Enhance Shaman: The Collected Works of Theorycraft, Vol I - Page 103

Malan · 09/14/07, 9:36 AM

Empirical evidence from people that have the Talisman says otherwise. Stigmata and several others who have the trinket and have been using it for some time have all said that they have tracked between 85-89% proc rates.
Berg presented his reasoning on this here.

If you believe otherwise, and have the trinket, go perform the test that Berg suggested and use a 2H weapon to SS and track the proc rate.

@Tristan - I'm looking at my armor XML page and its saying my base strength is 107, not 103, and my base agility is lower. Where is the difference coming from?

<baseStats>
        <strength attack="776" base="107" block="19" effective="398"/>
        <agility armor="636" attack="-1" base="59" critHitPercent="14.39" effective="318"/

forge · 09/14/07, 10:00 AM

Originally Posted by Malan

@Tristan - I'm looking at my armor XML page and its saying my base strength is 107, not 103, and my base agility is lower. Where is the difference coming from?

<baseStats>
        <strength attack="776" base="107" block="19" effective="398"/>
        <agility armor="636" attack="-1" base="59" critHitPercent="14.39" effective="318"/

I did a bit of grabbing of stuff and this is what i found out, purely from the armory which might be an armory bug or a real bug in the game. So this could be a nice idea to confirm from real game stats ?

Orc Shaman
base str 105
base agi 61
base sta 116
base int 105
base spi 123

Tauren Shaman
base str 107
base agi 59
base sta 116
base int 103
base spi 122

Troll Shaman
base str 103
base agi 66
base sta 115
base int 104
base spi 121

Draenei Shaman
base str 103
base agi 61
base sta 113
base int 109
base spi 122

Malan · 09/14/07, 10:02 AM

Bizarre, I hadn't realized that the races had different base stats.

forge · 09/14/07, 10:09 AM

Ah, did some more digging (using wowwiki) and apparently every class has different stat distribution depending on what race they pick. So it's not a bug or anything, it's most probably some class/race balancing around the racial abilities etc ?

vorda · 09/14/07, 10:10 AM

Originally Posted by Malan

Bizarre, I hadn't realized that the races had different base stats.

Has been known since beta actually, I remember even taken it into consideration at one point when choosing what race too start with (then I decided War Stomp > all :p )

Malan · 09/14/07, 10:10 AM

And yet, not a piece of information made available at the character creation screen. Another point of contention to put in the Information for the Masses thread!

Speaking of which, if you guys don't read that thread, here's the funny bit I posted in it this morning.

One of my guildmates was on his alt warrior and was tanking an instance, after awhile they realized the shaman in the group had been dropping Grace of Air. So they asked if he could start dropping Windfury for better threat generation. The shaman replied "No I can't, I already have it on my weapon."

vorda · 09/14/07, 10:20 AM

Originally Posted by Malan

And yet, not a piece of information made available at the character creation screen. Another point of contention to put in the Information for the Masses thread!

I even seem to remember that hunter RAP naked just doenst match their agility (as in, they should have XX more RAP) at all.

Anyway, enough OT from me now.

◊ Shalas · 09/14/07, 12:16 PM

Originally Posted by Malan

And yet, not a piece of information made available at the character creation screen. Another point of contention to put in the Information for the Masses thread!

Base stats don't actually do anything other than slightly modify the benefit you get from BoK. If you carefully watch your stats while leveling, you'll see plenty of times where you gain 2 stamina but under 20 health, 2 strength but under 4 AP, etc.

vokzhen · 09/14/07, 1:52 PM

Originally Posted by vorda

I even seem to remember that hunter RAP naked just doenst match their agility (as in, they should have XX more RAP) at all.

Anyway, enough OT from me now.

That's because stats aren't actually a straight 1agi=1RAP, but rather (totalagility - somenumber) = RAP, or something similar. Likewise, after deciding to put together a melee dps set on my priest for no apparent reason, I discovered my AP is str-10. For figuring out gear is doesn't make any difference, and would still be 1agi=1RAP, or 1str=1AP, in those examples. It's just small differences over the entire stat pool as a whole.

