The Ninja Superthread

There's no plateau for DEX, the more of it you have the better as it is NIN's primary stat. Potentially, in certain cases (maybe some of the Arachne), the primary+secondary totals on a particular piece (say, an over-melded craft item) can out-weigh a higher iLvl piece, but there isn't a given DEX cap where you can start stacking in favor of the secondaries and completely ignore more DEX.

Quote:

---

Originally Posted by Viridiana

http://ffxiv.ariyala.com/OX09 technically beats the other one by ~0.044 DEX weight, but that's at the cost of having 40 more Skill Speed, which is not really preferable.

---

Went back and played with the settings. Realized that this setup doesn't actually need Flint Caviar and can work just fine with Apkallu Omelets, with the same stats.

Quote:

---

Originally Posted by treuhavik

FFXIV Version 2.45
[*]If you have a Warrior in your party and they are using Storm's Eye with any consistancy, you can drop Dancing Edge in favor of Aeolian Edge. You may also drop Dancing Edge in favor of Aeolian Edge if you have another Ninja in your party and they are using Dancing Edge with consistency.

---

Changed that.

Now to add some other tid-bits.


DoT effects:

DoT’s on ninjas work a little “funny” in this game. Buffs such as the Kisses and Blood for Blood directly affect DoT damage. Dancing Edge’s slashing bonus does not increase the damage of the DoT effect, regardless of if it is applied before or after DE’s application, but it does increase the damage of the application strike on skills such as Shadow Fang and Mutilate. The tick count is as follows: [(DoT Effect Duration)/3].

Disclaimer: DoT's work on a "server-wide tick." This means that while the tick count of the skill does not change, the point in the spell where it occurs at may fluctuate (e.g. first ping in Mutilate occurs at 28 seconds rather than 27 seconds). The interval between ticks stays a constant three seconds, however.

Let’s use Shadow Fang, Mutilate, and Doton as our examples below:

Shadow Fang:

The DoT effect occurs once every 3 seconds. There are six ticks. An example visual below:

200 (18s) -> 40 (15s) -> 40 (12s) -> 40 (9s) -> 40 (6s) -> 40 (3s) -> 40 (0s)

Mutilate:

The DoT effect occurs once every 3 seconds. There are 10 ticks. An example visual below:

60 (30s) -> 30 (27s) -> 30 (24s) -> 30 (21s) -> 30 (18s) -> 30 (15s) -> 30 (12s) -> 30 (9s) -> 30 (6s) -> 30 (3s) -> 30 (0s)

Doton:

The DoT effect occurs once every 3 seconds. Note: Skill has duration of 24 seconds. There are eight ticks. The Heavy effect is applied upon the first tick and lasts for three seconds after the last tick. An example visual below:

30 [Heavy on] (22s) -> 30 (19s) -> 30 (16s) -> 30 (13s) -> 30 (10s) -> 30 (7s) -> 30 (4s) -> 30 (1s) -> 0 [Heavy off] (-2s)



Another "water under the bridge" topic concerns Goad usage:

Goad!

Is what I would like to say to get my fellow party members into action as a ninja. And you can too! Just... well, you cannot Goad yourself to action. That’s why many ninjas will choose Invigorate as a cross-class skill to make up for that. This leaves us with an important question: When do you Goad?

Let’s put this into a Do, Don’t, and Please Don’t format:

Do:

- Cast Goad on a Warrior if he/she is spamming Overpower (Animation looks like an Austin Power’s karate chop with an axe that spews enough flames to make even Ifrit proud).

- Cast Goad on a Warrior after some time fighting a mob, or pulling a large mob right after you eliminate a group.

- Cast Goad on a TP-based ranged/melee class only on request from them. Any other time it is used on them would fall under the “Don’t” section.

Don’t:

- Cast Goad on a Monk, Dragoon, Bard, or Ninja if a warrior tank needs it instead. Monks and Dragoons can use Invigorate, Bards have Army’s Paeon to restore their TP, and Ninjas can cross-class into Invigorate. Additionally, Paladins won’t need to be Goaded as they can use MP-based enmity generating skills.

Please don’t:

- Cast Goad on any class that uses MP as its primary resource. Big hint for what a MP class is: Someone with a book or staff in their hands. Don’t do it; save yourself the ridicule.

