The Sealed Dagger: A Ninja Guide

Actually a very good point. Kagero and Koki would decimate undead. I'll check later and report in.

GKT without /sam and hasso will be significant drop tho. Especially that you will need to get to 2000TP probably without tp bonus of Heishi. Also Kagero is one hit without offhand hit, its significantly weaker in concept than NIN hybrids. Btw idk why we havent heard about SAMs using hybrids yet (other than Jinpu on Apex or in wave 3)? SAM should have similar success using them now with Nyame, improving SAMs versatility even further in segments.

A buddy of mine had been using hybrids on his SAM. Apparently, they don't generate nearly as much hate. So hybrids + tank can make hate control much smoother.

Edit: BTW this is a very relevant factor to consider for NIN main tanking heavier stuffs, if your DD's are using hybrids, should help let NIN tank if at all a little smoother, so in addition to these thoughts, NIN should be using PHYSICAL SC components, as much as possible, to a skillchain set, with your SC partner using strictly hybrids.

IE: solo SCing, go hybrid, NIN tanking with moderate DPS function included, go PHYSICAL, with a real dedicated hybrid only other DD SC partner.

Bamp, NIN'DOYLE rulz!

How many support DPS, could step into some mean secondary be the hybrid only SC partner scenarios with NIN main tanking could be there?

Looking at you, RDM, GEO, THF, BRD, BLU, maybe a newb RUN or PLD, learning hate balance?

WAR, SAM, DRK, etc all, can learn not to show boat so hard.

But also, Mastered up RUNs and PLDs should be aware of this also.

Upon further review, I think pure magic WS enmity wise is the same as physical WS, but hybrids have something wonky in spaghetti code to make them have less hate

IDK, just think thoughts we've had of possible strategies, mulled over.

Ok alot of this is unintelligible to me but enmity for pure ele wses works NOTHING like enmity for physical wses. Elemental wses generate VERY little hate. I'm not terribly familiar with hate for hybrid wses since stuff you can hybrid ws is usually dead in 1 ws.

Ok, so with this then, the heavier your overall WS damage swings towards the magical numbers, then the easier it is for your tank to maintain hate. 10-4, wasn't sure how that part worked, thanks for the clarification.


Physical generates most hate, if susceptible (hello tanks), hybrid moderate (hello midgrade dps DD/supports), pure magic negligible (hello heavy solo SC/DDs that cannot mitigate the damage taken from the hate)

Correct?

Skillchains still generate a good amount of enmity no matter the source. Closing big skillchains too early on magical WS was the only time RNGs and CORs would pull off of tanks when using the magical WS strat in dyna.

Nice thanks for that. I know these thoughts I am trying to convey are detouring away a bit. But maybe it is still important to keep it fresh in the minds of newer NINs actually trying to tank, but wondering why their 99k hybrids may not be keeping hate as well as they were expecting.

Hybrid ws is like two WSs in one really. There is physical and magical WS there. Hate for physical part is calculated normally and hate for magical part is calculated like for magical WS. Simple as that :)

Yes but overall, the magic WS's and/or the magic portion of hybrid WS's some how have less enmity calculated through the XI spaghetti added to your hate table, with a significant margin as compared to the physical. Is what I am trying to clarify, and it seems as though that is correct.

Magic WS's point for point at 99k caps, are less enmity, than physical at the same 99k capped damage.

There is nothing spaghetti in it. Hate from magical WS and magical portion of hybrid WS is calculated from base damage before dstat and multipliers like mab, affinity, weather etc. Thats all and it's known since 2013 at least.

https://www.bg-wiki.com/ffxi/Enmity#Enmity_from_Damage_Dealt
https://www.bluegartr.com/threads/105852-Wildfire-emnity-testing?p=5638484&viewfull=1#post5638484

They are less enmity, is all I am saying.

@Ruaumoko did some tests myself, here are some additions or different experiences on my hand.
Small sample.
Gear used at the end of this post under Spoiler


Karakul: To >> Teki + Chi

Vulture: To >> Chi

Leopard: To works well (couldn't test the others)

Ziz: To = Teki >>>> Chi

Raptors: all 3 hybrids seemed to be working good (even got an 80k dmg at one point o.O)

Made the ambu gk a few back to goof around on nin + full on stp set. Thought it'd be fun to have almost every element hybrid on nin (minus darkness and thunder? bg says Tachi: Goten is hybrid but dmg/tp return looked like pure magic).

As /sam w/trust tp gain really wasn't too bad (able to self sc and closing ws @ 1.5k+ tp) might be fun w/sam roll for more multi attack from stp and ambu gk.

Guessing in 99% of cases it'd more for novelty/fun. Also like Simon said- no hasso, is going to sting a bit.

Sam main, koki/kagero definitely pump in dmg, esp if playing into mob weakness.

Hey all! New to Ninja here - How outdated are the equipment sets on the main page? I know Nyame is taking over when it comes to some(most) WS sets. Also - I'm using Selendrile Lua and can't seem to switch Stance, if anyone has any hints, I've tried everything I can think of. Thanks in advance.