/OT

◊ Rob · 09/14/07, 2:01 PM

Yo!:
I am getting a very different value for Haste Rating from the sim compared to earlier this week. 1.1 AEP then vs 1.9 AEP now. The possible explanations I could come up with this are:
* Haste is inflating the number of Mongoose procs (this appears not to be the case in Live & PTR testing)
* Code base accidentally switched to 2.1 value of haste
* Haste is or was being incorrectly combined with Flurry
* BoK is being applied to haste, or some other interference from the buffs tab

For reference, previous tests estimated the pre-2.2 value of haste at 2.22 AEP and so we expected the post-2.2 value of haste to be 1.48 AEP. I would happily accept this new 1.9 AEP value but for the fact that I was getting 1.1 AEP from the sim earlier this week. Any idea why this might be the case?

Also, a suggestion for the "Procs & stuff" tab: Add a checkbox for the 3% increased critical damage from the Relentless Earthstorm Diamond meta gem.

And finally, I don't think the Thottbot "Item Score" tool works properly. It doesn't seem to multiply the actual values in the way that I would expect. For example, Cataclysm Gloves score better than Skyshatter Grips on the page but Skyshatter Grips are better when I manually add up the value of each stat on the item. (Perhaps the tool isn't counting the socket on the Skyshatter Grips even though I have a value of 18 assigned to a yellow socket?) I would not suggest using this tool until someone can explain why this is occuring. My current theory is that it's computing some kind of "efficiency" by dividing the item's score by its item level.

Morelis · 09/14/07, 2:57 PM

Originally Posted by Rob

And finally, I don't think the Thottbot "Item Score" tool works properly. It doesn't seem to multiply the actual values in the way that I would expect. For example, Cataclysm Gloves score better than Skyshatter Grips on the page but Skyshatter Grips are better when I manually add up the value of each stat on the item. (Perhaps the tool isn't counting the socket on the Skyshatter Grips even though I have a value of 18 assigned to a yellow socket?) I would not suggest using this tool until someone can explain why this is occuring.

I can't explain why it happens but it is fairly consistent, for whatever reason it messes up on the value of a bunch of tier 5 set items (not just the shaman ones). I've also observed that sometimes shoulders and bracers have all of their values doubled for no apparent reason. I still think it's a great tool given all the stats it tracks that lootzor doesn't, it just requires the user to be on their toes for things that seem out of whack.

Yo! · 09/14/07, 5:09 PM

Originally Posted by Rob

Yo!:
I am getting a very different value for Haste Rating from the sim compared to earlier this week. 1.1 AEP then vs 1.9 AEP now

Buffs that are "on" now are cauising it most likely - many did not include buffs before and are getting different results with buffs now. Raid buffs provide huge bonuses.
If you still see something strange - you may also want to play with sim turning most mechanics off and maybe find where something wrong is happenning.
I used your armory profile (1540 ap, 31.5 crit <-- increased with str and agi from BoK, 16 hit, 0 haste) and with default buffs it produces 1012 dps and 1.67 crit, 1.52 hit, 1.77 haste (post patch 2.2 value). Without the buffs I get haste at 1.35 aep and 744 dps.
Sidenote: for those who think that Crit is always better than Hit - if you put on the very best armor (will post links later, still not finished finding that super-set) you will get crit so high that UR will be on 100%, almost all attacks will be flurried and hit (still below hit cap) becomes valued more than crit because by increasing hit you are increasing damage and chance of on-hit procs on normal attacks while increasing crit at that point is increasing damage mostly.
As for +3% to crit damage metagem and +haste metagem (with hidden cooldown) they were included into sim today on procs tab.

