How to Dragoon per Second: A DPS Paradigm

[aleph_null]

I don't think dragon sight can be double weaved reliably since you need to either macro it or manually target your party member.

[Umbrax]

Thank you Erys, that's actually pretty interesting info, I already like the opener you wrote.

About Dragon Sight being double-weaved, it's going to be tricky in someway but it's totally doable imo. If you macro it to a designated party member, I don't think you can double weave it but, if you set a keybind for the desired player, let's say your melee partner, it's just a key press more before you use Dragon Sight. It can take some practice to some players tho.

I don't think half a second without the boss being targeted affects auto-attacks? maybe it does, I'm not an expert in that matter sadly so if anyone is, please don't hesitate to correct me!

About the rotation itself, I rather wait and see the potencies on live servers before claiming an optimal rotation, we could be missing something (most likely because a lot of tooltips are conflicting with the Japanese version of themselves). Don't take me wrong tho, I really appreciate the effort.

Also I quote @patefoisgras on reddit

Dropping a 4th/5th tier in favor of a full Full Thrust combo is a tiny bit of a DPS loss, but the reward is that there's no positionals. Basically it's a moment-to-moment choice, if you can get 4th/5th tier positional off, go for it, if not, skip it and redo Full Thrust.

Sounds interesting even tho we have True North now.

[Nora_of_Mira]

Hello, fellow goons.

the only thing I see missing is considering that Wheeling thrust and Fang n Claw now combo off each other, making our combo's 5 GCDs long.

[KaiSunstrider]

the only thing I see missing is considering that Wheeling thrust and Fang n Claw now combo off each other, making our combo's 5 GCDs long.

This is correct, FnC and WT can now be used instantly after the other. With the change of these abilites only adding 10 secs to the BotD timer, the 20 sec addition for 2 GCDs will come in handy. Although, we may run into cases where the extra FnC/WT would give over cap durations which means unless the extra 290 potency outweighs starting over a TTT combo, we may have to adjust when and where we double up. At 290 potency Im going to assume atm that we will almost always use the 5 combo rotation regardless of timer.

[Erys]

About Dragon Sight being double-weaved, it's going to be tricky in someway but it's totally doable imo. If you macro it to a designated party member, I don't think you can double weave it but, if you set a keybind for the desired player, let's say your melee partner, it's just a key press more before you use Dragon Sight. It can take some practice to some players tho.

About the rotation itself, I rather wait and see the potencies on live servers before claiming an optimal rotation, we could be missing something (most likely because a lot of tooltips are conflicting with the Japanese version of themselves). Don't take me wrong tho, I really appreciate the effort.

About Dragon Sight I was also concerned with the chance of having to use a macro and get that delay or going the hard route and pressing buttons at the speed of light. But it was the best place that I thought of to buff the most of our opener. I also thought of focus target your designed party member instead of bosses and focus target the skill, don't know how that would work out for less buttons press, specially for me since I'm on ps4 x'D

On the potencies topic, I used the potencies from famitsu. I don't remember if we can post links but I will refrain until you confirm if we can, then I'll edit this with the link. Since this source has the most updated potencies it's safe to assume double Full Thrust combo is superior to 5 step combo strings. Math for this:
Ht-IDC4-TTT4-TTT4-Ht: this combo nets 3920 potency (180+1040+290+830+290+830+290)
While Ht-IDC44-TTT44-Ht-ID: again 14 gcds with a net potency of 3819 (180+1040*+290+290+830+290+290+180+190+130)
*[this one doesn't fully tick, it has 2 ticks left, miserable difference but still]
The difference is 101 potency, which we all know it might seem tiny but it's a gain over time anyways.

The only use I see for back to back refresh hits is in preparation for boss mechanics where we need to disengage, but since we will have 24s in our counter before the Nastrond phase this means we have 2gcds of leeway before re-engaging the boss and we wouldn't need that double refresh. Also I'm taking into account that thing they said about they don't want to force players to disengage as much, because that made losing rotations higher chance and what not, so I'm expecting more uptime in fights but dealing with more complex mechanics in the meantime.
I'd also like to note the potential of Mirage Dive having range and a 15s proc for disengages, to weave it in between piercing talons. We will have a nice extra tool to deal with those situations of dps tanking because of piercing talon spam.

If I forgot to adress something else, let me know.

[Dacen_Drg]

Since this source has the most updated potencies it's safe to assume double Full Thrust combo is superior to 5 step combo strings. Math for this:
Ht-IDC4-TTT4-TTT4-Ht: this combo nets 3920 potency (180+1040+830+290+830+290)
While Ht-IDC44-TTT44-Ht-ID: again 14 gcds with a net potency of 3819 (180+1040*+290+290+830+290+290+180+190+130)
*[this one doesn't fully tick, it has 2 ticks left, miserable difference but still]
The difference is 101 potency, which we all know it might seem tiny but it's a gain over time anyways.