Err, dot damage occurs every three seconds game time on a server-wide tick that has nothing to do with how long is left on the dot. You're just as likely to see Shadow Fang ticking at 17 or 16 seconds as at 15. Same thing with Mutilate and Doton.

Just curious, not a ninja but I'm a warrior ot still going up the fcob turns (t12 atm) and back when my nin is still nin (went back to monk this week due to composition issues between dps apparently), I used to apply the slashing debuff for him. I try my best but it's not always 100% up all the time (mostly just during add spawns) but it worked out fine, I think. Then a couple weeks ago he asked me to stop doing it because apparently he read somewhere it's better for both of us if he applies it for me instead. He mentioned the reason over voip but I didnt hear it clearly and we were already starting a pull and I didn't get to ask him again (forgot).

Anyway, that sounded odd to me. What's the deal?

I think the reason why he told u not to debuff anymore is prolly due to TP usage. That's what I think.

Quote:

---

Originally Posted by Butcherboy

I think the reason why he told u not to debuff anymore is prolly due to TP usage. That's what I think.

---

Maintaining the debuff would be less TP usage than using Aeolian.

DE = 50 tp right?
AE = 60 now
SF = 70 now

If it is then DE + SF + AE = 180 tp total.
SF + AE + AE = 190 tp total.

This can snowball over the course of a fight. Especially if you have a brd that doesn't wanna play paeon.

That might be why. We didnt have a full time brd. Our original was nin mnk smn blm. Now we are mnk brd smn blm. Mnk went brd and nin went mnk.

Thanks guys.

Quote:

---

Originally Posted by Viridiana

Err, dot damage occurs every three seconds game time on a server-wide tick that has nothing to do with how long is left on the dot. You're just as likely to see Shadow Fang ticking at 17 or 16 seconds as at 15. Same thing with Mutilate and Doton.

---

Just re-tested this statement. Yep; I'll have to edit it in there. Thank you for catching that.

Nevertheless, the number of ticks that occur (i.e. DoT duration effect/3 = tick count) stays constant regardless of where it occurs at in the server-wide tick.

Quote:

---

Originally Posted by Viridiana

Maintaining the debuff would be less TP usage than using Aeolian.

---

I mean that for tank to stop applying debuff. Ninja doing the debuff is less TP used

Quote:

---

Originally Posted by Butcherboy

I mean that for tank to stop applying debuff. Ninja doing the debuff is less TP used

---

Yeah, somehow I got really confused reading your comment and didn't realize that we were saying the same thing, sorry. @_@

Edit: Just noticed something on Ariyala's. . .can anyone with a new Zodiak weapon confirm if the minor stat caps are actually the same as the Dreadwyrm weapon despite a 10 ilvl difference?

Quote:

---

Originally Posted by Viridiana

Yeah, somehow I got really confused reading your comment and didn't realize that we were saying the same thing, sorry. @_@

Edit: Just noticed something on Ariyala's. . .can anyone with a new Zodiak weapon confirm if the minor stat caps are actually the same as the Dreadwyrm weapon despite a 10 ilvl difference?

---

Ariyala's is just really messed up for relic. Perfect Sasuke's is 49 crit 35 det. Got mine today! :D

Quote:

---

Originally Posted by Best-MNK-NA

Ariyala's is just really messed up for relic. Perfect Sasuke's is 49 crit 35 det. Got mine today! :D

---

Okay, that makes way more sense. It's not enough to sway anything either way, because WD is king, but I was looking at it like wtf?

Quote:

---

Originally Posted by Best-MNK-NA

Ariyala's is just really messed up for relic. Perfect Sasuke's is 49 crit 35 det. Got mine today! :D

---

If you want to be technical DET/CRIT relic is actually not the best relic for NIN, DET/ACC is.

Posted below is stats calculations with food.

DET/ACC relic 477crit(107.802)/406det(131.95)/385ss(82.775) = 322.527 Dex

DET/CRIT relic 483crit(109.158)/392det(127.4)/398ss(85.57) = 322.128 Dex

WD: 9.132
DEX: 1
DET: 0.325
CRT: 0.226
SS: 0.215

So not only is DET/ACC *SLIGHTLY* better, it is WAY cheaper.

But that is with current gear, depending on what Odin drops, along with other items from 2.5 this is subject to change.

Thanks for sharing that, Leggerless! OP updated with "DoT Effects" and note on NIN/DE added.

