No, it doesn't ignore armor on weapons, nor should it. The armor on weapons will remain, it just won't get multiplied.

I don't have any reliable data on Berserking, yet. Going just on wowhead comments on its proc rate (ie, take with a huge grain of salt), it has the same proc rate as Mongoose. That would make the 3 enchants provide, in optimal Naxx25 gear, about... Massacre=59, Mongoose=60, Berserking=76. Again, those are really rough numbers, based on rumors and here-say, so take them for what you will.

That said, even if Berserking is better, I wouldn't put it on Staff of Trickery. It costs 4-5k gold in mats on most servers, whereas Mongoose costs like 200g. Unless you think you'll never do Naxx10 or 25, I wouldn't drop that kind of gold on the VH staff, personally.

Thanks. I just saw Ghostcrawler's post on that today and was going to come back and remove my question. For those interested, here is the source

I know a few people have gotten close to this idea, but I don't think it's been suggested outright yet. Can/should Rawr.Bear and other tanking modules have a cap for Survivability? I know it's pretty well established that you need X amount of health and Y amount of damage reduction for a particular encounter, after which you should stack mitigation (and threat/dps) because more survivability isn't needed unless your healers die / fall asleep. The current method of deciding you have enough survivability and switching to Mitigation based sorting works, but throwing a cap on total Survivability score would make it more intuitive, allow Overall to sort by both Mitigation and Threat scores, and still accurately penalize the score of an item that would reduce you below your target Survivability.

It would vary by fight like boss armor / fight time / etc, of course, so it'd be an entry in the Options tab, and probably default to turned off. It could have a pull down with named defaults like Threat has, but based on either encounter or raid level (X for Heroics, Y for 10-man Naxx, Z for 25-man Naxx, and onward). I think it would be a pretty useful option to have.

I'd propose that you do something like a diminishing returns value with a soft and hard cap. Say, set the soft cap to 40k and the hard cap to 45k (user-changeable) and that each point above 40k got worse and worse until hitting the 45k cap, at which point it does become useless.

I'd also strongly recommend this be the default no matter what you end up doing, so that people using Rawr don't think that it's recommending huge stamina right out of the box.

Agreed. As is, I think Rawr.Bear values items quite well, assuming you're fighting challenging content... there just happens to be no such content available at the moment.

Awesome, I'm looking forward to it. The soft cap does sound even better than a hard cap, too, since additional health isn't entirely useless, just unlikely to come into play in most situations.

I originally had the idea of a health cap, which would let armor continue to provide both Survivability and Mitigation score (though the Survivability contribution would be based on the capped health), retaining most of its value even after you had "enough" health. But the survivability contribution of armor really isn't any more valuable than the survivability contribution of health after the cap, it's used entirely for mitigation, so yeah, a survivability cap is the way to go. Plus, the amount of health necessary for a worst case scenario does vary based on your damage reduction (the thought brings to mind a mage trying to tank Gruul).

I can agree with kalbear's suggestion on the default state. It's a balance between two slightly confusing situations ("Why are the Survivability values of all 20 of these gear pieces exactly the same?" vs "I stacked 60,000 health because Rawr kept rating the stamina items the highest, but our healers say I'm the hardest tank in the raid to heal"), but I'm all for defaulting to the more correct of the two.

The thing is, it is 0 at high enough levels. If you never are down more than 45k, no matter what, that 15k health is worthless. It also means that health becomes less valuable as you approach that cap, so that you can get the nice effect where health is valuable to the soft cap, then 'nice to have' as you get to the hard cap, then 'totally worthless' past that point.

Of course, you can make the hard cap infinitely large and get the same effect.

I don't know enough about how the Bear model works to say whether or not this would be worth the coding required for it, though. But you know end users - they always want a ton of features even though what they really want is simplicity.

Any plan for a Deep disc priest model ?

There already is one. The Holy Priest model really is for all healing priests. (And the Shadow Priest model is for all DPS priests. Yes, that means smite priests)

Would it be possible to add an option for ignoring the hit/expertise cap if you just want to check an item with these stats?