Enhance Shaman: The Collected Works of Theorycraft, Vol I - Page 110

Malan · 09/19/07, 9:37 PM

Originally Posted by Yakout

@Malan

The OP's list of top 5 MHs has "Wicked Edge of the Plains" instead of "Wicked Edge of the Planes". Not at all a big deal for anyone with two wits to rub together, but it does break the item link.

I blame Rob, I think I copied/pasted that from him.

Yo! · 09/19/07, 10:47 PM

Originally Posted by Morelis

What ppm/uptime are you using to model these procs? The sim is giving me results that are not consistent with our theorycrafting up to this point.

Dragonspine - 1.5 ppm
Ashtongue 75% per hand on SS.
Madness - 2.4 ppm
Hourglass - 0.9 ppm, crits only
Tsunami - 0,9 ppm, crits only
T3.5 - 1% of any hit
Please be more specific - what exactly looks odd to you?
Did you "add their static bonuses manually"?

Rob · 09/20/07, 1:43 AM

Originally Posted by Malan

I blame Rob, I think I copied/pasted that from him.

Maybe, but I corrected it in the TTT

Anyway, perhaps this will end the weapon selection discussion...

For non-orcs
You will want to choose maces or fist weapons since current itemization gives bonus skill to these but not axes. The first 5 weapon skill is worth 3% hit and 0.2% crit as well as an unknown amount of anti-parry and anti-dodge. Consider picking up a [Belt of One-Hundred Deaths] to gain this 5 weapon skill; it's better than the best BT belts because of it.

If you have access to [Syphon of the Nathrezim], dual wield it. (Put your first one in your OH if you have an MH-only weapon like Dragonstrike, otherwise it doesn't matter which hand you choose.)
If you don't have access to a Syphon, mainhand a [Dragonstrike] and offhand a [Merciless Gladiator's Pummeler].
If you don't have access to a Dragonstrike, you can mainhand a [Dragonmaw] and offhand an MG Pummeler, or dual-wield MG Pummelers.
Everything beneath this should be clear.

For orcs
You will want to choose axes since your racial gives you 3% hit and 0.2% crit as well as an unknown amount of anti-parry and anti-dodge if you equip an axe. (You will thus have the luxury of being able to select the best belt available to you, ignoring the weapon skill on Bo100D.)

If you have access to [Rising Tide], use it. (Put it in your OH if you have an MH-only weapon like Wicked Edge of the Planes, otherwise MH it.)
If you don't have access to a Rising Tide, mainhand a [Wicked Edge of the Planes] and offhand a [Merciless Gladiator's Cleaver] or dual-wield MG Cleavers.
If you only have one or even none of the above, [Black Planar Edge] is the next-best MH choice. (OH a MG Cleaver if you can.) A [Dragonstrike] or MG Pummeler would also be acceptable substitutes for the Black Planar Edge.
Everything beneath this should be clear.

Assumptions/Rationale:
Dragonstrike/Dragonmaw uptime is ~25% or less*. With the proc equal to 12% haste rating in 2.2, this is at most a 3% increase to white damage -- i.e. equivalent to 3% hit rating. It is thus not worth selecting a Dragonstrike over a Wicked Edge of the Planes for an orc; they gain the 3% hit from weapon skill and the larger statistics on the axe.

No test comparing wielding Syphon/Rising Tide combos has ever shown >3% increase to white damage, and many have yielded conflicting results, so we should defer to theory here, which says that slow weapons will tend to be better ceteris paribus.

Everything else should follow fairly clearly from this.

* This may be an inaccurate assumption; more to come on this point.

Morelis · 09/20/07, 1:53 AM

Originally Posted by Yo!

Dragonspine - 1.5 ppm
Ashtongue 75% per hand on SS.
Madness - 2.4 ppm
Hourglass - 0.9 ppm, crits only
Tsunami - 0,9 ppm, crits only
T3.5 - 1% of any hit
Please be more specific - what exactly looks odd to you?
Did you "add their static bonuses manually"?

I measured the performance of the trinkets in your sim using the stats from my armory profile and my usual raid buffs, I added and subtracted static bonuses as necessary. I then ran the sim again putting in the static bonuses of each trinket as well as the approximate passive contribution of each proc based on observation and the results were a bit off. For haste stuff I assumed 2.2 values as well as reduced uptime on DST.