• Disquette · 09/14/07, 5:27 PM

As for finding strange things - I was getting low dps numbers (like 200 or less) when I turned on Crusader in the offhand, which returned to the 900 to 1000dps level when I switched it back to mongoose.

rava · 09/14/07, 6:01 PM

I remember Yo asking for armor values of bosses so I went through my TargetArmor saved variables and pulled out what I saw, couldn't find Illidan anywhere in there for some reason, but:

["Hydross the Unstable:-1"] = {
["MeleePercent"] = 90.65588344578481,
["TargetDamage"] = 12086851,
["RangedHits"] = 0,
["TargetPoints"] = 361,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["MeleeHits"] = 135,
["The Lurker Below:-1"] = {
["MeleePercent"] = 72.61636977902651,
["TargetDamage"] = 14461548,
["RangedHits"] = 0,
["TargetPoints"] = 384,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["MeleeHits"] = 106,
["Leotheras the Blind:-1"] = {
["MeleePercent"] = 90.4753846722851,
["TargetDamage"] = 12706970,
["RangedHits"] = 0,
["MeleeHits"] = 131,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 361,
["Fathom-Lord Karathress:-1"] = {
["MeleePercent"] = 48.85333228480048,
["TargetDamage"] = 7065359,
["RangedHits"] = 0,
["MeleeHits"] = 67,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 398,
["Morogrim Tidewalker:-1"] = {
["MeleePercent"] = 248.9926252935603,
["TargetDamage"] = 36014657,
["RangedHits"] = 0,
["TargetPoints"] = 636,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["MeleeHits"] = 358,
["Lady Vashj:-1"] = {
["MeleePercent"] = 182.5408140306064,
["TargetDamage"] = 20643325,
["RangedHits"] = 0,
["TargetPoints"] = 435,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["MeleeHits"] = 246,
["Al'ar:-1"] = {
["MeleePercent"] = 85.85027964112118,
["TargetDamage"] = 10978248,
["RangedHits"] = 0,
["TargetPoints"] = 486,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["MeleeHits"] = 127,
["Void Reaver:-1"] = {
["MeleePercent"] = 65.54655061008738,
["TargetDamage"] = 17031095,
["RangedHits"] = 0,
["TargetPoints"] = 377,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["MeleeHits"] = 101,
["High Astromancer Solarian:-1"] = {
["MeleePercent"] = 74.72848204403455,
["TargetDamage"] = 10930465,
["RangedHits"] = 0,
["TargetPoints"] = 365,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["MeleeHits"] = 103,
["Kael'thas Sunstrider:-1"] = {
["MeleePercent"] = 176.3647025877637,
["TargetDamage"] = 16137281,
["RangedHits"] = 0,
["TargetPoints"] = 452,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["MeleeHits"] = 218,
["Rage Winterchill:-1"] = {
["MeleePercent"] = 74.98291001343254,
["TargetDamage"] = 10207405,
["RangedHits"] = 0,
["MeleeHits"] = 100,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 240,
["Anetheron:-1"] = {
["MeleePercent"] = 110.0134920667448,
["TargetDamage"] = 10566031,
["RangedHits"] = 0,
["MeleeHits"] = 151,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 290,
["Kaz'rogal:-1"] = {
["MeleePercent"] = 34.02481719993865,
["TargetDamage"] = 4079367,
["RangedHits"] = 0,
["MeleeHits"] = 45,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 96,
["Azgalor:-1"] = {
["MeleePercent"] = 28.99024631795682,
["TargetDamage"] = 3797048,
["RangedHits"] = 0,
["MeleeHits"] = 39,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 91,
["Archimonde:-1"] = {
["MeleePercent"] = 136.3754397064038,
["TargetDamage"] = 11542858,
["RangedHits"] = 0,
["MeleeHits"] = 187,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 256,
["High Warlord Naj'entus:-1"] = {
["MeleePercent"] = 78.98313304195391,
["TargetDamage"] = 13013574,
["RangedHits"] = 0,
["MeleeHits"] = 114,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 353,
["Supremus:-1"] = {
["MeleePercent"] = 25.74575904314052,
["TargetDamage"] = 7992773,
["RangedHits"] = 0,
["MeleeHits"] = 37,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 182,
["Shade of Akama:-1"] = {
["MeleePercent"] = 19.31851167371959,
["TargetDamage"] = 2898413,
["RangedHits"] = 0,
["TargetPoints"] = 295,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["MeleeHits"] = 26,
["Teron Gorefiend:-1"] = {
["MeleePercent"] = 219.0667478831767,
["TargetDamage"] = 25422540,
["RangedHits"] = 0,
["TargetPoints"] = 508,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["MeleeHits"] = 300,
["Gurtogg Bloodboil:-1"] = {
["MeleePercent"] = 164.3253322910583,
["TargetDamage"] = 21503518,
["RangedHits"] = 0,
["MeleeHits"] = 236,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 386,
["Essence of Suffering:-1"] = {
["MeleePercent"] = 115.9768434646575,
["TargetDamage"] = 15004562,
["RangedHits"] = 0,
["MeleeHits"] = 137,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 663,
["Essence of Anger:-1"] = {
["MeleePercent"] = 31.58787810114778,
["TargetDamage"] = 14179397,
["RangedHits"] = 0,
["MeleeHits"] = 71,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 472,
["Essence of Desire:-1"] = {
["MeleePercent"] = 186.3774201634872,
["TargetDamage"] = 17665047,
["RangedHits"] = 0,
["MeleeHits"] = 215,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 582,
["Mother Shahraz:-1"] = {
["MeleePercent"] = 926.288844926868,
["TargetDamage"] = 48750263,
["RangedHits"] = 0,
["MeleeHits"] = 1221,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 1330,
["Gathios the Shatterer:-1"] = {
["MeleePercent"] = 101.5948107769565,
["TargetDamage"] = 11759165,
["RangedHits"] = 0,
["MeleeHits"] = 147,
["LastLevel"] = 70,
["RangedPercent"] = 0,
["TargetPoints"] = 970,

