Warrior Weapon HATE Tables

[Dillusional]

Originally Posted by mycoolrausch [You must be logged in to view images. Log in or Register.]

Does this change if a proc is resisted? I know hate is calculated before resist but some of these procs are ticking dots, what about that hate?

The way i calculated hate for the procs with DoTs was that I assumed the non-damage component scaled to 400 and then i gave you all the hate for a DoT up front. So if Trident was supposed to do 25 damage per tick for 20 ticks. You got 400 for the slow component and 500 for the dot up front. I also assume you got the same hate if it was a resist or not. I never looked at the emu source to verify that was how DoT hate is calculated. if someone says otherwise, now is the time to correct me. Because if that isn't how it works, weapons with DoTs are much weaker than I was thinking they were. I should go look at the emu source heh.

[Dillusional]

For slow, stun, poison counters just count each of those as 400 agroe. Then add the total damage a dot or dd will do to that. A warrior with 255dex will proc twice a minute on average. So multiply the proc value by 2. The agroe generated by a warrior swinging with the primary hand will be: (swings per minute) * (weapon damage + damage bonus). At level 60 the damage bonus is 11. That is how you calculate the agroe generated by a weapon per minute if you want to build your chart correctly @OP. At least that's what i did above and how I believe an accurate chart would be calculated now.

[Cerate]

@Dillusional You're not considering haste, which would improve the performance of white hate generation. Consider this formula for swings per minute, following from the wiki's haste guide:

60/((DLY/10)/(1+Haste)) = swings per minute

Haste is expressed as a decimel, so 100% haste would be 1.00 here. You can use 0.21 if you want to figure it unbuffed with, say, an FBSS, but I don't think that's the situation for most warriors, most of the time, in raids.

Into the whole formula for total hate per minute with 100% haste:

60/((DLY/10)/(1+Haste)) * (DMG + Bonus) + (Proc Hate) * 2 (rate @255 Dex) = Total Hate per minute.

As for DoTs, I'm simply adding on 10 ticks, considering continuous uptime but that's really liberal and probably not realistic.

[Dillusional]

Originally Posted by Cerate [You must be logged in to view images. Log in or Register.]

@Dillusional You're not considering haste, which would improve the performance of white hate generation. Consider this formula for swings per minute, following from the wiki's haste guide:

60/((DLY/10)/(1+Haste)) = swings per minute

Haste is expressed as a decimel, so 100% haste would be 1.00 here. You can use 0.21 if you want to figure it unbuffed with, say, an FBSS, but I don't think that's the situation for most warriors, most of the time, in raids.

Into the whole formula for total hate per minute with 100% haste:

60/((DLY/10)/(1+Haste)) * (DMG + Bonus) + (Proc Hate) * 2 (rate @255 Dex) = Total Hate per minute.

As for DoTs, I'm simply adding on 10 ticks, considering continuous uptime but that's really liberal and probably not realistic.

Sure, factoring haste into the formula is a good thing. I'm glad you were able to put together the formula out of my rambling post. I'd like to point out though, that many warriors tank with 99% haste, not 100%. The best worn haste items are 41% and most people ask enchanters for the 42minute spell Visions of Grandeur that gives 58% haste 20 attack and 25dex.....In fact, in my years of raiding as an enchanter, I've only ever had one warrior consistently ask for a different haste spell when he didn't need to use null aura.

I don't know what you meant by "adding 10 ticks." What I was doing was saying that you get all of the hate from a DoT effect up front. 1 hate per damage. Not as the ticks occur. The assumption was you get this whether there is a resist or not. The durations and damage per tick data was taken from the wiki even though I have all those dot weapons in game =/ (yeah, lazy right).

Weapons I am unsure about how to calculate: Feverblade and Wess. Do each of the non-damage components of Wess count separately (they might). I have a Feverblade but haven't tested how the dot or "heal" actually works here.

[Cerate]

Sure, 99% haste would be more accurate.

DoTs deal aggro both up-front and on ticks. It's not very apparent for warriors because most DoT ticks we get are quite small, but it's obvious for shamans, necros and druids. NPCs turn on ticks. I've added 10 ticks of damage to represent one minute of DoT damage with the initial proc hate. As I said, if it's not continuously maintained (resists or unlucky with procs) then that's going to overestimate the value. Ticks also do less damage when the NPC is moving, so that's a factor, too. Thankfully, even the upper bound for tick hate is relatively small.

I'm also not sure about Feverblade ticks yet. When I parsed it, I had my control wait until I got the worn off message, then told him to start swinging. So the Feverblade proc plus any hidden ticks, if they generate hate or not, added up to about 550. I'm not sure what all the ticks are even supposed to add up to, from the description it sounds like it starts at 22 then the ticks go down in damage from there, like shaman epic in reverse? So that's just kind of a mess.

WESS parses for me to about 400-450, probably getting capped at 400 as per the new post-chardok cap for non-damaging procs. I don't have access to a bio orb or ragefire BP, but that might be a way to try to isolate a blind's hate for warriors, if it's less than that cap.

[baakss]

Unless they changed it from EQEmu, there should be no difference between 99% haste and 100% haste ever.

Haste rounds down to the nearest whole number because of integer math.

[Dillusional]

Originally Posted by baakss [You must be logged in to view images. Log in or Register.]

Unless they changed it from EQEmu, there should be no difference between 99% haste and 100% haste ever.

Haste rounds down to the nearest whole number because of integer math.

I don't believe this to be the case.

I just looked at the emu source and it seems to work like this....

attackSpeedInMillisecond = (originalDelayInMilliseconds / hasteMod)

speed and delay are both integers but hasteMod was a floating point (number with a decimal point)


if you have 99% haste, your hasteMod value would be 1.99 if you have 100% haste, your haste mod would be 2.0

So let's say you have a 35 delay weapon with 100% haste

3500 / 2 = 1750 ....so you will attack once every 1.75 seconds

and let's say you have 99% haste

3500 / 1.99 = 1758.79.... rounded down to 1758 ....

you are taking 8 fewer milliseconds between swings with 100% haste versus 99% haste on a 35 delay weapon. The server will round down to the nearest millisecond because of the Integer math but you'll still have an 8 millisecond difference. I'm pretty sure the attack timer can handle 8 milliseconds of precision and there will be a tangible difference.

[baakss]

You're absolutely right. I missed the multiplication of delay by 100 back when I had originally looked at this, which matters tremendously because the cast to int is now with milliseconds instead of seconds.

Thank you for correcting me!

[baakss]

Well, that's possible too, but you were still correct on how that function works.

[Dolalin]

One last consideration that's missing here is the fact that proc hate scales based on mob HP. Raid mobs have tons of HP.

If I get a stun proc (400 aggro) on a mob with 32k HP, it generates more than 400 aggro. It's (400 * C * 32,000), where C is some constant scaling factor.

In that light, for mobs with very high HP, you can effectively discount aggro from the direct damage component of procs I would think, as I believe it is added on separately and does not scale up with mob hp.

That would change how these various weapons rank in terms of one another, to the point where for mobs with high HP, the effectiveness of a weapon's hate generation takes this order of precedence:

1. Scaling proc hate (400 * C * HP)
2. White damage hate
3. DD / recurring DoT hate