Quote:

---

Originally Posted by Ephier

DET/ACC relic 477crit(107.802)/406det(131.95)/385ss(82.775) = 322.527 Dex

DET/CRIT relic 483crit(109.158)/392det(127.4)/398ss(85.57) = 322.128 Dex

---

Acc/Det: 572crit(129.272)/374det(121.55)/482ss(103.63) = 354.452 Dex

Crit/Det: 567crit(128.142)/393det(127.725)/464ss(99.76) = 355.627 Dex

Those being much closer to actual "relic BIS". Crit/Det still wins.

Quote:

---

Originally Posted by Viridiana

Acc/Det: 572crit(129.272)/374det(121.55)/482ss(103.63) = 354.452 Dex

Crit/Det: 567crit(128.142)/393det(127.725)/464ss(99.76) = 355.627 Dex

Those being much closer to actual "relic BIS". Crit/Det still wins.

---

Nevermind I see it. hyperlinks weren't showing on phone.

i got bored and did tests

2.00 GCD exactly, 30 fps video. tests in estate area, mudra speed was pretty good, about as fast as this video (https://www.youtube.com/watch?v=NqutYpkQqmg)

1 mudra->ninjutsu (fuma) - 0 frames clipped, GCD goes off as soon as it's up
2 mudra->ninjutsu (raiton) - 14 frames clipped after GCD goes off CD; about 0.467s of gcd clipping
3 mudra->ninjutsu (huton) - 35 (!) frames clipped after GCD goes off cd; about 1.167s of gcd clipping

edit: over 3 mins, doing this rotation for ninjutsu would be 9.337s of GCD clipped by ninjutsu. that's assuming you get the first seal off as soon as your GCD's animation lock is over always, and that the mudra speed stays consistent throughout.

suiton>raiton>raiton>huton>
suiton>raiton>huton>
suiton>raiton>raiton>huton

if you replace the non-kassatsu raitons with fuma shuriken, you save 1.401s of gcd, losing about 129.6 potency in exchange. if you replace all of them, you save 2.335s of gcd, losing 259.2 potency or so. shoddy math warning.

That'll actually help as now I have to rework the Ninja model slightly.

Can you replicate those numbers constantly, or do they change each time you do your rotation? Also, what's your latency?

they're very consistent for the fps range. if you're consistently over 60 fps during recording, recording at 60 fps would be more exact; my frame estimates should be within -1 or +1 of the number stated for my personal mudra speed.

if anyone wants to try and see how their own mudra speed/ninjutsu gcd clipping compares

record @ 30 fps (or 60).
put huton up, try to get 2.00 skillspeed (since it's just an easier number to work with when it comes to allotting gcd to frames. not required; just divide your huton gcd by 60 for gcd per frame for 30fps)

do gcd>any ninjutsu>gcd; as soon as your first gcd goes off cooldown, play the video frame by frame (ctrl and -> arrow on MPC HC) until your gcd gets used again (skill icon goes entirely black) the amount of frames from your gcd going off cd to your gcd getting used is how much your ninjutsus clip into your gcd in frames; 1 frame=0.033s.

latency via resource monitor is like 90ms or so. pinging my server (behemoth) i get 55ms~

It must be nice to be able to play with so little latency and swing Raitons inbetween GCD with only 0.93 clipping like that even at 2.00 GCD :|
*envies, envies a lot*

Just out of curiosity, can you try fitting a Shuriken inbetween GCD and see if it comes out before GCD refreshes? (aka, no clipping at all) So then I can envy you even more :U

under huton, gcd>mudra>ninjutsu (shuriken)>gcd will not clip the gcd at all. didn't for me at least.

it's actually 0.467s gcd clipping for my raitons, i did some bad math before i edited the post lol

