We've been using it on Heroic LK during Valk phases for extra swipes.

Swipe till no energy, GotW Macro, where the first press casts GotW on the raid, and then the second press shifts back to cat form. Swipe once with the OOC proc from GotW, then swipe again with the energy you've regenerated, repeat. It is excellent.

It is quite useful anywhere you don't have enough targets to keep constant OOC procs flowing, assuming it is still beneficial to swipe.

Assuming perfect reaction time, you'll lose 4 melee hits at the least (melee hits with buffs and ICC-level gear are around .7 speed). The OoC proc loss is obviously a wash. Assuming you're doing it below x-30 energy, where x is how much furor you have (so with 3 points you should do it below 30 energy) it should be a gain provided that you get that OoC proc. Which isn't guaranteed.

It's probably viable, but the actual gain is very slight. It almost certainly isn't a good choice in 10-mans where the chance of an OoC proc off of GoTW is closer to 40-50%.

In cataclysm the choice is tremendously worse given the emphasis of savage roar on damaging autoattacks only vs. specials, at least right now.

You're spending one GCD outside of cat form; that's about 1.25 seconds assuming ~12% spell haste (~16% melee haste) and the 5% and 3% raid buffs. However, this isn't the bottleneck for autoattacks; the bottleneck is your weapon attack speed outside of cat form. GotW resets the swing timer, and you're now in caster form, so a caster form attack timer begins. A staff generally has base attack duration ~2.4 seconds; a polearm generally has base attack duration ~3.4 seconds. That's about 1.7 and 2.4 seconds respectively with 16% haste and the 20% and 3% raid buffs (the amount required for a 0.7 second swing timer in cat form). Assuming a cat form attack duration of .7 seconds, you're losing on average .5 + 1.7/.7 - 1 = 1.9 attacks with a staff, or .5 + 2.4/.7 - 1 = 2.9 attacks with a polearm. I add 0.5 because on average you'll interrupt the attack timer halfway; I subtract 1 because we've counted the full time between two successive autoattacks, which would have occurred at the normal separation (not 0 separation) otherwise. Of course, this conclusion is actually independent of haste, since we're just taking the ratio of two hasted attack durations.

Conclusions: You lose on average <weapon speed> - .5 (minimum <weapon speed> - 1) autoattacks per GotW, where <weapon speed> is the listed "speed" (really duration) of your equipped weapon. For example, a typical staff might cost you 1.9 autoattacks per GotW while a typical polearm might cost you 2.9 autoattacks per GotW.

Assumptions: Casting GotW resets the caster form swing timer, which persists into cat form. I have tested this fairly rigorously and am confident it is true. Here is a combat log excerpt I produced just now on a training dummy to demonstrate this:


Edit:

Here is a primitive analysis of the viability of using GotW in a single-target rotation, in my gear. I am using Distant Land, which has a 2.4 second attack duration, so I lose on average 1.9 attacks per GotW. If GotW hits 28 targets (25 players, 3 pets - probably a low estimate for a 25-man raid), then GotW has a 1-(1-3.5/60)^28 = 81.4% chance to proc clearcasting. I will ignore combo points and assume the clearcast is used to shred. Hence, casting GotW is worth .814(shred damage) - 1.9[(melee damage)+3.5/60(shred damage)]. According to Rawr, my shred does 14926 average damage per swing while my melee does 3315 average damage per swing. Thus, GotW is worth 4197 damage in this estimate.

If I were using Oathbinder (3.6 second attack duration), then I'd lose 3.1 attacks per GotW and in this case each GotW would be worth -826 damage in this estimate -- a loss.

I've been trying to avoid all Cataclysm discussion and 4.0.1 discussion for as long as possible as to not be spoiled, but I guess with the impending release of patch 4.0.1, it's time to start looking at rotations.

Just looking at the new talents, It seems that Ravage will finally have a use on our bars. Just from my initial glance, this appears to the be the DPS spec I'd use at 80

WoW Talent Calculator - Sigrie

Ravage would appear to be our opener now, with a solid 100% crit (Assuming 50% crit from gear, perhaps a bit lower at entry level 85 gear), and also Ravage would be usable on every Feral Charge. So in ICC, I'd be able to Ravage multiple times on every fight with the exception of Saurfang and Festergut.

