[top] .75 x.4 (40% increase)

Grandleaf · 03/18/08, 9:16 PM

Kioga, I was wondering the same thing about Vaccine's list as well. As it stands, it looks like he is 33 def short.

As for Vaccine's list, it is pretty solid and I will be going for the same items. In addition, here are a few additions I will be going for:
[Demontooth Shoulderpads]
[Figurine - Empyrean Tortoise] - with Vaccine's list, I should be able to reach the defense cap with this.

I personally will gem my gear to favor stamina.

Krag · 03/19/08, 1:15 AM

Originally Posted by Grandleaf

Kioga, I was wondering the same thing about Vaccine's list as well. As it stands, it looks like he is 33 def short.

As for Vaccine's list, it is pretty solid and I will be going for the same items. In addition, here are a few additions I will be going for:
[Demontooth Shoulderpads]
[Figurine - Empyrean Tortoise] - with Vaccine's list, I should be able to reach the defense cap with this.

I personally will gem my gear to favor stamina.

He might be counting in an [Elixir of Ironskin]

Vaccine · 03/19/08, 6:11 AM

Originally Posted by Grandleaf

Kioga, I was wondering the same thing about Vaccine's list as well. As it stands, it looks like he is 33 def short.

As for Vaccine's list, it is pretty solid and I will be going for the same items. In addition, here are a few additions I will be going for:
[Demontooth Shoulderpads]
[Figurine - Empyrean Tortoise] - with Vaccine's list, I should be able to reach the defense cap with this.

I personally will gem my gear to favor stamina.

Yes I was counting the elixir.

The problem with using the demontooth shoulders is that you break the 4xT6 bonus which I've found pretty crucial in maintaining threat over our DPS Crazy warlocks. I think dropping the swipe bonus would drop me back down to Lacerate as the main threat device and even with the new AP buff it doesn't do much for the threat. Have to experiment with them.

As for the agility gems I Was just really using them as kind of an experiment. I'll be socketing +15 stam and +5agi/7stam in red slots to make up the set bonus which are for the most part excellent.

I may look at making up the defense with a trinket instead of the MT heroic one, possibly the hydross one or the new alchemy one.

Inaiwae · 03/19/08, 6:14 AM

@ Vaccine: i tried to re-gem your setup to favor stamina. The rules: red sockets 5/7, blue and yellow 15 stamina. When yellow socket is on a item, put 15 stamina everywhere.

Your gemming: 133 agility, 43 stamina --> 9.3% dodge, 664 HP
Stam gemming (5/7, 15 gems): 42 agility, 195 stamina --> 2.9% dodge, 3012 HP

So total stats, buffed (your numbers):
HP: 23556
Armour: 35592
Dodge: 67.6%

Total stats with stam:
HP: 26029
Armour: 35390 (-200 armor from agility)
Dodge: 60.8%

Numbers i used: 1 stam = 15.45 HP, 1% dodge = 14.7 agi. I multiplied agi by 1.03, and also for total stats i multiplied both agi and HP by 1.1 (kings) before changing the total numbers.

In case you took 15stam only path, it would be:
0 agility, 228 stamina (socket bonus loss compared to 5/7 way: -7 agility, -15 stamina) --> 3522hp

Total stats with pure stam:
HP: 26590
Armour: 35290 (-300 armor from agility)
Dodge: 57.65%

I agree that the 5/7 15 gems is the best way.

edit: fixed armor values

Maeltne · 03/19/08, 1:19 PM

Ignoring armor, if you took the three method and averaged out over a long period what the effect of the dodge would be on the efficiency of health you would get the following

Method 1: max dodge: 23556 / (1-.676) = 72,703 effective hit points
Method 2: balanced: 26029 / (1-.608) = 66,400 effective hit points
Method 3: max stam: 26590 / (1-.5765) = 62,786 effective hit points

This *seems* to indicate that method 1 would be the preferred. But as I understand it the bosses in Sunwell hit so hard that dodge isn't really an effective method of survival. Our buffer really needs to be large enough to survive.

