How to Dragoon Per Second: A DPS Paradigm

[SunnyHirose]

I've moved defensive CDs to the third set. It's really messing with me. Think I'm gonna go with 1st/3rd set for toggling and use triggers for the 2nd (since that's what I normally do), but man that still throws a monkey wrench in my whole thing. I used to pop whichever one would help, especially in a clutch... but now the entire rotation is kinda clutch, hehe.

(EJ has plenty of offensive uses, so... you get to stay, li'l buddy :> )

[Jonnycbad]

Yeah, I do a backturn+EJ to move across the map to get to adds more quickly, also good for things like Odin, to EJ out of his AoE then Spineshatter back.

[Talec]

I really don't get those new skills. They have a lower Attack potency than full thurst beside being a higher tier? Geirskogul with only a attack potency of 200 doesnt look that amazing too.

Are they really gonna make us stronger?

[whiskeybravo]

I'm still having a hard time wrapping my head around 2 new, nearly identical abilities. They could've at least made one a dot or something.
>.<

[Neyka]

I really don't get those new skills. They have a lower Attack potency than full thurst beside being a higher tier? Geirskogul with only a attack potency of 200 doesnt look that amazing too.

Are they really gonna make us stronger?

For the true thrust combo, the average potency is 236.6 per gcd ((150 + 200 + 360)/3). Adding in an appropriately positioned fang/wheel raises the average potency per gcd to 250, so yes, it's a damage increase, not even counting extending blood of the dragon. Also, Geirskogul is an off-GCD attack, so it's basically free additional damage, and even more if you hit multiple targets with it. Factoring in half a Geirskogul raises that to about 275 potency per GCD, which is roughly a 16% increase in damage.

As an aside, the TT combo drops to 202.5 potency per GCD if you miss the positional, which may be a dps loss. If you assume that extending the timer for blood is worth half a Geirskogul, then you end up with an average potency of 227.5 per GCD (unless the timer allows you to get in a jump, but that's hard to account for).

[Jonnycbad]

Hope they ease things up a bit. I get the whole "get gud" thing, but these are a tall order to pull off without impeccable timing/positioning, and a good connection. I'm sure they'll adjust things soon, just gotta grit it for a few weeks till the complaints pile up. If they nerfed the HT positional, then nerfing the tight restrictions on our new abilities are nearly guaranteed. Would also like a formal response on how Jump animation was adjusted. The difference is nearly impercievable.

[Thendiel]

In this post, I want to compare a rotation that involves Disembowel and CT clipping:

HT - ID-Dis-CT-4 -Phleb - TT-VT-FT-4 - (repeat)

...to a rotation that does not involve Disembowel and CT clipping:

HT - ID-Dis-CT-4 -Phleb - TT-VT-FT-4 - HT -TT-VT-FT-4 - Phleb - ID-Dis-CT-4 - HT - TT-VT-FT-4 - Phleb - TT-VT-FT-4 - (repeat)

Basically, I was wondering if and when it would be useful to clip Dis and CT vs. letting them fall off for a few seconds.

On the one hand, if we clip, we maximize our ticks of CT, which weigh in at 35 x 1.15 raw potency per tick, plus extra damage from getting the Disembowel buff on Disembowel and/or Impulse Drive, which comes to 220 x 1.15 x 0.10 (possibly + 180 x 1.15 x 1.10) = 25.3 potency to 46 potency.

On the other hand, if we don’t clip, there is a loss of CT ticks, but with the advantage of favoring our TT-VT-FT-4 combo--which has higher initial potency values--over our ID-Dis-CT-4 combo.

The initial hit damage of our clipping approach is, (170 + 180 + 220 + 250 + 290 + 170 + 150 + 200 + 360 + 290) x 1.15 x 1.10 = 2884.2 potency across 10 GCDs.

The initial hit damage of our non-clipping approach is ((170 x 6) + ((180 + 220 + 250 + 290) x 2) + ((150 + 200 + 360 + 290) x 4)) x 1.15 x 1.10, or (1020 + 1880 + 4000) x 1.15 x 1.10 = 8728.5 potency across 30 GCDs, minus the loss of Disembowel’s bonus on Dis and/or ID’s hits. I’m guessing that, for typical levels of Heavensward Skill Speed, ID gets the Disembowel bonus but Dis itself does not. As we saw, that’s a loss of 25.3 potency per Disembowel, of which there are two. So, 8728.5 - (25.3 x 2) = 8677.9 potency across 30 GCDs.

SO. We should be able to construct an equation that tells us when clipping is or is not worth our time, based on Skill Speed:

((2884.2 potency / 10 GCDs) x (1 GCD / X seconds)) = [(8677.9 potency - ((40.25 potency / 3 seconds from an HT-buffed CT) x ((30 collective GCDs between CTs x (X seconds / 1 GCD)) - 60 seconds of CT uptime))) / 30 GCDs] x (1 GCD / X seconds)

In other words, we want to take the potency per second of the initial hits for clipping, then compare it to the potency per second of the initial hits for non-clipping minus the penalty we get for missed ticks of Chaos Thrust. Let’s solve the equation for X, which should tell us the threshold of GCD length underneath which NOT clipping Chaos Thrust (rotation #2 above) will be a benefit.

When I solve for the equation, I end up with an X-value of 2.0629 seconds. In other words, holy shit, we are never going to have enough Skill Speed for a non-clipping approach to be worth it. (A rudimentary check of the numbers does suggest that ID’s initial potency receives a Disembowel bonus at this GCD length, whereas Disembowel itself does not, which is consistent with my mathematical assumptions.)

Now, to beat a dead horse even further: keep in mind that Disembowel buffs our auto-attacks, in addition to the damage of any bard(s) or machinist(s) in our group. This means that those tiny pockets of downtime, even the ~1 second or so that we would have at a 2.13 second GCD, would cost even more than this equation suggests.

Conclusion: HT - ID-Dis-CT-4 - Phleb - TT-VT-FT-4 - (repeat) is officially our new GCD rotation.

[spelley]

Conclusion: HT - ID-Dis-CT-4 - Phleb - TT-VT-FT-4 - (repeat) is officially our new GCD rotation.

So our ridiculously long rotation was replaced with a much more concise one, just with more strict positioning requirements and a really "dense" opener with oGCD management being the primary complexity we have to deal with.

[Nominous]

Good info, dude. I sort of skimmed through, but you're saying that unless we have 2.1 or so GCD, we need to clip IDC even if it's at like 8-10 seconds remaining?

[Dervy]

Yup. That's exactly what I'm saying. The deficit of not having 10% DE is huge.