sp00n

It should be rather easy to test.
While slamming, you don't do any auto attacks, so just let a mob beat upon you until you have enough rage for a slam. Then let somebody else equipped with a 1 DPS weapon and possibly naked beat on the mob until he gains aggro.

 

Enkidu

Some (belated) Spell Reflect numbers:

The aggro is on a 1:1 ratio, stance modifiers apply. Did not test if Defiance applies however.
numbers:
Defensive Stance 311 bolt reflected, expected 444.73 aggro swap, aggro swapped at 445.
Battle Stance 281 reflected. expected 247.28 aggro swap, aggro swapped at 250 (mob was casting at the time however so was probably 248-249).

 

Whiteknight

It's been a while since I've updated this data. I finally got around to comparing Threat/Omen and KTM.

The most significant differences / discrepancies are - revenge - the KTM and Threat estimates differ by a factor of 2.
And Mangle - Omen lists a multiplier of 1.3, while KTM uses 1.5.


There are also various cases where one or the other threat library simply doesn't implement specific ranks of abilities. Big ones are shield bash/cleave max rank (not in ktm) and many of the lower spell ranks for Threat

 

ramenchef

I was under the impression that the druid talent for 15% threat was additive, not multiplicative.

 

Melthar

Mangle was changed to 1.5threat with 1.0x weapon damage with the "druid nerf patch". A later patch increased the damage on mangle to be 1.15*weapon damage, but redcued the threat coefficent so that "overall threat would remain unchanged" This fits with the 1.3 figure (1.15*1.3=1.495)

 

Darkstar

Im sorry if i missed this going thru this thread, but any info on the various agro reduction trinkets out there ? Im thinking of getting the one from KoT for my dps warrior.

-Darkstar-SSL

 

Roywyn

General consensus is that they are pretty junk. They reduce your threat on nearby targets (usually 30 yards) by a fixed amount.
You can get their data by looking them up on wowhead/thottbot and click their "Use:" ability to see how much threat they reduce.

Timelapse Shard - Items - World of Warcraft -901 threat, Shrouding Potion - Items - World of Warcraft -1500 threat

Prism of Inner Calm - Items - World of Warcraft causes your physical hits to reduce your global threat by 150, and your spell crits to reduce your global threat by 1000. It's considered to be quite nice for dual wielders.

Edit: Thanks for the info on the Prism.

 

Duilliath

I honestly believe the biggest threat 'advantage' those trinkets and similar items (such as the cloak from Sporeggar) offer is that you're gimping your overall damage output, making you less likely of actually needing the 'On Use" ability...

 

Abbi

KTM is wrong on Revenge. He's using the old threat values; in 2.1, Revenge was altered to do more damage but less threat.

 

Antiarc

I forget how KTM handles this, but Threat guesses threat values for lower ranks by linear extrapolation. They likely aren't accurate, but they're close enough to provide reasonable data, and can be tested and entered explicitly if it ever becomes clear that a guessed value is grossly incorrect.

 

Wuff

You are wrong about the Prism of Inner Calm. It does not reduce the threat of your physical/spell crits at all. It removes the amount (150/1000) from your global threat.

As a destruction warlock I found some fun ways to exploit this mechanic (let my void taunt a melee mob once, then play aggro ping-pong with rank 1 shadow bolts for hours). You can constantly reduce your threat to 0 if you are using low dmg high crit rate spells.

Imagine two casters using this trinket with a very high crit rate and lots of long range instants (moonkins perhaps?) ...
This item is pretty much broken at the moment, I expect it to be changed a 3rd or 4th time.

PS: Sadly, this thing is worthless for warlock AoE (seed does not proc it) and it does not work with Omen currently.

 

Prinsesa

Not quite. I believe the change was that Revenge's innate threat was reduced while upping the damage to keep threat values the relatively the same. I say relatively because when translating the threat to damage, you're probably going to see some reduction if your target has a lot of armor.

"We do want Sanctuary to be the tanking seal"

- Ghostcrawler

 

JamesVZ

For the purposes of how KTM tracks threat, that would be a significant shift upwards in TPS if true (old threat value + new damage = damn near twice the threat from revenge). However, I thought he patched that in like 20.1, if not much earlier?

 

Asmodeu

The AQ40 trinket is still a flat 50%?

 

Whiteknight

From what I've read in the KTM code, it has a table with all the known values for each rank. I wasn't aware that Threat uses an extrapolation technique - but that should provide for reasonable estimates.

I suspect Kenco will go and add the one or two missing ranks and such, but still I think the biggest difference between the threat libraries is the Mangle multiplier.


[edit:]
Looking at the latest KTM code, it seems Kenco has the proper value for revenge and mangle. It's done in a patch function and has probably been in place for a while - I just missed it when reading the file. (I haven't gone back and checked how long it's been in place - I just assume Kenco put it in when the relevant patch went live).