//gs c toggle stance

Or
//gs c set stance Yonin
//gs c set stance Innin

Amazing Thank you!

if you can find and open the lua file for NIN, and look around line 18 and 19, they have binds already setup to shift between the two modes.

Looks like this

send_command('bind ^` input /ja "Innin" <me>')
send_command('bind !` input /ja "Yonin" <me>')

the ^ and ! represent either ctrl or alt, so it'd be basically ctrl+~ and or alt+~.

Hope this helps

Quick question about the Ninja DPS spreadsheet's PDIF calculations.

The BG wiki page for PDIF states that "1H melee weapons have an effective ratio cap of 3.625" when referring to Ratio=Attack/Defense before calculating cRatio. The Ninja DPS spreadsheet does not apply this effective cap in the cRatio cells for weapon skills (in the Data sheet: cells D76 and D140 for example).

Without the effective cap of of 3.625, the lower and upper qRatio limits are identical and equal to the PDIF cap adjusted for PDL. This means your qratio spread is 0, and your overall damage spread is decided entirely by the +0-5% random multiplier at the end of the PDIF calculation, leading to a very narrow range of possible damage. It also means that any increase to PDL through gear or traits is always a direct increase to damage dealt when attack capped. My understanding was that PDL increases only the upper limit of your spread of possible damage, but in this case PDL is acting almost like WSD that applies to all hits when attack capped.

If you are capped attack and you do use the effective cap of 3.625 before calculating cRatio, then your lower and upper qRatio values are centered on your wRatio (3.625) and have a large spread (wRatio+-0.375). This means your upper qRatio limit is defined by your weapon type and PDL+ gear/traits, while the lower qRatio limit is defined by your attack/defense ratio, which leads to larger damage spreads for each weapon skill.

Somebody mentioned to me that the "effective ratio caps" listed on BG may simply be relics of the past that were never removed from the page after we determined the "pre-randomized caps by weapon type." This makes sense considering that these "effective ratio caps" all coincidentally line up with the "pre-randomized caps by weapon type" after adding 0.375 for capped wRatio. Has anyone followed the history of PDIF enough to confirm whether or not we should apply an effective Attack/Defense cap before calculating cRatio, wRatio, and qRatio? The BG wiki page should be updated if these effective ratio caps aren't actually a thing, at least.

That sounds like an error in adjusting for the randomizer aka "qRatio" in the sheet. Or simply ignoring it in favor of assumed maximums for the sake of simple comparison.

You also seem to be hung up on "cRatio" which is the level correction term. They've been slowly removing it from the game in chunks and pieces and it's not for any iLv content, so if it's effectively unused in any spreadsheet, that's gonna be moot and perfectly fine.
(In other words, it's only a thing that happens for ***that doesn't matter because you're 50 levels higher than it.)

"wRatio" is just the critical hit modifier for the pDIF cap. Generally a +1, but pewpews get a 1.25x.

These terms are all relics of whoever originally decided to name them like this. (Probably a math person, as a chemist, my instinct is to use Δ for everything like this.)

Essentially:

• "pDIF" = "physical damage limit" = "Ratio" = your attack÷target's defense

• "cRatio" = "α" = level correction (and is only relevant for old stuff)

• "wRatio" = critical hit pDIF bonus

• "qRatio" = the stupid physical damage randomizer that's been a thing in JRPGs since time immemorial

The goal is also to get capped pDIF to maximize damage, so that's why even though pDIF can be less than a weapon type's cap... no one cares unless you're capped.

Also: yes. PDL+ does indeed work "like WSdmg that applies to all hits when attack capped" in essence; because the "final" multiplier to physical damage is your attack stat divided by the target's defense stat, aka pDIF. And since we only care about cap, increasing said cap will be a direct increase to overall damage.

There's quite a bit of misinformation in here.

PDIF and Ratio are not the same term. Ratio is simply your current attack divided by their current defense before any other terms modify them, while PDIF is a term for the whole series of attack/defense calculations (though confusingly, PDIF cap refers to the value at which qRatio is capped).

cRatio is the modified version of ratio after level correction has been applied, not the level correction itself. As such, when there is no level correction, Ratio and cRatio are exactly the same term and can be used synonymously.

Similarly, wRatio is the variable that you get after applying the critical hit modifier onto cRatio, not the bonus itself.

qRatio is the name for the upper and lower limits of the bounded term after you plug wRatio into the damage range formula; while the equation is a bit more complex at lower wRatio values, at higher ones qRatio (Upper Limit) = wRatio + 0.375, and qRatio (Lower Limit) = wRatio - 0.375.

The qRatio values are then capped, and a final 1.0x ~ 1.05x randomizer is applied.


Example: You are using a Scythe, you have 3000 attack, the mob has 1000 defense, and it's 2 levels higher than you. You do not critical.

