Kenco: A Guide To Threat. - Page 2

• Polleke · 12/04/06, 11:33 AM

Feint does -600 threat, or -800 threat with rank 5, from the AQ book. Note that it is affected by your threat modifiers, so that the rank 5 ability reduces only 800 * 0.8 = 640 threat by default, or 800 * 0.56 = 4448 threat with blessing of salvation.

Typo, should be 448.

sp00n · 12/04/06, 12:02 PM

Also the modifier is now 29% (as stated in your description), thus the equation would be
"800 * 0.71 = 568 threat by default, or 800 * 0.71 * 0.7 = 397.6 threat with blessing of salvation."

modhelm · 12/04/06, 3:52 PM

Originally Posted by Bury

Possibly. The exact wording of the tooltip appears to be "threat generated by your shadow spells", not "threat generated by your damage shadow spells". Does the game consider healing from shadow spells to be shadow-related threat? I hope so.

Vampiric embrace is on the "affected spells" list for Shadow Affinity on thott, so it seems likely.

• Vinsent · 12/04/06, 4:23 PM

(D) Paladin
Paladins receive 50% threat from healing compared to other healers. This is designed to stop them tanking instances by healing everything in sight. As a side affect, this gives them a decent advantage over other healers where healing threat is an issue.

When the Righteous Fury buff is active, Holy damage causes 1.6 threat, or 1.9 threat if you have 3/3 Improved Righteous Fury.
Holy Shield damage gives another 1.2 multiplier, for 1.9 * 1.2 = 2.28 threat per damage with talented Righteous Fury.

Holy sheild is 1.35 in 2.0

I believe that paladin healing aggro is 67% not 50% but I havent tested this in a while, that is the number I remember though.

Also RF is changed (at least according to the tooltip) to affect all holy spells (assuming it inculdes healing now)

Might want to add something about DS, and Iceblocks mechanics, as well as FD/Vanish.

Good compliation though, thanks for reposting it.

Irshish · 12/04/06, 4:48 PM

These forms of buffs all have infinite range; they will cause threat to all mobs on whose threat list you are on. Furthermore, the threat caused is split equally among all the affected mobs. If you are on one mobs threat list, a 1000 point heal will cause 500 threat to that mob. If 5 mobs are aware of you, the same heal will cause 100 threat on each mob.

Can anyone provide some links or info that backs up this claim? I've seen it refuted in a number of places and it just doesn't seem very logical to me. From a game play standpoint I could see developers arguing a mob's thoughts are "I've got friends so I don't feel as threatened by that." But from a coder standpoint that would require an object to know about all the other objects attacking the player creating the threat. As a healer I seem to be able to pull a 2ndary mob the same if there are 2 mobs or 5 mobs, but that is purely anecdotal and I've never tested it.

◊ Praetorian · 12/04/06, 5:05 PM

Here's the best evidence of shared heal threat: Gluth with a kiting DPS class.

A mage who kites Gluth mobs isn't constantly hammering on all of them, and yet a mob they fireblast once can still be on them a minute later. I can shock a zombie (they spawn on everyone's hate list) once and it'll stick to me for pretty much the whole phase unless another DPS class pulls it off. Meanwhile you have healers landing 2k+ heals fairly often on the tanks, but they don't get aggro. That's simply impossible if a single heal on the Gluth tank generates 1000 aggro on all zombies in the room.

• Arawethion · 12/04/06, 5:12 PM

Originally Posted by Praetorian

Here's the best evidence of shared heal threat: Gluth with a kiting DPS class.

A mage who kites Gluth mobs isn't constantly hammering on all of them, and yet a mob they fireblast once can still be on them a minute later. I can shock a zombie (they spawn on everyone's hate list) once and it'll stick to me for pretty much the whole phase unless another DPS class pulls it off. Meanwhile you have healers landing 2k+ heals fairly often on the tanks, but they don't get aggro. That's simply impossible if a single heal on the Gluth tank generates 1000 aggro on all zombies in the room.

Similarly, if you use a healing class to kite, their aggro numbers are deceptively low. Mages trying to help out with Rank 1 CoC can get aggro easily, and the Warriors trying to build Rage for PH also get hit easily. KTM will show the Priest's threat as being far above the tanks, and he won't pull aggro.