The lua from the mod says:

--FIRST STEP: CALCULATE ARMOR
local pr = 0;
if ( TotalMeleeHits > TotalRangedHits ) then
pr = 1 - ( TotalMeleePercentage ) / ( TotalMeleeHits );
else
pr = 1 - ( TotalRangedPercentage ) / ( TotalRangedHits );
end

local ea = TargetArmor_PercentToArmor(pr, PlayerLevel);

pr = floor(pr*10000+.5)/100;
ea = floor(ea+.5);

I'd post specific #'s for them but I do not have a calculator handy and I am far too lazy to click with my mouse!

Also, any chance of adding Solarion's Sapphire to the buff window? 87.5 extra ap if the warrior has Mind Mastery

Yo! · 09/14/07, 6:02 PM

Originally Posted by Disquette

As for finding strange things - I was getting low dps numbers (like 200 or less) when I turned on Crusader in the offhand, which returned to the 900 to 1000dps level when I switched it back to mongoose.

Yes it was a bug. Thank you for such direct indication of strange behavior.

Originally Posted by rava

I remember Yo asking for armor values of bosses so I went through my TargetArmor saved variables and pulled out what I saw, couldn't find Illidan anywhere in there for some reason, but:

Thank you. Will add bosses as drop-down menu. Illidan is estimated to have 3500 as someone posted earlier.

everwatch · 09/14/07, 6:12 PM

Originally Posted by Malan

Shab, since you've been using the Prism extensively, I had previously done some envelope calculations when its value was tweaked and estimated it as a 10% threat reduction for anyone sitting at around 30% crit. Do you find this to be reasonably true?

@Rob - I agree, but you know how it is, people will want to know the *exact* theorized gain for each class for each totem.

I've had the Prism for at least one month now, but I believe it's closer to 2 months in actuality. In a raid environment with our group makeup I sit at 38%-42%+ buffed crit (no mongoose - I'm at 30.23% unbuffed). I've found the Prism to be at least a -10% threat reduction no questions asked. The difference between using the Prism and not using it is literally the difference between night and day for me. My dps has jumped up significantly because I rarely have to worry about stealing agro. I use it on most boss encounters because I still get threat capped, however there are a number of them where it would be more efficient to swap something else out probably where melee threat is low and tank threat is high.