DST:
1203 - Sim
1151 - Passive 5.14% haste (325 haste rating with 25% uptime)

MotB
1152 - Sim
1143 - Passive -120 armor (300 with 40% uptime)

Ashtongue
1175 - Sim
1160 - Passive 220 AP (275 with 80% uptime)

I don't have any empirical data for uptime on the hourglass and TT but using 0.9 ppm returned values that were about 25dps lower than the value from using the sim option for the trinkets.

The results seem to consistently over-value the trinkets especially the DST, is there something else that might be accounting for it? I also wonder if the same thing might be happening with the weapon enchant options.

Boratz · 09/20/07, 2:29 AM

@Rob

Rising Tide is unique, thus can't dual wield.

Otherwise that's a top breakdown analysis.

Morelis · 09/20/07, 2:38 AM

Originally Posted by Rob

Dragonstrike/Dragonmaw uptime is ~25% or less. With the proc equal to 12% haste rating in 2.2, this is at most a 3% increase to white damage -- i.e. equivalent to 3% hit rating. It is thus not worth selecting a Dragonstrike over a Wicked Edge of the Planes for an orc; they gain the 3% hit from weapon skill and the larger statistics on the axe.

For what it's worth, after I noticed the lowered uptime on DST on the PTR I tested Dragonstrike right away. I was relieved to find that it hasn't had any kind of internal cooldown added and the proc rate appears to be unchanged. I've gotten uptimes from 35-40% whenever I've tested it which would be 4.5-5% haste.

everwatch · 09/20/07, 3:27 AM

Begging a few answers on the topic of GoA vs. WF totem again. And please, listen to my explantion.

Before the WF totem nerf I ran WF totem exclusively and was spec'd into it. After the nerf, my rogues (3) and fury warrior (1) wanted to test some things out. The main reason why we decided to test this was because we had one rogue who regularly put poisons on his MH weapon even if WF was out. Regardless, we ran several runs with WF and then several runs with GoA with poisons/sharpening stones. The melee players in my guild unanimously all told me that they wanted me to run GoA while they used sharpening stones/poisons. I was spec'd into both of the totems at this time, and this occured after the WF Totem nerf. In addition, keep in mind that at several occasions I will swap out the current air totem for Tranquil Air. This is more to benefit our Fury warrior who has severe threat issues. And I'm not a big fan of totem twisting myself. I tried it. I've decided to not use that tactic.

Now on to the question.

Are my boys just seeing what they want? Or are they right? Is it possible their results are specific to our guild/group makeup with gear/specs/playstyles? I did not track their numbers myself. I simply told them on the specific days what I was running for them, and to then tell me which one netted them more dps in the long run. Is someone planning on doing extensive testing on the new results of using GoA with poisons over WF? Their standpoint is that GoA is superior, at least for them. I saw mention of Disquette organizing some tests on PTR, but did not see any details.

I listed below some of the WWS reports from Boss Kills so you can look at the rogues/warrior if you like: Hashassasin, Veneno, Polymer, and Darip. These are using GoA for them. I don't have any with WF totem because we only recently started using WWS within the last few weeks. I wish we had used it sooner so I could compare some fights.

The Officers of my guild typically always run 1 druid in our party even if all 3 rogues and the 1 warrior are there. This is an extreme rarity btw. Usually one of the 4 is not in the raid. So standard makeup is either:

1 Feral Druid, 1 Fury Warrior, 1 Enhance Shaman, 2 Rogues
1 Feral Druid, 1 Enhance Shaman, 3 Rogues

The party members, and my guild officers have made a few decisions (totem use and party makeup) that drastically difffers from some commenly held "truths". But the issue that I come to is I find their arguements and their positions solidly argued and made. I've read every single page of this thread. However considering my guild mates found results that differ from the commonly held truths, I was asking for opinions/reasons on why they may be right or wrong. Thanks again.

This thread taught me just about everything I know about Shaman. You have no idea how much I appreciate the hard work all of you have put in to make the information available to other Shaman like myself.