• frmorrison · 12/04/06, 5:18 PM

Originally Posted by Malan

As I understand it, both Earth Shock and Rockbiter are losing their threat component in the patch.

Earth shock doesn't generate double aggro anymore, but on the PTR Rockbiter's tooltip says increases dps by x and increases aggro generated by attacks. Not sure on the beta version, but two weeks ago it did not have the word aggro in the tooltip.

I am interested to see what the devs decide on tomorrow.

• Whiteknight · 12/04/06, 5:46 PM

Originally Posted by xarg

To be slightly nitpicky as well - there are two different thunderfury debuffs. One of these is the attack speed debuff (current target only) and the other is the NR debuff (4 targets). I'm not sure which to go with - entirely different values are cited in this thread.

Updated to include the thread link for the source of the TF data. Note the discrepancy arises from one document including the warrior def stance modifier and the other leaving it out. Kenco's data is the combined hate for both procs. Also, the numbers differ by a fairly small amount even when you take into account the stance modifiers - I'm assuming that's the result of refining the test data.

Originally Posted by Arawethion

This is the stuff that's been known for a while, correct? Just never posted here in a consolidated way before?
Agreed with above--a 2.0 version would be great.

Correct. Kenco had posted this data on the WOW R&D forum where it got stickied, but lost after Bliz changed their forum format. It's also posted on the linked website I quote above, however that is the guild website for a guild that no longer exists. I wanted to repost it here on an active forum so that it didn't get lost, and that we can use it as a reference.

Obviously updating it to include 2.0 information is desirable, but I won't pollute the document with speculation. As the data becomes available, I'll update the document as needed.

Roana · 12/04/06, 5:48 PM

About split healing threat:

Originally Posted by Irshish

Can anyone provide some links or info that backs up this claim?

Simple, I've run tests. Have one unit generate a fixed amount of aggro on one or more mobs (Warrior with Demoralizing Shout, Voidwalker with Suffering, etc.). Then see how much healing it needs to pull aggro. You'll need twice as much healing to pull aggro on two mobs compared to one, thrice as much to pull aggro on three mobs, and so forth. It's easily reproducible. You can do it by yourself with even a fairly low-level warlock (Voidwalker + Suffering + Health Funnel).

I've seen it refuted in a number of places and it just doesn't seem very logical to me.

Whoever refuted it never did any actual testing, I am afraid, but was just arguing from gut feeling.

Also, it has nothing to do with being logical, but simply with the game remaining playable. If the threat by healing didn't diminish with the number of mobs, tanks would soon be unable to hold aggro on them, unless you gave them ridiculous AE aggro tools (even Consecration with Improved Righteous Fury would not hold up to Greater Heal spam if healing threat weren't split).

In particular, rage mechanics wouldn't scale: Rage is a linear function of the number of mobs (assuming all mobs do equal damage), total healing threat that would have to be offset if it weren't split would be a quadratic function of the number of mobs.

From a game play standpoint I could see developers arguing a mob's thoughts are "I've got friends so I don't feel as threatened by that." But from a coder standpoint that would require an object to know about all the other objects attacking the player creating the threat. As a healer I seem to be able to pull a 2ndary mob the same if there are 2 mobs or 5 mobs, but that is purely anecdotal and I've never tested it.

Just go and test it. It's 100% reproducible.

Irshish · 12/04/06, 6:28 PM

Thanks for the responses. After Praetorian post I started going through the numbers on what it would take to pull aggro in a typical RG scenario. If a hunter did a single arcane shot to a dragonkin and 20 mobs are active at any time, a healer could do close to 10k healing before pulling aggro on that dragonkin (w/o any threat reducing talents or buffs). This seems pretty close to my own experience with the fight.

Now I have to go eat some humble pie and admit to a guildmate I was wrong.

• Tharas · 12/04/06, 8:18 PM

Could you please slap a version tag on the title of this thread (please)? (And perhaps make a new thread for the 2.0 threat numbers as they are discovered?)

If that's too much work, I understand, but it might be nice to have these open in two tabs and see the comparison.
Alternately, a clever format showing the changes from 1.12 to 2.0 would also work.