Unless I am looking at this wrong, for the double Full Thrust combo rotation you forgot the 290 for fourth hit after Chaos Thrust, which would therefore make it worth even more. So this would therefore be very interesting to look into more. The issue I do see is that it is one less "4th" hit (we need new name) which means 10s lost on BotD which might mean that when we enter LotD it won't be at 30s? which is not ideal.

[Erys]

snip

Sorry for that I will edit it at once, the potency output is correct. I just forgot to write the potency in the string, the total is correct, my bad.
As for BotD timer, let me edit the first post quick and I will edit this one explaining BotD timer.

Assuming a 2.4 gcd:
Ht-IDC4-TTT4-TTT4-Ht
We activate BotD right before Impulse Drive, with 15s on the counter that means we are going through 3 gcds before our first refresh. That'd be 2.4x3s=7.2s that we are losing, then we refresh with 10 seconds 15-7.2s=7.8+10=17.8seconds.
Now we go through another 3 gcds for our second refresh 17.8-7.2=10.6+10=20.6 seconds.
Another 3gcds and we are at 23.4seconds on out timer.
Then we repeat our rotation and this time we go through 4 gcds before refreshing, 23.4-9.6=13.8+10=23.8 seconds.

The calculations I did were based on the assumption that we'd have to go through 4 gcds everytime, as you can see with the last refresh being only a 0.4 we gain against the timer. Therefore we have even more leeway than I said, but again, this would have to take into account mechanics; so putting 1 more gcd per combo to represent that possibility of mechanics messing up with our thing - +, while not correct (because I can't math that much, duh) it's the best I can come up to give us a "realistic" scenario of what to expect towards our timer.

In conclusion when approaching Nastrond phase we should be above 20seconds, give or take. We would Geirskogul after a refresh hit which would make us hit 30 seconds and nastrond right after, trying to get those 4 Nastronds. If 4 Nastronds is not possible, any amount of time above 20 will allow us to hit 3 Nastronds all of them under blood for blood which will be 960 potency anyways.
If the mechanics incoming will drain a great deal of BotD timer, we will have to deviate from our new rotation and double refresh. Same if when approaching nastrond and we won't have 20+s after refreshing once, which will be hard, we just use double refresh before Geirskoguling into Life.

[SunnyHirose]

That is an inappropriate basis of comparison, though. You can't just give potencies with a hard cutoff and call it a rotation, because that's not the average GCD value you are calculating. Please respect the loop!

Anyway, I for one find it curious that SE would "bake in" the ~5% difference for most skills but not Wheeling/Fang. The balance was precarious before, and mostly justified by BotD accounting for tons of damage (if you may recall that brief period in 3.0 where the non-positional potency was 100... they were still worth using from the front, barely, but if it’s visibly weak and you need some sophisticated theory work to prove it... so yeah, going to 200 was much appreciated). It rubs against the grain of basically all their older job designs to make skipping the ends of combos a potentially DPS-optimal option and I will find it a bit surprising if it lasts past 4.1.

(Yes, I realize how some--or a non-negative number--of you must feel: right when the world needed that ornery math-hat the most, I vanished. Lo, I am not dead, just burnt out and struggling to keep abreast of standard developments, much less high-falutin’ guesswork theory. Haven’t even checked my Reddit messages in months; if you have sent one or intend to, I intend to reply when I can, but that will likely not happen soon. Still, it is only fair I give notice.)

[Erys]

That is an inappropriate basis of comparison, though. You can't just give potencies with a hard cutoff and call it a rotation, because that's not the average GCD value you are calculating. Please respect the loop!

If there's a method I can only ask one thing... teach me senpai? So I can be more independent and participate, very please? xP

[Deathgiver]

As far as the loop thing it means (as far as looking at purely gcds) that you have to start at ht and end at the last skill before the next ht, then divide by the number of skills. So, using your example with a 2.4 sec gcd and dot calculated separately:

HT-IDC4-TTT4-TTT4 = (180+690+290+830+290+830+290+350 dot) = 3750 potency over 13 gcds for an average of 288.4615 potency/gcd

HT-IDC44-TTT44 = (180+690+290+290+830+290+290+308 dot) = 3168 potency over 11 gcds for an average of 288 potency.

You also need to consider how well each rotation lines up with your oGCD buffs. For example, with the 11 hit combo, you can use Life surge on every second Full Thrust with only minimal activation delay. Additionally, the 13 hit combo actually takes 31.2 seconds to execute at 2.4 sec GCD, meaning you have 1.2 seconds of time where your disembowel falls off and 1.2 secs where heavy thrust falls off. Even if you're executing the gcd before the debuff falls off, that means that roughly 7.7% of your autos and possibly some of your oGCDS aren't being buffed by one or the other.

Without looking too deeply into how all of the oGCDs line up, I'd say that the 0.16% dps loss from the double finisher combo is worth it for the extra safety and the slight increase to AA damage.

Edit: Also to add, as GCDs get faster, the double TTT rotation actually pulls ahead a little further, so once you get to the point where gcds are lower than 2.31 secs (the point where the rotation takes 30 seconds or less) the double TTT might be the stronger option depending on latency.