I think I'm going to have to come up with a simulation that calculates the max, min, mode, and average hits until dead, based on supplied min, max, average hit and whether or not the boss crushes. Most of our current simulators work off of average, but I think the most important is actually min. In theory I see dodge as having little/no effect on the minimum number of hits until dead.

tedv · 03/19/08, 2:13 PM

Originally Posted by Spookeh

Monkens- ya. you need a /cancelform as the first line (i think?), but otherwise it works perfectly.

DO NOT press it while you're under a global cooldown though, or all that happens is you cancelform :P

I use:

/cancelform
/use item:22105
/use item:22104
/use item:22103
/use item:32905
/use item:32784
/use item:28100
/use item:22829
/cast !Dire Bear Form

Which is healthstone, and then one of: [nethergon, red ogre, volatile, super] in that order of priority.

Is it possible to include some basic lua scripting that checks your form before you shift out and then shifts you back into that form when done, so you can use the same macro for both cat and bear form?

Septus · 03/19/08, 2:31 PM

So, I don't know if EJ's strategy prohibitions apply to Magister's Terrace as well, but if not, has anyone done the heroic several times to get a feel for it?

I'm really not optimistic about main tanking that place, as opposed to using a pally with direct spell and damage reduction, or a warrior with the same plus spell reflect. And I don't know how useful an offtank might be.

Druidus · 03/19/08, 3:09 PM

Since patch 2.0 (I guess) its not possible to use lua to cast spells... that was making the game too automatic...
You also cant check buffs (like forms) and hp afaik. We must go manual on this one...

Toskk · 03/19/08, 3:44 PM

I hate to sidetrack a conversation, but I'm searching for actual data (tested with different amounts of expertise rating on lots of different bosses over tens of thousands of attacks) on raid boss dodge rate. Anyone happen to know of a source? I'd really like to confirm a solid number on this one, for the feral DPS model.

Allev · 03/19/08, 4:15 PM

May I ask which solid number you're looking to pin down? Dodge rate from the back or parries from the front? As far as I know, the rogues pinned down a decent number in their threads where they no longer saw dodged attacks from behind, but there hasn't been enough expertise gear to figure the exact parry number in front.

Edit: With the addition of 2.4, we should see enough expertise gear to find the parry number, though it may take both new trinkets, T6 boots, badge/Sunwell neck, Vashj belt, [Gloves of the Searing Grip], and even possibly [Clefthoof Hide Leggings] to reach that point for a feral. Warriors may have more luck.

Toskk · 03/19/08, 4:23 PM

Originally Posted by Allev

May I ask which solid number you're looking to pin down? Dodge rate from the back or parries from the front? As far as I know, the rogues pinned down a decent number in their threads where they no longer saw dodged attacks from behind, but there hasn't been enough expertise gear to figure the exact parry number in front.

Ooh.. that's a good point. Dodges from the back is what I'm looking for

I'll take a look over in the Rogue section to see what I can find.

Uhman · 03/19/08, 5:20 PM

Originally Posted by tedv

Is it possible to include some basic lua scripting that checks your form before you shift out and then shifts you back into that form when done, so you can use the same macro for both cat and bear form?

/use item:22105
/use item:22104
/use item:22103
/use item:32905
/use item:32784
/use item:28100
/use item:22829
/cast [stance:1] !Dire Bear Form; [stance:2] !Aquatic Form; [stance:3] !Cat Form; [stance:4] !Travel Form

Im using the last line for my Powershiftmacro. This could work

Allev · 03/19/08, 5:30 PM

Originally Posted by Toskk

Ooh.. that's a good point. Dodges from the back is what I'm looking for

I'll take a look over in the Rogue section to see what I can find.

I found a reference to references of 6.25% for the cap: Rogue Gear Spreadsheet

Rogue gear spreadsheet thread, posts 1517 and about the next 5 after that. The WWS reports it discusses have long expired, however.

Edit: Different than the 5% the calculator uses, looks like!