I have really learned a lot from the Kenco threat posts on the old WoW forums and used this information as much as possible while becoming a decent tank. Some of our raid strategies (Nefarian, Fankriss, etc) originated from a couple of the guild tanks discussing the threat potential of Demo Shout, Battle Shout, etc. and how to make effective use of those abilities with the personnel, specs and gear level we had. I think it's especially important for the tanks in a guild to understand the threat mechanics and how they change.

(Basically, thanks for doing this and please keep on trucking)

• Whiteknight · 12/04/06, 10:49 PM

Originally Posted by Whiteknight

Originally Posted by xarg

To be slightly nitpicky as well - there are two different thunderfury debuffs. One of these is the attack speed debuff (current target only) and the other is the NR debuff (4 targets). I'm not sure which to go with - entirely different values are cited in this thread.

Updated to include the thread link for the source of the TF data. Note the discrepancy arises from one document including the warrior def stance modifier and the other leaving it out. Kenco's data is the combined hate for both procs. Also, the numbers differ by a fairly small amount even when you take into account the stance modifiers - I'm assuming that's the result of refining the test data.

Minor correction - I'd missed the fact that the numbers measured in the TF thread were done with defensive stance and defiance, prior to the game update that made these multiplicitive. So, in order to compare the TF thread measured numbers to Kenco's unmodified threat values, you have to divide by 1.45.

Thus the value 350/1.45 = 241 (within the error bounds of the measurement data).
I.e. Kenco's numbers for TF proc threat exactly match the research done in the thread you've quoted.

xarg · 12/05/06, 12:28 AM

Originally Posted by Whiteknight

Originally Posted by xarg

To be slightly nitpicky as well - there are two different thunderfury debuffs. One of these is the attack speed debuff (current target only) and the other is the NR debuff (4 targets). I'm not sure which to go with - entirely different values are cited in this thread.

Updated to include the thread link for the source of the TF data. Note the discrepancy arises from one document including the warrior def stance modifier and the other leaving it out. Kenco's data is the combined hate for both procs. Also, the numbers differ by a fairly small amount even when you take into account the stance modifiers - I'm assuming that's the result of refining the test data.

Minor correction - I'd missed the fact that the numbers measured in the TF thread were done with defensive stance and defiance, prior to the game update that made these multiplicitive. So, in order to compare the TF thread measured numbers to Kenco's unmodified threat values, you have to divide by 1.45.

Thus the value 350/1.45 = 241 (within the error bounds of the measurement data).
I.e. Kenco's numbers for TF proc threat exactly match the research done in the thread you've quoted.

Ahh, nice sleuthing, I must have missed that. Separating the two for a program like KTM is a bit of a waste of time, I agree, so that makes good sense :)

Vytae · 12/05/06, 12:38 AM

All warlock curses cause threat apon initial application (possible exception CoA,Siphon life however has initial threat application before the ticks start). They are very small amounts ( no more then 100 for max rank spells) but its worth noting.

It's hard to test because it depends on very small numbers with max rank curses and tanks missing.

Example: Raid was running into Heigan's room warrior taunts but is resisted i see him swing and cast CoR anyhow but his first swing was unluckily missed/parried and i pulled aggro from 10 yards (the tank not realizing right away he had lost aggro kept running into the room for half second,so i got 2 shot).

I had no regen's going (was full life,only mana missing was that from the casted curse) wasnt using pots/buffing etc. Ive unluckily managed to duplicate it only a few times (not that i was trying on purpose,i died every time).

Updating this for tonights patch would be spectacular,i suspect i will be showing our mages this thread within our first raid night =P

Nikisa · 12/05/06, 7:55 AM

Originally Posted by Whiteknight

(E) Priest

Fade reduces your threat temporarily by a fixed amount - 820 at max rank, and is not affected by threat modifiers, such as Silent Resolve. This is a good thing since a priest would only have negative threat modifiers on. When the buff ends, you get the same threat back, so there is no net gain from spamming Fade you should leave it as a panic button instead. Note that you cant fade to below 0 threat.

Since we don't have paladins we let the priests and only the priests heal on aggro sensitive pulls like the Drakes, Chromaggus and the Twin Emperors. If we fade before casting any heals we never draw aggro. Healing pulling aggro during the pull was one of our main problems on Firemaw, and since we started doing this we have never had any issues with it, so I believe it is possible to get lower than zero threat with fade.

