 spell dmg
1014 x .08 = 81
81 x 115% = 93

93-45 = 48 extra dmg to searing totem


I believe searing totem goes off every 2 seconds so with max buffs you will get an extra 24dps over unbuffed. It's not a whole lot, but I guess every bit helps. I am not 100% sure on when the call of flames talent is calculated either so I may be a bit off.

Remove all expertise gear and head to blasted lands. You'll be able to replicate this eventually. Additionally, you should see a hit + dodge combo on your SS. You'll notice the hit is always MH damage.

Right, other abilities like Mutilate work the same way, the MH has to land for the OH to land. I'm sure that would account for a tiny bias in SS procing a MH WF. What I'm asking about though it the assumption that when both hands land the MH is always checked first for WF. I'd like to see the testing that has drawn people to this conclusion, or it it just speculation that's been around long enough to be passed off as truth?

Well, I suppose the assumption is that WF procs are calculated on hit, and since the MH hit is calculated first, it would follow that the MH WF proc is checked before the OH when they both hit.

For that Dragonstrike discussion I tried to see where my model diverges from Yo's simulator, so I tried a few things:

1.)
Boss Armor 0, 2800 AP, 100 DPS, 2.8 Weapons, 0% Crit, 100% Hit, 100 Expertise, 0% Haste, No buffs, no Windfury, no Stormstrike (disabled evverything on the other pages):

So this should be swinging a 100 DPS Weapon every 2.8s, including glancings:

Let's assume 1000 swings, 750 will be normal, 250 will be glancing (fits with the analysis in Yo's simulator)
Glancing probability is 25%, glancing reduction is 25% too (some say it's 24% but that's not the question here)

(750*840 + 250*(840*0.75))/(1000*2.8);
it should do 281.25 DPS,
Yo's simulator shows 273 DPS instead.

I'd like to know where that difference comes from.

2.) If I switch on Dual Wielding, the DPS is correctly 1.5x the MH only value

3.) If I switch on Weapon Mastery, the DPS is correctly 1.1x the previous value

4.) If I switch on Haste, the DPS is correctly 1.Nx the previous value

5.) If I switch on Crit, the calculated Crit% is constantly 3% Crit less than entered. I wonder why.

6.) If I switch on Stormstrike, the calculated DPS is using the proper dmg formula.

7.) If i switch on Windfury, the calculated DPS is using 16% proc rate (which should be correct, since we have to disallow proc chains).

If we take dual wield windfury, the avg wf dmg we should get is:

wfmh := 2.8* 1.4 *2 * ((2800+475)/14+100);
wfoh := 2.8* 1.4 *2 * 0.5 * ((2800+475*2)/14+100);
wf := 0.5*wfmh + 0.5*wfoh

wf:= 2030

Yo's sim outputs 278 dps, 2030/278 == 7.3s for each wf proc, for a 2.8s weapon combo this means a proc rate of ~38.3%, considering that we might have a tad more mh wf than oh wf (since it's evaluating the mh first), this should be correct



It would be very nice if Yo could explain me/us why his simulator outputs less white dmg than it should do, until then I would be very careful with decisions made based on his simulator. The Anyway a proper source code review might reveal other problems.

So I'd like to know:
1.) Whitedmg too low
2.) Crit% too low

Edit: I will update this as soon as I find more noteworthy things.