Anetheron Kill
Rage Winterchill Kill
Teron Gorefiend Kill
High Warlord Naj'entus Kill

P.S. Before you rag me about spec'ing into Imp Ghost Wolf, I'm the Totem/Kite/Puller for Trash Mob waves in BT. We're not 100% to speed on every wave, so sometimes I'm hauling ass to make it out past the NPC's and to drop totems, and get all of the Banshees and Shadowy Necromancers casting on me and my totems while I run like hell so no one gets alpha striked. For the role my guild has me perform there, I find it amazingly useful. The 1 second cast changed the situations from me dying some of the time, to never dying.

Rob · 09/20/07, 3:38 AM

@ Boratz -- thanks, I will edit to reflect this.

@ Morelis -- Hmm, other posts I had seen said 25% uptime (perhaps the proc rate changes between incarnations of the weapon and I was looking at the wrong one).

Does anyone else have uptime figures on Dragonstrike proc?

GamingManiac · 09/20/07, 4:15 AM

Just picked up Belt of a Thousand Deaths off Vashj thing today (was using Belt of Deep Shadow). However, I have several issues. Right now I'm MHing Big Bad Wolf's Paw MH, Merciless Gladiator Axe (bought it months ago) OH, while Rising Tide/Syp are in the near future (missed them both dropping when I would have had prio since I was gone the nights they've dropped).

First off, the belt only benefits my main hand. My rogue friend (who theorycrafts pretty heavily) is pretty sure that if the MH is benefitting from the belt and the MH alone, it's still the best belt in the game (ahead Boneweave Girdle). If it was OH and not MH, it wouldn't. Is he correct in one or both assumptions? I know that the belt is conclusively the best one if you can get the appropriate weapons (as has been mentioned many times over), but my question's relevant to if I only have one weapon slot that's of the right type.

Secondly, the same question as above, but if I was running with Rising Tide/Syp. I apologize if these specific questions have been answered before but I'm positive they haven't, and I'd run simulators, but I have a Mac and no apple ID to download Java 8 for Yo!'s sim. Thank you very much.

Diogo · 09/20/07, 4:17 AM

Originally Posted by everwatch

Begging a few answers on the topic of GoA vs. WF totem again. And please, listen to my explantion.

Before the WF totem nerf I ran WF totem exclusively and was spec'd into it. After the nerf, my rogues (3) and fury warrior (1) wanted to test some things out. The main reason why we decided to test this was because we had one rogue who regularly put poisons on his MH weapon even if WF was out. Regardless, we ran several runs with WF and then several runs with GoA with poisons/sharpening stones. The melee players in my guild unanimously all told me that they wanted me to run GoA while they used sharpening stones/poisons. I was spec'd into both of the totems at this time, and this occured after the WF Totem nerf. In addition, keep in mind that at several occasions I will swap out the current air totem for Tranquil Air. This is more to benefit our Fury warrior who has severe threat issues. And I'm not a big fan of totem twisting myself. I tried it. I've decided to not use that tactic.

Now on to the question.

Are my boys just seeing what they want? Or are they right? Is it possible their results are specific to our guild/group makeup with gear/specs/playstyles? I did not track their numbers myself. I simply told them on the specific days what I was running for them, and to then tell me which one netted them more dps in the long run. Is someone planning on doing extensive testing on the new results of using GoA with poisons over WF? Their standpoint is that GoA is superior, at least for them. I saw mention of Disquette organizing some tests on PTR, but did not see any details.

I listed below some of the WWS reports from Boss Kills so you can look at the rogues/warrior if you like: Hashassasin, Veneno, Polymer, and Darip. These are using GoA for them. I don't have any with WF totem because we only recently started using WWS within the last few weeks. I wish we had used it sooner so I could compare some fights.

The Officers of my guild typically always run 1 druid in our party even if all 3 rogues and the 1 warrior are there. This is an extreme rarity btw. Usually one of the 4 is not in the raid. So standard makeup is either:

1 Feral Druid, 1 Fury Warrior, 1 Enhance Shaman, 2 Rogues
1 Feral Druid, 1 Enhance Shaman, 3 Rogues

The party members, and my guild officers have made a few decisions (totem use and party makeup) that drastically difffers from some commenly held "truths". But the issue that I come to is I find their arguements and their positions solidly argued and made. I've read every single page of this thread. However considering my guild mates found results that differ from the commonly held truths, I was asking for opinions/reasons on why they may be right or wrong. Thanks again.