Also, KTM has a similar function for inferring missing ranks of abilities from the known ones. So for the most part it should get lowbie tanking right too.

 

Ronoran

My GM/MT makes the following claim:
Is this real, made-up, ancient and patched, or just a bug in his threat meter?
Is there any known way that one warrior's casting sunder armor can contribute threat to an entirely different warrior (except by reducing physical damage mitigation)?

 

Prinsesa

@Ronoran:

Let's do some quick and dirty math. Assuming both Warriors are not auto-attacking, are in Defensive Stance and have 0/3 Defiance:

Each application of Sunder Armor rank 6 is going to apply (300 * 1.3) or 390 threat.

Warrior A Sunders the mob. The mob attacks Warrior A.
Warrior B Sunders the mob. The mob still attacks Warrior A, since both A and B have equal amounts of threat and B has not stepped over the 110% melee threshold yet.
Warrior A Sunders the mob. The mob is still attacking A. A now has 780 threat on the mob, B has 390 threat.
Warrior B Sunders the mob. The mob is still attacking A. Both Warriors now have 780 threat.
Warrior A Sunders the mob. The mob is still attacking A. A now has 1170 threat on the mob, B has 780 threat.
Warrior B Sunders the mob. The mob is still attacking A. Both Warriors now have 1170 threat.

Basically, the mob is never going to stop attacking A, assuming all the two Warriors do is keep Sundering in sequence, because B is never going to overstep the 110% threshold. If the rate and amount of threat generation from any x sources is equal, then aggro will always stay with the person who initially drew aggro.

"We do want Sanctuary to be the tanking seal"

- Ghostcrawler

 

• Playered

I think you misunderstand, he means everytime either of the warriors use sunder, both get the threat for it rather than just the one which casts it I believe

 

Antiarc

This is incorrect. Each warrior only receives threat for their own sunders.

 

Lazare

Because Sunder is a rolling debuff, it's a plausible area for odd behavior to crop up. It'd be easy to test however:

Warrior 1 builds up some threat without using sunder, then stops attacking.
Warrior 1 sunders (so the debuff stack might be considered to be "owned" by him).
Warrior 2 sunders until aggro swaps.

If your GM is correct, Warrior 2 is generating threat for Warrior 1 by refreshing Warrior 1's sunders, so aggro will never swap.

Edit: I can't imagine this actually occurs, however. But that is how you'd test...

 

Fierystix

Being the druid of note in this post, I wanted to mention I have reconducted this test with more accuracy in regards to the end heal of lifebloom, and have videos to show how the tests were conducted and concluded.

For a brief summary of the first test:

1. The warlock body pulled the mob with a health deficit (for no overheal).
2. I taunted the mob off the warlock without attacking it.
3. I cast lifebloom on the warlock.
4. The lifebloom was then purged by a horde shaman before the HoT could tick, triggering the end heal.

The mob did not turn from the druid to the warlock, even though the druid started with 0 threat. This suggested the end heal did one of the following: a) generated threat for neither parties, b) generated threat for only the druid, c) generated the same amount of threat for both parties, or d) generated more threat for the druid than the lock.

To rule out these possibilities, we had to conduct a second test. Here is a brief summary:

1. A warrior body pulled the mob with a health deficit (for no overheal).
2. I taunted the mob off the warrior.
3. A second druid taunted the mob off of me.
4. The second druid cast lifebloom on the warrior.
5. The lifebloom was then purged by a horde shaman before the HoT could tick, triggering the end heal.
6. I then did exactly one damage to see if I would pull aggro.

The mob ended up aggroing me after only one damage, meaning the druid had 0 threat on the mob after the end heal of the bloom was triggered. And because the warrior failed to pull aggro from the druid that cast lifebloom, this meant the end heal also generated 0 threat for the warrior.

So we can conclude the end heal on lifebloom generates threat for neither the druid nor the recipient of the lifebloom.

A more detailed rundown of the tests with videos showing them can be found on my blog: Lume the Mad Blog Archive Lifebloom and Threat: The End Heal

 

Intendrer

I had a question in regards to Aggro and Threat, and figured this was the best place to bring it.

I have seen lately on my hunter where I pull agro from my pet despite the fact that I haven't crossed the 130% ranged limit, let alone surpassed 100%. Sometimes I even pull at 80 or 90. I used Omen for TM and it's up to date. Any ideas? I'd love to use KTH instead, but my guild uses Omen so hence I'm stuck.

 

Liar

Playermade Threat Meters are never 100% accurate (most often if the mob has an aggro resetting/reducing move) so that is probably all there is to it.

Come WotLK though, we will have 100% accurate meters since the data will use Blizzard threat values instead of tested ones.

Unexpected TankPoints error Clearly someone doesn't understand how EH works. That, or upgrading from T8 to T9 is optional in beating Arthas. Clearly.