Allev · 03/19/08, 6:27 PM

In hopes of introducing a topic that's appropriate with the debut of tons more expertise gear on tanking items:

I've recently wondered how to value it exactly in relation to other stats for tanking-- current calculators don't seem to give much help. Obviously it's a great stat for threat, and a solid hybrid DPS stat, but what about about tanking stats? Is it better point-for-point than, say, agility? Is it a better survival stat than stamina? Can you quantify spreading out the damage?

Specifically, it seems to me that the first item you wear with expertise on it would be the best, because of instagibs. Let's define an instagib where two consecutive attacks by the player are parried, causing the boss to have back-to-back attacks far faster than expected. He needs to hit you to instagib you, so we'll throw that component in as well.

The chance of an instagib if you get parried 10% of the time and dodge 50% of the time requires this sequence:

boss hit/crush -> player parry -> player parry -> boss hit/crush

The chance of this sequence happening is .5 * .1 * .1 *.5 = .0025: 1 in 400 swings. This seems to quantify to 1-2 pure instagibs a night, with a good amount of those instagibs being crushes.

Say you get 16 expertise rating for 1% fewer parries, bringing our parry chance to 1%. This gives us an instagib possibility of .5 * .09 * .09 *.5 = .002025, or 1 in about 489 swings. We only get .002025/.0025 = 81% of the instagibs we were seeing before by only having to eliminate 1/10th of our parries.

Say you get 1% dodge instead. The math goes to .49 * .1 *.1 *.49 = .002401, which means we still get .002401/.0025 = 96% of instagibs we saw before.

Now, say you get 90 expertise rating: 5% fewer parries, which brings our parry chance to 5%. Then, the instagib sequence has a chance of happening to be .5 * .05 *.05 * .5 = 1 in 1600 swings. Without the expertise, there would have been 4 instagibs in 1600 swings, so you're effectively eliminating 3/4 of the "worst-case" scenarios.

This is in addition to the overall mitigation outlined in the expertise thread, which put the value somewhere around 1 expertise rating equaling .25-.5 dodge rating depending on the boss's specials ("mitigation" in Rawr) and the obvious time-to-live it generally gives for parries (Rawr's "survival").

Can anyone actually provide a weighting on this stat? Is it still "personal preference"? I've nearly convinced myself that [Brooch of Deftness] over [Necklace of the Juggernaut] is a bigger tanking upgrade now than the new badge ring over the Mag ring, but I don't have "hard" numbers, just napkin math like above.

Toskk · 03/19/08, 7:24 PM

Originally Posted by Allev

I found a reference to references of 6.25% for the cap: Rogue Gear Spreadsheet

Rogue gear spreadsheet thread, posts 1517 and about the next 5 after that. The WWS reports it discusses have long expired, however.

Edit: Different than the 5% the calculator uses, looks like!

Hmm.. that 6.25% number seems to come down to the 3-5 wws logs of the author of the Rogue DPS spreadsheet.. and it looked mostly like he just used that number because his gear had it, and he wasn't recording any dodges. From a thorough search, I'm still seeing five different values (5.00%, 5.60%, 6.00%, 6.25%, and even 6.50%) used as the dodge chance for a raid boss.

Anyone happen to have any really solid data on boss dodge chance? I'd really like to find a body of data on par with the data collection that was used to determine raid boss miss chance. For reference, that link is:

[Rogue/Warrior] Weapon Skill Adjustment Discussion

kalbear · 03/19/08, 7:40 PM

I would think you'd want to keep both the Brooch and the necklace around, just in case you need the defense to stay capped. But you raise a good point - the chances of being parrygibbed go way, way down with expertise relative to dodge. Dodge will in general reduce your overall damage more, but expertise will reduce your spike damage significantly. I would imagine you'd want more dodge when the amount of damage over time is great but the individual hits are small, whereas you'd want expertise where the hits aren't as frequent but are much harder.

So the necklace for prince, the brooch for gruul (for example).

Krag · 03/19/08, 9:44 PM

Originally Posted by tedv

Is it possible to include some basic lua scripting that checks your form before you shift out and then shifts you back into that form when done, so you can use the same macro for both cat and bear form?

