I am with Astrylian here on what to drop, while expertise is nice in the tanking aspect of it is a threat increase.
So it comes down to the fight, is threat or mitigation most important?
While expertise does hold some value for mitigation it isn't very high, I think it was Dukes who did some maths on it in this thread somewhere.
I don't recall the maths accurately but I recall thinking, yes it does make a difference but it isn't a very big difference.

Adding sockets to a weapon and/or a fifth stat to split the item budget with would make it easier for them to give us a choice on weapon upgrades.


I like the new macro's but I must admit some concern as to what Blizzard will do as they do something they tried to stop, it is a fairly specific case so maybe they will let it slide but I don't plan on getting used to it.
It isn't too hard to keep track of when you can and can't press the emergency button.

Don't use the healing macro when you tank anything that can stun you.

Edit: Due to Sunwell Info/Strats EJ policy

As Wowjutsu has passed 50 kills now, I assume it's ok to mention tanking aspects of Kalecgos now?

I also noticed that my threat was up a lot last night, from ~1k tps to ~1.4k tps in a 5 man (no shaman). I need to parse a run properly to work out if it is just the increased damage from lacerate, but it feels like there is something more to it - I'm wondering if they gave us something to offset the effect of cutting the GCD on mangle.

If they gave us something, it'd only be reflected in Omen if Antiarc knew about it, right? Or am I missing something stupid again?

50 kills and one week is what the post by Gurg says.

As it stands, I'd be happy for any comments in this thread such as "I geared towards stamina because of xx boss, and found a good point of xx total health for surviving the fight, what have other people who did the fight found?" without specific comments about abilities/enrages/timers/etc.

Falk: If they change the damage on abilities (such as give mangle 1.3* the damage it used to have) without changing the static threat, then Omen will still pick that up as mangle threat has always been based largely on damage->threat conversion. It's only if they actually change the direct threat values that abilities have (such as changing the base lacerate threat from 285) that there would be a need to change Omen/other threat mods.

Double post

I tried the new macro last night, and the new GCD monitor in it made the macro much less stressful to use. It's now an excellent addition to a Bear's tanking options, for a grand total of 1.

Per the expertise thread, expertise is worth .5 dodge rating for overall mitigation if the boss only auto-attacks, less if his specials hit the MT.

However, I find parry reduction to be pretty valuable. The entire point of tanking is to not die, and I believe that expertise keeps you from dying much better than agility does. Especially as we get into Sunwell, it's not damage over the course of a fight you need to mitigate, but reduce/eliminate the chance you'll get bursted down.

Think about it this way: a boss swings for 5k damage on a 1s timer. His DPS is 5k. Suppose the tank got parried .25 seconds after a swing-- thus the boss has .75 of his swing timer reduced 40% to .45s. His DPS is then 5k/.7 = 7143 DPS. As long as being parried is in the worst-case scenario, the worst-case gets a LOT worse. And the first points mitigate the worst case the fastest.

Looking at it another way: At 15% parry, you have a 16.5% chance to get parried at least 2 times in any sequence of 5 attacks. At 10% parry, you have an 8% chance to get parried at least 2 times in any sequence of 5 attacks. At 5% parry, you only have a 2.3% chance in 5 attacks for at least 2 parries. If you're swipe or lacerate spamming, those 5 attacks can happen in a little over 3 seconds.

Not getting parried won't keep you from getting hit, but on the other hand, it buys healers valuable time to get off the next heal, and buys you time personally to hit emergency buttons.

In addition to its passive benefits, the increased threat will help you get out to a larger threat lead, meaning you can use fewer special attacks, and fewer attacks = even less parries. And more specifically, your parries cluster up less often.

It's hard to quantify exactly (you can come up with a ton of anecdotal worst-cases like this), but isn't short-term mitigation and reducing the worst-case significantly more important than long-term mitigation and reducing the average-case? Isn't this why it's an absolute rule to get crit-immune?

The problem is the amounts of expertise we're getting on tanking gear. Of all seriously viable bear items, there are 4 items with expertise on them. Only 2 of which are suitable for use in Sunwell. We'll definitely want to use one of them (the boots), so the only question is whether to use one of the necks. It's certainly very viable to do so, but it's still only one item, nowhere even remotely close to making the effect as noticeable as we'd like (15%->10%->5%). I certainly wouldn't mind a new bear staff/mace with expertise on it, that'd be spiffy.

I tried finding this in the Rawr wiki (and the description of the Tossk model), but didn't see anything about it, so I thought someone here might know the answer...

In calculating DPS, does the model take into account the fact that an "on use" item can be timed with a Rip while a trinket that procs cannot?

Last night in my first heroic run through the Magister's Terrace I was lucky enough to get [Shard of Contempt]. Rawr shows it as being almost a 7 DPS upgrade over my [Crystalforged Trinket]. I'm just wondering how much that will really translate to in actual fights. I plan to try to get some numbers together using our WWS logs to make my own best guess, but maybe someone else has done that work?

