afaosfdbh

My issue was that the 2.4 number kept being tossed around as ideal. It was very confusing for me when I first read it because I thought someone had figured out the value of skillspeed. I think we're putting the cart before the horse when we are theory crafting for armour that will be obsolete in 2 weeks and every couple patches after that. All this theory crafting is operating under the assumption that 2.4 seconds is ideal when nothing has been established. We're basically theorycrafting for something that hasn't proven to be superior. Perhaps we should focus on proving its superiority before building upon it as our base?

100% HT/PH uptime is overrated. It's not a bad thing but putting it as your goal just doesn't make sense. +15% on HT is 25.5 potency. PH seems to drop for a fraction of a second and equal less than 10 potency.  It's roughly ~35 potency and it's hardly a game changer... roughly 1% increase in damage (considering our 10 skill rotation has ~2850 potency in just GCDs alone). And it forces further clipping of CT not that clipping is inherently a bad thing so much as the fact that you were using CT combo which had a face value of 650 instead of FT combo 710 (that and the overwriting of triple buffed CT)

While it's *nice* to have 100% uptime on these skills, the potency gains on them seems very small for how much skillspeed you need. It's sort of like pre-patch where PH would drop for 2 seconds per rotation and we had 0 grace period for any of our DoTs. Any slipups and we suffered dropped DoT's in CT and Phb. Skillspeed would have been nice there, but it wasn't as nice as DET or Crit and so we stuck with our potential downtimes as well despite the fact that a simple 1-2-3-4-5-7-8 combo might have done more dps due to the grace period granted to CT even though the full rotation netted maybe a 60 potency difference over the other (1 extra FT cast instead of 1 CT cast, both had same uptime on DoTs and heavy thrust) over a full minute rotation.

also my numbers for the leeway say that we actually have a lot more leeway to get multiple GK casts per 3 minutes even with a 2.5 second GCD. You should take into account that oGCD abilities are cast inbetween GCD abilities meaning that Disembowel cast 5 seconds into a rotation means a BotD cast 6.5 seconds into the rotation. according to my numbers you have a 4 second leeway. Simply because the second GK you only need to make sure there are 20 seconds left on the buff instead of 22.5. 20 seconds because 10 second cost + 10 second cooldown on GK to recast (instead of 5 GCDs to hit the refresh). Whereas the 22.5 number is the number you need to hit in order to refresh and make sure GK doesn't drop (12.5 seconds to cast 5 skills for a refresh and 10 seconds for the GK cast). while im not saying my math is foolproof, perhaps there is an error in your math where you are forcing the dragoon to cast GK on combos 1/3/5 where casting on combos 1/4/5 is just as reasonable as far as DPS goes?

another thing, while it might be advantageous to cast GK on combos #1, #3, and #5... there are times where it might be better to cast it on combos 1/4/5. There is no punishment for late GKs so long as you get your 3/min but it does give you more of a cushion for mistakes if you delay the second cast by 1 combo.

my personal rule for GK is this: to get 3 casts in a minute all you have to do is watch for the second one.... That's the key. If there's 20 seconds left on the buff post refresh you will be able to cast 2 more geirskoguls so long as you cast them the moment they're up. The number is 22.5 seconds for a refresh, 20 seconds for a quick double cast (10 seconds to cast GK, 10 seconds to have GK come off CD). Maybe 21 seconds to give yourself leeway.

buttttt if you are on a time frame of less than 60 seconds (boss about to die, dps check, need burst) moving up gk's earlier might be a good idea. otherwise there are cases where delaying GK is advantageous because like i've mentioned GK's casts/min is reliant on BoTD and not if you can cast it the minute it comes up (like every other oGCD ability). check my numbers in the next post, perhaps there's an error