This thread taught me just about everything I know about Shaman. You have no idea how much I appreciate the hard work all of you have put in to make the information available to other Shaman like myself.

Anetheron Kill
Rage Winterchill Kill
Teron Gorefiend Kill
High Warlord Naj'entus Kill

P.S. Before you rag me about spec'ing into Imp Ghost Wolf, I'm the Totem/Kite/Puller for Trash Mob waves in BT. We're not 100% to speed on every wave, so sometimes I'm hauling ass to make it out past the NPC's and to drop totems, and get all of the Banshees and Shadowy Necromancers casting on me and my totems while I run like hell so no one gets alpha striked. For the role my guild has me perform there, I find it amazingly useful. The 1 second cast changed the situations from me dying some of the time, to never dying.

I'm sorry, but I dont know what you want us to tell you. From what you describe (no wws from when you used to drop WF) we really have no useful information to impart other than what has been stated time and time again: for rogues and warriors, windfury is much, much better than GoA. Nothing else scale with gear as it does. You've given us no information to try to understand why your melee classes feel like that, and, from your statement that you do not have the data from your wf runs, they dont seem to have that info either.

Now, I dont mean to sound like I am flaming you, but this has been asked time and time again.

As to why your melee classes do not realize how good WF is, one can only guess. Maybe they dont have sct and therefore do not notice when they get the extra attack, whereas the extra agility is right there in the tooltip. In any case, WF is still better than GoA.

Dacora · 09/20/07, 9:00 AM

Got ya Rob, thank you. Lets say though that the only OH weapon I have is faster then 2.6, maybe an epic dagger or something. Would I then want to put WF MH and FT/RB on OH? I only want to know this for general purpose. My Shaman is an Orc and I plan on dueling Gladiator axes unless I come across a better one.

Yo! · 09/20/07, 9:07 AM

Originally Posted by Morelis

I measured the performance of the trinkets in your sim using the stats from my armory profile and my usual raid buffs, I added and subtracted static bonuses as necessary. I then ran the sim again putting in the static bonuses of each trinket as well as the approximate passive contribution of each proc based on observation and the results were a bit off. For haste stuff I assumed 2.2 values as well as reduced uptime on DST.

DST:
1203 - Sim
1151 - Passive 5.14% haste (325 haste rating with 25% uptime)

MotB
1152 - Sim
1143 - Passive -120 armor (300 with 40% uptime)

Ashtongue
1175 - Sim
1160 - Passive 220 AP (275 with 80% uptime)

I don't have any empirical data for uptime on the hourglass and TT but using 0.9 ppm returned values that were about 25dps lower than the value from using the sim option for the trinkets.

The results seem to consistently over-value the trinkets especially the DST, is there something else that might be accounting for it? I also wonder if the same thing might be happening with the weapon enchant options.

Thank you for the test. My sim is not assigning dps value to trinket directly - it is a real simulation in that it runs step-by-step just like in game though much faster (Hours value represent it). Your uptime values are "plain ones" - if you are using "uptime" you have to include cross-interactions between game mechanics. Did you include flurry, mongoose, winfurry, SS to come up with those uptimes? How did you account for interactions between theese mechanics? Anything that affects speed will affect uptime. Furthermore - benefit of this uptime will be different for each stat combo. This is exactly what sim solves - shows benefit for your gear with your set of procs.

EDIT: Regarding DST - sim uses post 2.2 value of haste raitng but pre 2.2 value for ppm. I did not see new ppm values for all the trinkets yet.

EDIT: Regarding Hourglass and TT - it procs on crits only!

Nanaho · 09/20/07, 9:09 AM

Hello. Just wanted to ask if I should replace my Fel edged axe with blue pvp reward in offhand ofc. In mh I have Drakefist hammer and in the morning I was planning to put mongoose on Fel edged axe but after reading this thread im not sure if I it is a wise thing to do.

Tornhoof · 09/20/07, 9:15 AM

@Yo, I think you can remove the dragonslayer buff from your sim, it does not work with level 70 anyway. Sunflower Serenade does not work either.

Disquette · 09/20/07, 10:03 AM

Nanaho, please you the simulator linked on the front page (do a search for "Yo!" to find it), put in your stats, and compare the weapons.