Posted this in Eliroos thread on reddit

Also went ahead and got some numbers for you,
Fuma: http://puu.sh/dxBfd/54aad59b4d.png Raiton: http://puu.sh/dxBiX/7cfc1c9756.png
In the parses you can see the damage variance on fuma is 626~673 non crit (not counting crits because I do not have enough), while Raiton is 702~746
So for the high end I lose 120 dmg per mudra (746 - 626), while on the low end you lose 29 damage (702 - 673), so the damage variance (for the small sample size I did anyways) is 29~120 dmg lost per mudra.
If you average the numbers in the parse out you get 718.4 for raiton, we shall round up for worst case scenario, 719, and 651.8 for fuma, we shall round down for again for worst case scenario.
719 - 651 = 68.
On average you can expect to lose 68 damage per mudra based on the numbers I presented.
Using the numbers you showed in your original post 10 fuma (3160) vs 10 raiton (3600), you would lose a total of 440 potency over that course of time. If we use the average we are losing a total of 680 damage (1200 if we use the highend number). Using spinning egde (again assuming the worse case scenario, the 2 gcds that you gain from using fuma will be spinning edge), which averages 390 for me, the 2 GCDs you would gain (4.6 seconds) will be made up for damage wise, but not potency wise. 390 (2 spinning egdes) x 2 (2 gcds gained from using fuma) = 780 but, 440 - 300 = 140, so while you may lose 140 potency, the damage can either be made up for a gain of 100 damage or a loss of 420 damage.

This is all assuming the worst case scenario, if you end up on anything other than spinning edge you are most likely looking at a DPS gain over raiton.

I know its meaningless and pointless to just post for this but


OTL!!!

I want that latency...
Seriously, this Mudra system feels like bloodline-skills from birth for our characters <- Given that it practically depends on where you play from.

Quote:

---

Originally Posted by Ephier

Shuriken VS Raiton

---

Well, it confuses me a bit there, even at the speeds in Aiurily's video, Raiton bites on the GCDs by about half a second, while Shuriken, as the person herself (based on char) assures, executes and ends before the GCD refreshes.

Given this, it would mean that Raiton is undoubtly stronger than Shuriken, but its pushing the GCD by about half a second per execution, while Shuriken almost plays like a normal oGCD.

So then in theory, based on those numbers Raiton compensates the timeloss by striking harder and Shuriken by not clipping on GCD?
In other words, if Shuriken is used instead of Raiton unless Kassatsu, the same DPS-output is to be expected?

Certainly numbers are weird...

should keep in mind these numbers for gcd clipping are on the generous side for me personally. my mudras are slower in pve instances and open world; estate is generally always very responsive for mudra speed, and a bit unrealistic because everything else isn't as smooth in comparison lol. realistically in a raid, you're gonna clip your gcd even more.

Quote:

---

Originally Posted by Arkenne

I know its meaningless and pointless to just post for this but


OTL!!!

I want that latency...
Seriously, this Mudra system feels like bloodline-skills from birth for our characters <- Given that it practically depends on where you play from.

---

asian mmo games have to have a certain ping-carried factor! it's just how it is~

Yes, by the end of the day is a tests played against a striking dummy and hence a lot lighter to the overall system... still, lag on the Mudras is terrible :/

Quote:

---

Originally Posted by Arkenne

Well, it confuses me a bit there, even at the speeds in Aiurily's video, Raiton bites on the GCDs by about half a second, while Shuriken, as the person herself (based on char) assures, executes and ends before the GCD refreshes.

Given this, it would mean that Raiton is undoubtly stronger than Shuriken, but its pushing the GCD by about half a second per execution, while Shuriken almost plays like a normal oGCD.

So then in theory, based on those numbers Raiton compensates the timeloss by striking harder and Shuriken by not clipping on GCD?
In other words, if Shuriken is used instead of Raiton unless Kassatsu, the same DPS-output is to be expected?

Certainly numbers are weird...

---

The whole point of what I posted was to show, that the dmg you would lose from using Fuma over Raiton would be made up from lack of GCD clipping. All the numbers I posted were assuming the worst possible outcome, and results could actually come out MUCH better.

Quote:

---

Originally Posted by Aiurily

should keep in mind these numbers for gcd clipping are on the generous side for me personally. my mudras are slower in pve instances and open world; estate is generally always very responsive for mudra speed, and a bit unrealistic because everything else isn't as smooth in comparison lol. realistically in a raid, you're gonna clip your gcd even more.