Polymer, our dps warrior, got his a few weeks ago. He described the exact same results as I had from an out of class example.

Dukanull · 09/14/07, 6:27 PM

Originally Posted by Celetroll

[Ashtongue Talisman of Vision] info remains still wrong.

It does not have 89% uptime. You can do successful SS but no buff is applied.

It does exactly what tooltip says - 50% chance on hit to proc. So two times 50% chance per stormstrike since both weapons get a new roll. That gives 75% proc chance. Add to this dodge/parry on stormstrike, and it goes even lower.

I'm tracking an 84% uptime on my talisman, and I'm sitting at 95ish Hit Rating atm. I've heard the same said in multiple other places, so I'm sticking to thinking the Tooltip is wrong.

falonub · 09/14/07, 8:41 PM

Hey sorry to be interrupting...again :P was just wondering what you guys thought on which of these two necklaces:

[Pendant of the Perilous] was what I had before, gave me 30.08% crit 1572ap (with my current gear unbuffed)

last night I finally got [Choker of Serrated Blades] and with it equipped takes me to 29.04% crit 1578ap (with current gear unbuffed) was wondering if you guys think the ignore armor makes up for the 1.04% crit or is the crit more important

Tornhoof · 09/14/07, 9:56 PM

Originally Posted by Tristan

@Tornhoof, when you rip the armory page in WoWEquipOptimizer do you remove buffs manually (or add missing) etc that the player logged out with/without?

From what I can see on the armory page there is only two values being passed (based and effective):

<baseStats>
  <strength attack="528" base="103" block="13" effective="274"/>
  <agility armor="538" attack="-1" base="61" critHitPercent="12.43" effective="269"/>
</baseStats>

<melee>
  <power base="668" effective="1068" increasedDps="76.0"/>
</melee>

As I read it 103 is the strength base without additions from items/buffs and 528 is with items/buffs included. Otherwise wouldn't the program suggest very different when I logout with resp. without buffs?

Actually I use hardcoded base values for each class, then add the item attributes, I don't use the actual attribute values from the xml (and honestly I didn't know they include base values, back when I implemented the parser, they did not). This is not as accurate as having the exact values but it is fine enough.

Malan · 09/15/07, 12:51 AM

Originally Posted by falonub

Hey sorry to be interrupting...again :P was just wondering what you guys thought on which of these two necklaces:

[Pendant of the Perilous] was what I had before, gave me 30.08% crit 1572ap (with my current gear unbuffed)

last night I finally got [Choker of Serrated Blades] and with it equipped takes me to 29.04% crit 1578ap (with current gear unbuffed) was wondering if you guys think the ignore armor makes up for the 1.04% crit or is the crit more important

Originally Posted by Malan

Strength = 2 (2.2 w/Kings)
Crit Rating = 2
Agility = 1.8 (2 w/Kings)
Hit Rating = 1.4
Attack Power = 1
Armor Penetration = 0.25 (low ball estimate)

You do the math.

• Krish · 09/15/07, 1:57 PM

I've been trying to accurately weigh the value of the 4-piece bonus on T5, and running into a question that doesn't seem to have been answered here before. If my searching capabilities for this thread have failed, I apologize (though I have tried).

The issue that I am running into is this: It has been suggested that a way to value it would be to take your flurry uptime from something like Yo!'s simulator and multiply it by the value of haste rating equal to 5%. For me, this equation looks like this:

4SetBonusValue = NewHasteRatingValue * 15.77 * 5 * FlurryUptime

With my results being:

115.91 = 1.75 * 15.77 * 5 * .84

115.91 AEP is an awful lot to give up. Even the 4-set bonus on T6 should only be worth 70 AEP. If you are willing to forego T6 with the assumption that some of the non-set leather and mail pieces from Hyjal/BT are better, you'd have to save up probably 2-3 of the best ones before giving up that 4-piece bonus is worth it.