Everwatch - we're going to have much cleaner data on this soon, hopefully. You could even be part of the testing crew that shows it! Please check the thread linked below, and sign up if you have the time/inclination:

PTR 2.3 (not here yet) testing issues / signup

Malan · 09/20/07, 10:04 AM

@Rob - I'm incorporating the weapon discussion into the OP.

@Everwatch - without a WWS for identical groups, on identical fights, using WF and then GoA, there's no comparison to be made. All you provided were WWS of GoA groups with several different class makeups, on several different fights.

Originally Posted by GamingManiac

I'd run simulators, but I have a Mac and no apple ID to download Java 8 for Yo!'s sim. Thank you very much.

You can sign up for the Apple dev site regardless of whether you have an appleID or not, its open to the public as long as you register for it.

Zerath · 09/20/07, 10:22 AM

Originally Posted by Malan

I blame Rob, I think I copied/pasted that from him.

That was my mistake, sorry.

Nite_Moogle · 09/20/07, 10:50 AM

Originally Posted by Dacora

I'm not so sure I am fully understanding the Windfury Weapon topic. Are you saying it is still better to put WF on both weapons, as long as the OH weapon is 2.6 or slower, or that putting WF on your MH with Flametouge (if the target is not immune to fire) on your OH, or even put WF on your OH and FT on your MH?

WF/WF is better than every other weapon combination, as long as you are using slow weapons.

Also, with the subject of haste, you say there may be a valley between 1.5 and 1.41 that may lower our dps so try not to get that fast. What if our haste makes use go faster then 1.41?

It gets even worse in terms of WF procs, but by then your autoattack DPS will have skyrocketed and you'll still be doing oodles of damage.

Malan · 09/20/07, 10:51 AM

Originally Posted by Nite_Moogle

It gets even worse.

No, the theory about that valley said that at 1.41 - 1.51 you drop WF DPS (although you gain a lot of white DPS) and then once at 1.40 or lower you gain all that WF DPS again.

Oh you ninja edited, I saw that.

Stigmata · 09/20/07, 10:58 AM

0.5 MH/OH in Hyjal with the necromancer buff last night, that was 12 swings for every available WF proc, insane dps and alot of fun for the melee.

The term "gets worse" is a bad one, your dps gets better, that is the main thing to remember, and even with this theoretical "valley" it has been proved to not be the case in some very limited testing.

Nemaa · 09/20/07, 10:58 AM

I don't think it gets even worse. Theorycally 1.49 is bad, because you'll need 3 more hits after a WF proc to proc it again "wasting" 1.47 seconds after the WF internal cooldown ((3x1.49)-3) . Let's say you are at 1.01 weapon speed, you still need 3 more hits after a WF proc to proc it again (3x1.01=3.03) But you "waste" only 0.03 seconds so you just filled the WF cooldown with 2 good hits.

Shabadu · 09/20/07, 11:05 AM

I haven't had a 2 stack of Necro buff in a while, either our priests have given up on me, or more likely, they're double and triple stacking the feral tanks for threat.

Malan · 09/20/07, 11:07 AM

I no longer use the phrase "lose DPS" as its totally inaccurate. I prefer to say that its a "drop in potential WF output" or "increased delay between leaving the WF cooldown state and having an eligible hit for a WF proc." Both of those are much more descriptive of what the problem is.

Edit - btw, just how common of a drop is the Syphon and the Rising Tide? Every shaman in BT seems to have 2 Syphons and a Tide to play with.

Shawndreya · 09/20/07, 11:12 AM

From what I've read, Syphon is one of Supremus' most frequent drops.

So obviously, I'm pretty psyched that soon I may have them.

-Edit- I had read that the syphon effect was amplified by warlock and shadow priest debuffs, yet in your SS it shows it being a stable 20. Are these rumors not true? Or were your casters targeting something else?

Nemaa · 09/20/07, 11:29 AM

We killed Supremus and Naj'entus 16 times and only 2 syphons and 3 rising tides dropped. The syphon effect is amplifyied by every debuff the mob has (misery, cos, etc). The bad side is that mechanical mobs like Crystalcore robots, Void Reaver in The Eye are immune to lifesteal what means no extra damage from the proc. Reliquary of Souls is immune as well.