---

^

Quote:

---

Originally Posted by Aiurily

should keep in mind these numbers for gcd clipping are on the generous side for me personally. my mudras are slower in pve instances and open world; estate is generally always very responsive for mudra speed, and a bit unrealistic because everything else isn't as smooth in comparison lol. realistically in a raid, you're gonna clip your gcd even more.

asian mmo games have to have a certain ping-carried factor! it's just how it is~

---

Every time I mudra in final coil I be like

http://thenanobyte.com/wp-content/up...s_u18chan1.png

I pretty much have to depend on programs just to lower my latency issues with mudra. Everything else I do on any job is fine and my latency without it in general is great, but its sad to see I have to result to using them just to get mudra to function as it should in Coils. Its given me great results however. SE does need to either find a way to make mudra lag more bearable or change how you cast it to make latency of the ability less relevant. It is one thing if you lack experience to pull it off fast enough, but its a pretty terrible concept when latency itself is the major determining factor if you can play Ninja well or not. In short, I pay an extra 9.99 a month to play Ninja. :o

So how big a difference is there between the true BiS and the non-crafted BiS?

Quote:

---

Originally Posted by ThanatosPrimordialis

So how big a difference is there between the true BiS and the non-crafted BiS?

---

Crafted BIS wins by about 6 DEX. Considering that base DEX before considering DEX equivalency from the secondary stats is over 600, we're talking less than 1%.

My nina is only lvl 48 so far, this priority system became clear to me very fast as I was lvling but I was wondering about this 1 thing though...

Refreshing dots the second they drop isn't possible most of the time.. So what's the sweet spot for reapplying dots and DE? Do you refresh them when they have a few seconds left so there's zero downtime BUT a few ticks/seconds wasted by overwriting them? OR do you just finish the combo with AE and refresh them the on next combo BUT having a few seconds of downtime...

Hey all, Wondered if I could get some advice on my Yoshimitsu Novus quest. What is the best stat allocation for these? Crit is obvious I guess but not sure if I should put the rest on Det or some on something else aswell. Any advice on this would be great, Thanks :)

31 det/44 crit is best
31 det/33 crit/11 acc is a bit cheaper and still good depending on how much you need/don't need acc
23 det/33 crit/19 acc is the budget build with no tier IV materia

Ok thank you very much :)

Guys, I'm running into a serious dps problem with ninja and it's beginning to frustrate me enough to want to switch back to drg because I can not pull the numbers I think I should be pulling.

Today I could not parse higher than 445 on a training dummy at I lvl 116 with 115 weapons. I know the rotation, keeping all ogcd moves on cool down. I got the opener down pact and even had other ninja's come parse me and watch me.

No one could find a single flaw in what I was doing, kept dots/debuffs up. Kept ninjustsu rotation clean and flowing nicely. The tried to explain the priority of moves which I already knew. They even parsed themselves in similar gear to mine and was doing close to 500 dps with no buffs.

The only things they could even point out was that the at times got more dots up during TA than I did and their jutsu casts were faster (was using wtfast).

All things considered, I don't know why they were consistently at around 500 and I was consistently around 440 and we were doing practically the same thing.

Can someone possibly shed some light on what may be happening? It is incredibly frustrating when I and everyone else can't see anything wrong with what I'm doing but wonder why my numbers are so low.

Quote:

---

Originally Posted by Leonus

Can someone possibly shed some light on what may be happening? It is incredibly frustrating when I and everyone else can't see anything wrong with what I'm doing but wonder why my numbers are so low.

---

Did you have a Kiss up? I've done that a few times, drove myself mad trying to figure it out, and the numbers look similar - 440 + 20% = 528.

Quote:

---

Originally Posted by Pyroclastic

My nina is only lvl 48 so far, this priority system became clear to me very fast as I was lvling but I was wondering about this 1 thing though...

Refreshing dots the second they drop isn't possible most of the time.. So what's the sweet spot for reapplying dots and DE? Do you refresh them when they have a few seconds left so there's zero downtime BUT a few ticks/seconds wasted by overwriting them? OR do you just finish the combo with AE and refresh them the on next combo BUT having a few seconds of downtime...

---

Keep an eye on your DoTs/Debuff. If you're ready to start a new combo and SF is roughly 5s, refresh that. Mutilate is a bit easier to get away with, as there's no build-up. If it's around 5s, go with AE/DE first. For Dancing Edge, you want to start that chain with no less than 10-7s left, depending on your Skill Speed.

You'll pick up on these things naturally too, as you gain more experience with the job. Happy Stabbin'!

Quote:

---

Originally Posted by treuhavik

Did you have a Kiss up? I've done that a few times, drove myself mad trying to figure it out, and the numbers look similar - 440 + 20% = 528.

---

I thought about this too and on my buff bar it was. Maybe I should dropped it and reapplied it just in case.

Other than that, I have no idea.