Ok, so that being said, here is my question: How do haste rating and the haste from flurry stack? For example, If you have 10% haste from gear, and you have no T5 set bonus affecting flurry's normal 30% haste increase, then what is your overall haste while your attacks are flurried? Is it 37% (multiplicative) or 40% (additive)? The answer to this question must surely affect the validity of my equation above. If it is truly additive, then simulator results and my equation should be accurate. If, on the other hand, it is multiplicative, then an equation to determine its value would be more complex, as the relative value of haste rating itself would be greater for flurry downtime than flurry uptime, and calculations would have to first determine how beneficial haste is while not flurried and while flurried separately.

If anyone knows the answers to these questions, it is surely a few of the people on these boards who have so meticulously tested and parsed some of their results. If we could nail this down perhaps Yo! could add it into his simulator as both a checkbox to enable it and as an AEP output box. I would be most grateful for any feedback that could help sort this out.

Yo! · 09/15/07, 3:38 PM

Originally Posted by Krish

I've been trying to accurately weigh the value of the 4-piece bonus on T5, and running into a question that doesn't seem to have been answered here before. If my searching capabilities for this thread have failed, I apologize (though I have tried).

The issue that I am running into is this: It has been suggested that a way to value it would be to take your flurry uptime from something like Yo!'s simulator and multiply it by the value of haste rating equal to 5%. For me, this equation looks like this:

4SetBonusValue = NewHasteRatingValue * 15.77 * 5 * FlurryUptime

With my results being:

115.91 = 1.75 * 15.77 * 5 * .84

115.91 AEP is an awful lot to give up. Even the 4-set bonus on T6 should only be worth 70 AEP. If you are willing to forego T6 with the assumption that some of the non-set leather and mail pieces from Hyjal/BT are better, you'd have to save up probably 2-3 of the best ones before giving up that 4-piece bonus is worth it.

Ok, so that being said, here is my question: How do haste rating and the haste from flurry stack? For example, If you have 10% haste from gear, and you have no T5 set bonus affecting flurry's normal 30% haste increase, then what is your overall haste while your attacks are flurried? Is it 37% (multiplicative) or 40% (additive)? The answer to this question must surely affect the validity of my equation above. If it is truly additive, then simulator results and my equation should be accurate. If, on the other hand, it is multiplicative, then an equation to determine its value would be more complex, as the relative value of haste rating itself would be greater for flurry downtime than flurry uptime, and calculations would have to first determine how beneficial haste is while not flurried and while flurried separately.

If anyone knows the answers to these questions, it is surely a few of the people on these boards who have so meticulously tested and parsed some of their results. If we could nail this down perhaps Yo! could add it into his simulator as both a checkbox to enable it and as an AEP output box. I would be most grateful for any feedback that could help sort this out.

Speed bonuses are applied multiplicative, there is "blue" confirmation on how haste works.
If you have flurry, mongoose, 10% from haste rating and 2.6 speed weapon - the next attack will be after 2.6/(1.3*1.02*1.1) seconds.

Varag · 09/15/07, 3:46 PM

Is it possible to add a changelog to the first page of the thread, or some sort of dating so that its easy to identify at a glance what's changed recently for those of us that prefer to just watch that page vs. the entire thread?

vorda · 09/15/07, 3:53 PM

which may have already been addressed

It has been, alot of times. I suggest you use the search function.

edit: nice edit..

Tristan · 09/15/07, 4:13 PM

Originally Posted by Tornhoof

Actually I use hardcoded base values for each class, then add the item attributes, I don't use the actual attribute values from the xml (and honestly I didn't know they include base values, back when I implemented the parser, they did not). This is not as accurate as having the exact values but it is fine enough.

So you parse through all gear and download them pages (couldn't find any item info in the base page but I might be just stupid) for each item and then add them to the total?