There's a macro somewhere back in the thread that does this (with the help of conditionals and stopmacro, you cannot use lua checks). I'm not on my wow computer now, but if it's impossible to find I'll post my macro at some point later.

raffy · 03/19/08, 11:39 PM

Is it possible to include some basic lua scripting that checks your form before you shift out and then shifts you back into that form when done, so you can use the same macro for both cat and bear form?

Originally Posted by Uhman

/use item:22105
/use item:22104
/use item:22103
/use item:32905
/use item:32784
/use item:28100
/use item:22829
/cast [stance:1] !Dire Bear Form; [stance:2] !Aquatic Form; [stance:3] !Cat Form; [stance:4] !Travel Form

Im using the last line for my Powershiftmacro. This could work

The above will not work.

Make one macro with all your /uses.
Put it on your bars somewhere.
Move your mouse over the button and run this command:

/run DEFAULT_CHAT_FRAME:AddMessage(GetMouseFocus():GetName())

It should print out something like "TheButtonX".

Make a second macro:

/click [noform] TheButtonX
/stopmacro [noform]
/cancelform [form:1]
/click [noform] TheButtonX
/cast [noform] Dire Bear Form
/stopmacro [noform:3]
/cancelform
/click TheButtonX
/cast Cat Form

Remember, you CANNOT make a potion/hs macro GCD safe. Using !Cat Form or !Bear Form will not work, since you can only consume consumables in caster. autoUnshift will not exit form to use consumables.

Astrylian · 03/19/08, 11:58 PM

Originally Posted by raffy

autoUnshift will not exit form to use consumables.

Technically, this part is false; it will shift you out to use a consumable. However, using consumables can be done while you're in GCD, so it will shift you out while you're GCD'd, which is still bad, and means that as you say, we still can't make a GCD-safe pot macro.

Thessaly · 03/20/08, 1:38 AM

I use

#showtooltip
/cancelform [stance:1]
/use [nostance] Master Healthstone
/cast [nostance] Dire Bear Form
/stopmacro [nostance]
/cancelform [stance:3]
/use Master Healthstone
/cast Cat Form

I got it off one of the other Feral threads that has since fallen by the wayside ages ago, it basically uses the /stopmacro [nostance] as a conditional of sorts, I don't actually completely understand how it works. The way it does work, is if I'm in bear, it'll use the HS and shift back to bear, if I'm in cat, it'll use the HS and shift back to cat. Insert consumable of choice, enjoy. You could fit multiple consumables if you drop the tooltip line, I think.

Tasonir · 03/20/08, 3:02 AM

@Allev

I asked about expertise as a purely defensive stat a long ways back in this thread (Somewhere around page 40 or 50 I think)...The basic answer given (by dukes, again if my memory holds) that I agree with is that it's nothing considerable. With some basic assumptions he worked out parry adding about 2.4% damage over the course of a fight. I don't see it as being all that major that you must remove parries to smooth out damage. I look forward to getting expertise on my gear that isn't otherwise a downgrade (like the current badge neck which has no avoidance or anti-crit), but I really don't think the mitigation of parry is a huge concern.

Your instagib situation really doesn't look that scary to me. Let's say the boss has a 1.5 second swing timer - the next swing comes in on average .48* seconds faster. I would really expect to survive that, I don't see it being an instant death at all. You should have the hit points to always take 2 hits (preferably 3) - they aren't doing any increased damage, unlike a critical strike, which is something you really don't want to take.

*random value in range [1.5 to 0]

[top] .75 x.4 (40% increase)

.3 seconds. I am doing this correctly, right? Let's say two occurred at the same time just to make it simple - an increase of .48 seconds would result.

Spookeh · 03/20/08, 7:12 AM

Maeltne - In theory I see dodge as having little/no effect on the minimum number of hits until dead.

of course not, you can always fail to dodge 100 times in a row, its random after all..

tedv - no i dont think you can get it to go back into the previous form for you
Uhman - same thing, it won't work because once you take the potion you're not in a form, so the shift line at the end will think you're in stance:0

raffy's macros look like they'd work tho

Divinefactor · 03/20/08, 10:27 AM

Originally Posted by Tasonir

parry adding about 2.4% damage over the course of a fight. I don't see it as being all that major that you must remove parries to smooth out damage.