While that part seems solid, one change that absolutely baffles me is the Endless Carnage talent. I'd consider our best set bonuses throughout this expansion being 4pc T10, 4pc T8, and 2pc T9. This new talent seems to be an essential combination of 4pc T8 and 2pc T9, only buffed versions of them. My question is, will this overly simplify our rotation like I think it will? At crit cap in heroic ICC gear, it just seems like they're removing the variables from our rotation with this talent.

I'm not sure how it will work out at 85 with scaling, but for the 2 or so months we have left to play in ICC, it just seems like that our rotation will be spamming Ferocious Bites while barely having to worry about keeping up Savage Roar or Rake as much due to the increased timers. I remember back in Ulduar, part of the reason DPS felt so easy back then was due to our 4pc set bonus. That extra 8 seconds turns a 1-2 CP SR into 3-4 CP SR. Correct me if I'm wrong, but will our DPS rotation essentially remain the same, only with the ability to use Ferocious Bite in between almost every refresh of Savage Roar and the usage of Ravage on Feral Charges?

I actually meant to take Feral Aggression instead of Brutal Impact. There aren't many fights in ICC requiring interrupts and at this point it's just fun to see how much damage i can squeeze out as raid utility isn't very important at the moment.

I would use Feral Agression instead of Brutal Impact for pve purposes. And yeah Ferocious Bite will see it's way through our rotation a bit more than it has before.

Brutal Impact is there so that we can function as an interrupter, as 2/2 gives us a 10 second cooldown interrupt, making us as useful as the other melee dps classes in that department. I'd consider it a nearly mandatory talent.

Predatory Strikes gives you energy when you want it - when you're doing extra damage from TF or when you're operating at half energy costs from Berserk. I agree it's a minor damage buff, but it's certainly not the only talent in that category. Predatory Strikes is barely useful in a pve environment, often coming up once per combat unless you have regular adds (and even there, you may well prefer to start with mangle/rake, and then it's below 80%). 2/2 Stampede also strikes me as weak over 1/2, as it's a minor energy reduction you aren't likely to get much on most fights.

As kalbear points out above, there's almost no reason to take Natural Shape Shifter prior to the expansion, as you can't get Master Shapeshifter until 81 at the earliest.

Think of it as redundancy. Sometimes our rogues get stupid on LDW and die to the ghosts. Sometimes they are on vacation or watching a football game (or watching a football game while we are doing LDW...). The new Brutal Impact + Skull Bash is huge for a feral - providing a utility we did not have before both as a cat and as a bear.

I am very dubious about the overall value of Stampede and Predatory Strikes on a lot of the fights in PVE. It is 4 talent points for a single attack damage increase once every 30 seconds. Also, until they adjust our damage coefficients we don't really know how much better it will be than our other attacks. We may find that opening with mangle is more important in a raid environment even when we use feral charge to get in position.

An interesting question for our openings is whether we should get to 5 combo points, TF, RiP, refresh Rake and then use one or more combo points to get SR up or get SR up first. Lining up the timers - TF 30s, Rip 16s or 22s with Glyph of Shred, Rake 15s, SR 22-42s. It looks like we may not want to use Glyph of Shred, and instead line up every other Rake and Rip to occur after TF. There is also a question of using the Glyph of TF - but taking 3 seconds off of TF will make it harder to sync it with Rip and Rake. With mastery and the Glyph of Rip, lining up TF with Rip and Rake may take precedence over everything else in the rotation. The rules for when to FB will probably be more restrictive now and we may find the only time we use FB is during the execute phase. Current FB use is predicated on the idea that some downtime on Rip is acceptable.

Here is another interesting question, if we refresh a Rip with FB during the execute phase, if the Rip was cast during TF does it retain the 15% extra damage?

I agree it's a good or maybe great extra bit of utility, I just don't believe it to be mandatory when a number of others bring it for free.

Yeah it will be most likely mandatory for 85 raiding. But seeing as we got away with it for the time being we can do without for the less than 2 months left raiding at 80.

Jaymz, healers are going to struggle for a while, getting used to the new mechanics, etc. I'll take the +20% buff in healing done to me over something else, personally.