Ayr · 12/05/06, 8:24 AM

Originally Posted by Nikisa

Originally Posted by Whiteknight

(E) Priest

Fade reduces your threat temporarily by a fixed amount - 820 at max rank, and is not affected by threat modifiers, such as Silent Resolve. This is a good thing since a priest would only have negative threat modifiers on. When the buff ends, you get the same threat back, so there is no net gain from spamming Fade you should leave it as a panic button instead. Note that you cant fade to below 0 threat.

Since we don't have paladins we let the priests and only the priests heal on aggro sensitive pulls like the Drakes, Chromaggus and the Twin Emperors. If we fade before casting any heals we never draw aggro. Healing pulling aggro during the pull was one of our main problems on Firemaw, and since we started doing this we have never had any issues with it, so I believe it is possible to get lower than zero threat with fade.

It makes more sense to me that the Fade buff acts as a constant modifier to threat for the duration of the buff, rather than an initial -aggro. So ActualThreat = Max(0, Threat-820) as long as Fade is up, not actually reducing Threat by 820 in the instant it's cast and doing nothing after that.

Evert · 12/05/06, 9:23 AM

Originally Posted by Nikisa

so I believe it is possible to get lower than zero threat with fade.

Lower than zero threat would put you out of combat. I think?

• mutagen · 12/05/06, 4:51 PM

With Fade active and Silent Resolve a priest will have to do somewhere over 2000 healing (I'll leave the exact number to the mathcrafters) to have threat above 0. At the rate tanks can build aggro there is isn't much of an opportunity to pull aggro unless the tank has a terribly unlucky string of miss / parry / dodge.

I'm sure Blizzard has explicit mechanics built into Fade that prevent it from taking you out of combat, which may include setting a Faded priest's threat to a floor of 0 as Ayr points out. Otherwise we'd have all sorts of fun tricks pretending to be a hunter if we hadn't generated much threat.

Has any rigorous testing been done on threat generated by Power Word:Shield? I did some tests that seemed to indicate it was lower threat than healing, leading me to the conclusion that it was 1/4 the amound shielded in white damage threat or half the threat of an equivalent heal. However, while I was attempting to follow Kenco's methodology closely my testing partner wasn't so interested in the details and I was unable to get exact numbers. The 130% threshold may have been a factor as well. Overall though it did seem to generate less threat than the equivalent heal, though I'd be more confident saying that if someone did some more rigorous testing.

Snow · 12/05/06, 5:17 PM

Originally Posted by Evert

Originally Posted by Nikisa

so I believe it is possible to get lower than zero threat with fade.

Lower than zero threat would put you out of combat. I think?

Unless you have evidence otherwise, I doubt this. Makes a lot more sense if it's just a in-combat/out-of-combat flag. I know for instance that you can leave combat with the wickerman abominations, yet still be on their aggro list(and they will pull to you when they leash off someone else).

Nikisa · 12/05/06, 7:05 PM

Originally Posted by mutagen

Has any rigorous testing been done on threat generated by Power Word:Shield? I did some tests that seemed to indicate it was lower threat than healing, leading me to the conclusion that it was 1/4 the amound shielded in white damage threat or half the threat of an equivalent heal.

This has been confirmed in rigorous testing.

• Althor · 12/05/06, 7:46 PM

It's already been tested and confirmed that Fade can make your threat go negative.

Also, while Fade is only a temporary agro reduction, the end of Fade doesn't force a recheck on whether or not the Priest pulls agro. You need to cast another agro changing spell for that to happen. So in effect, Fade lowers your threat level for a minimum of 10 seconds or whatever up to whenever you cast a new threat increasins spell after that time (a tick of Renew that isn't all overheal counts).

This makes Fade good on touchy boss pulls. Once the Priest enters combat from the boss pulse, they can Fade and heal a bit without pulling agro. As long as they stop healing when Fade is about to wear off if the warrior still doesn't have a strong agro hold then they'll be fine.

• Whiteknight · 12/07/06, 3:55 PM

Updated the OP to cover all the 2.0.1 changes that I'm aware of. Mainly just by reading Kenco's v18.8 KLHThreatMeter and the associated change notes.

(edit: oh, as requested I've kept the 1.12 version as an archive/reference)

Two things that I still need to verify:
- The quantified change to battleshout AOE threat.
- Does Paladin Righteous Fury really affect heals, or was the word