Thanks!

You can't get a rip in, in 20sec?

I'm pretty certain it's the lacerate. The initial threat on lacerate does scale with damage, as evidenced by crit lacerates doing significantly more threat. Since lacerate was buffed a lot, we should expect our single-target threat to go through the roof. I was sustaining around 1600 threat in Magister's Terrace spamming only mangle and lacerate on the robots - I'm looking forward to trying Teron Gorefiend with the new lacerate.

The original post is a bit off, maybe a better question would be comparing a 15s buff like Bloodlust Brooch, where you time it to get it on two Rips, to the Shard, which could potentially only benefit one Rip if it happens to proc after one Rip, then you get the benefit on Rip ~12s later and the buff is gone by your third Rip. For that to happen, it'd have to proc within 4-5s of putting Rip up. It's more likely to proc in the other 7-10s of your cycle and then you should be fine to get the buff on two Rips I think. Maybe if you had an internal CD timer you could fiddle a little (go for 5cp rather than Rip uptime if it's up?), but anything more than a minor delay would probably outweigh the benefit of getting the buff on two Rips I'd imagine.

Of course I can, but if I can time it myself rather than it being a proc, I can easily fit two Rips into that 20 seconds. With the proc, if it happens to proc right after I've used Rip I can't fit two in unless I change my cycle which diminishes DPS in other ways.

The timing to which he may be referring is using a trinket like [Berserker's Call] just before a Rip, going through a normal cycle, and having it active for a few more seconds including the next Rip. This is more benefit than would be captured in a model that simply describes the "on use" as a flat 60 AP over the entire damage time.

(Moreover, even assuming every activation is good for two Rips (side research question: does the benefit of getting a third Rip during the trinket's active time outweigh the loss of clipping one of them?) per 2 min provides a minimum value which will be exceeded whenever the fight ends before the trinket's cooldown expires. Lhivera's theorycraft-o-matic for mages takes fight length as an input for valuing abilities with various cooldowns.)

Considering the fact that the [Shard of Contempt]'s proc only has a 45s cooldown with a 10% proc rate, it should proc at least twice within the 2 minute cooldown of the [Bloodlust Brooch]. The controlability of click trinkets may be more desirable for fights with specific dps phases.

Ohh, good point. I completely forgot about the additional procs/minute of Shard over Brooch/Call.

Ok, my reply came amid a sea of others pointing out the same first thing. But I still wonder about both the second issue (whether the fact that <the value of a usable effect with benefit x once per time period y exceeds x/y unless the fight duration is an exact multiple of y> is modeled in various tools druids use) and the side question (whether it's worth clipping a Rip to get a third in during the bonus AP).

I noticed a HUGE increase in the TPS I was outputting last night. Its possible that this was due to the new version of Omen, but no one elses TPS was any different then it normally is.

While offtanking Gruul (using only mangle and lacerate) I was sustaining 1400-1500 TPS which resulted in me pulling aggro off our Warrior MT (despite him having 4 hunter chain MD to him) Once I pulled aggro off him, my TPS soared into the 1800+ range with spikes over 2k. Again, I'm not sure if this an issue with the new omen or if my TPS was actually that high.

After talking with Astrylian, we did some napkin math and came up with the tentative assumption that the new lacerate does ~12% more threat then the old. 12% more threat from "one" move shouldn't equate to a 50-80% TPS increase.

Aside from demo roar having no threat multiplier in the Threat 2.0 Lib, its exactly the same as the Threat 1.0 Lib. which "rules out" omen 2.0 as the culprit.

Anyone have any insight into this? Not that I'm complaining of course...

It's not like all the benefit of a trinket comes from Rip.

230ap from SoC, on a cat grouped with an Enh Shaman will increase the damage on each attack by:

Melee: 21.87
Mangle: 34.90
Shred: 63.98
Rip: 95.51

If SoC procs, it's going to get 20 melee attacks, 1 Mangle, 3 Shreds, 1 Rip, and either [another Mangle, another Shred], [another Shred, another Rip], or [another Rip, another Mangle]. (I think, really depends on subtle timing and energy procs, but I think that's usually right, correct me if I'm wrong)

So it's always going to do 20*21.87 + 1*34.90 + 3*63.98 + 1*95.51, plus either [34.90+63.98], [63.98+95.51], or [95.51+34.90]. Totals in those cases are 858.63, 919.24, and 890.16. That's only a 6.6% difference from worst case to best case. And Rawr averages it out so it's in the middle.

Sure, you could probably do about 3.3% more damage per proc, on average, if you could control when it procs. However, that small difference is made up for by the fact that you don't use on-use trinkets the instant they come up, you wait for the next perfect moment. Wait just 4sec after the cooldown is up on Bloodlust Brooch before using it, and you've already lost the 3.3% bonus you would get by perfectly timing it.

The T is constant, but is the time constant between omen1 and omen2? If Omen2 discounts non-active time that would easily do it.