*random value in range [1.5 to 0]

[top] .75 x.4 (40% increase)

.3 seconds. I am doing this correctly, right? Let's say two occurred at the same time just to make it simple - an increase of .48 seconds would result.

The problem with Parry is not that it adds 2.4% dmg over the course of a fight, but when that 2.4% dmg comes.

More often than not if your tank dies its not because your healers ran out of mana but rather the tank got spiked. Parry strings (when multiple parrys occur one after the other) greatly increases your risk of dieing. For instance:

0s Boss swings 4000dmg
1s He parries
1.5s Boss swings 4000dmg
2.5s He parries
3s Boss Swings 4000dmg

Numbers are a bit skewed to help the example and my ultra simple math. But basically in this example, you swing twice and get parried. Your incoming damage goes from 8000 to 12000. If you are full health, sure that spike wont kill you, but if you were not full health, it could.

Also, I dont think anyone is advocating stacking as much expertise and hit as you can, but as far as neck goes, I would take [Brooch of Deftness] over [Necklace of the Juggernaut] because the slight amount of dodge vs the massive amount of hit and expertise is an easy choice for me. It would get a bit tougher of a choice once you have access to [Pendant of Titans]. Life is full of choices.

Allev · 03/20/08, 11:07 AM

Originally Posted by Tasonir

@Allev

I asked about expertise as a purely defensive stat a long ways back in this thread (Somewhere around page 40 or 50 I think)...The basic answer given (by dukes, again if my memory holds) that I agree with is that it's nothing considerable. With some basic assumptions he worked out parry adding about 2.4% damage over the course of a fight. I don't see it as being all that major that you must remove parries to smooth out damage. I look forward to getting expertise on my gear that isn't otherwise a downgrade (like the current badge neck which has no avoidance or anti-crit), but I really don't think the mitigation of parry is a huge concern.

Your instagib situation really doesn't look that scary to me. Let's say the boss has a 1.5 second swing timer - the next swing comes in on average .48* seconds faster. I would really expect to survive that, I don't see it being an instant death at all. You should have the hit points to always take 2 hits (preferably 3) - they aren't doing any increased damage, unlike a critical strike, which is something you really don't want to take.

*random value in range [1.5 to 0]

[top] .75 x.4 (40% increase)

.3 seconds. I am doing this correctly, right? Let's say two occurred at the same time just to make it simple - an increase of .48 seconds would result.

The reason the "instagib" mechanic is so bad is it couples with other things. What if those hits crush? What if the boss uses a special? The "instagib" still gives him a chance to autoattack before and after the first attacks, meaning you could very well take 3-4 hits in about 2 seconds. At that point, every fraction of a second you can make available for a healer to get a heal in is huge. If you have 3 healers on the MT with 2 second heals, you may only need to buy another fraction of a second to increase your shoft-term effective HP pool by whatever that heal is.

I do think it's "all that major" to reduce spikes-- after 10-man content, preserving healer mana from healing the tank isn't really an issue issue. So "over the course of a fight" damage isn't an issue at all, it's not what will kill your tanks. The fights that are stressful on healer mana are because of raid damage, not tank damage. Also, the largest advantage a druid as over any other tank is "smoother" the damage curve-- we often take more damage overall, but it's "easier" to heal-- doesn't expertise play into this philosophy?

Why, exactly, is the pure tanking theory for warriors "stack short-term mitigation to the maximum, stack avoidance when there isn't anything left to get" but druid philosophy seems to almost be the reverse, even when we're tanking the big hitters? At the very least, developing some serious math about its value so it can be added into existing tanking models. "Not considerable" just doesn't look right.

bluenote · 03/20/08, 12:47 PM

Any of you guys plan on getting near the hit cap and some expertise before you start tanking Sunwell? I am pretty sure we need to get hit capped for the bosses in there. I am considering using [Insidious Bands][Band of the Abyssal Lord] along with [Earthwarden] because of the expertise. Also, let's not forget about [Pendant of Titans], might have to re-gem a couple of items with +5 agi +5 hit and [Boots of Natural Grace].