Ratio: 3000/1000 = 3.0
cRatio: 3.0 - 0.05 * 2 = 2.9
wRatio: 2.9 + 0 = 2.9
qRatio (Lower Limit): 2.9 - 0.375 = 2.525
qRatio (Upper Limit): 2.9 + 0.375 = 3.275

Multiplying these by the final 1.0x ~ 1.05x randomizer term gives you a final range result of 2.525 - 3.43875, which is then multiplied by your total DMG and any other factors to get how much damage you do. If you had a 300 DMG weapon and 30 DMG from fSTR, on a neutral enemy, you would do 833 to 1134 damage per hit.

PDL starts having an effect when your qRatio Upper Limit surpasses your pre-PDL PDIF/cRatio cap, at (pre-PDL PDIF cap - 0.375), though it only reaches maximum effectiveness when your qRatio Lower Limit surpasses your pre-PLD PDIF cap, at (pre-PDL PDIF cap + 0.375).

To fully cap attack, it requires a cRatio of 0.375 greater than your weapon's PDIF cap, not counting PDL effects. Counting PDL effects, your attack cap is ((Weapon PDIF Cap + PDL Traits) * (1 + PDL Gear)) + 0.375.

The most extreme case possible, a DRK with a Voluspa Scythe inside Domain Invasion, 5/5 Sakpata, and Abyssal Beads +2, would require a cRatio of (4 + 0.5) * (1 + 0.7) + 0.375 = 8.025 to cap.

Thanks. That matches my understanding of PDIF and tells me that the "effective cap" of 3.625 for 1-handed weapons as listed on BG-wiki is not to be taken literally or applied to attack/defense ratio directly. It is instead the attack/defense ratio you would need to be at to reach "attack cap" before any form of PDL+ is added in.

EditEdit: Nevermind.

Thinking about it a bit after stepping away for a few hours made me realize it must be the 2nd case:

I was curious how the use of elemental Ninjutsu would impact the damage produced by hybrid weaponskills. I did some really random testing last night and found a pretty dramatic difference between the damage done by Blade: Chi before Raiton: San and after, but it wasn't against the same mob and it is possible trusts were debuffing or the mob was changing resistances.

While I was waiting for my segfarm to stop, I went and hit up some Apex bats to do some preliminary testing out of curiosity. I tried to WS at the same TP, but did overtp a few times. I also know there is some variation due to DA/etc. The difference wasn't as dramatic as I had observed the night before.

I admit this is limited, but I'm curious what other peoples experiences are or if this is something that has been evaluated further or not.

Blade: Chi (w/o Raiton)
Average: 10639
Low: 7506
High: 18216 (I was over-TPd here)

Blade: Chi (w/ Raiton: San)
Average: 11759
Low: 6099
High: 17089

Blade: To (w/o Huton)
Average: 8312
Low: 4788
High: 11950

Blade: To (w/ Huton)
Average: 9112
Low: 8409
High: 9929

Blade: Teki (w/o Katon)
Average: 8618
Low: 7401
High: 11230

Blade: Teki (w/ Katon)
Average: 8954
Low: 8039
High: 10704

While it is an interesting idea, I believe that the elemental ninjutsu only affect resist rates, not damage. It's always good to test these things to be sure, though.

I'm curious to know what your sample size is for your trials. How many weapon skills were used before and after using ninjutsu for each hybrid weapon skill? With a small sample size (perhaps even less than 1000-2000 weapon skills per trial), your average damage may not be a good representation of the true average, especially if you're sampling from a skewed or multi-peaked distribution.

If we assume that elemental ninjutsu does not affect physical damage in any way, then we can test this more efficiently using Lode Golems. Since Lode Golems are immune to physical damage, the damage dealt by hybrid weapon skills will be entirely decided by your magic damage stat and magic multipliers without any variance due to differences in TP, PDIF randomization, or multi-attack.

A quick test on Lode Golems suggests that elemental ninjutsu does not affect the magical damage in any way.

Makes sense! Sorry I neglected to include that information, I put the raw data below. I was running out of time around Teki/To (mainly because I had to restart due to a trust using Dia), so I had to cut those a bit shorter. I was planning on going to gather more sometime soon-ish.

I do agree the sample is too small to draw conclusions from (esp To/Teki), but also likely skewed due to physical damage, which your test eliminates. It was more of a start to see if anyone else had evaluated it before I go further at it.

Blade: Chi - No debuff
9769
14420
18216
9294
11939
10058
10210
8739
9247
9894
8382
7506

Blade: Chi - Debuffed
17089
6099
10659
10919
10049
10652
12392
10165
16895
11804
12617

Blade: To - No debuff
10331
11950
9068
7674
4788
6061
8850 

Blade: To - Debuff
9929
8409
8576
9532

Blade: Teki - No debuff
11230
9344
8530
7751
7401
7835
8241

Blade: Teki - Debuff
8039
9643
10704
8277
8108

I'm currently taking a break from Odyssey but before I did I largely finished the chart I posted a few weeks back.





When you get this down the only jobs you're losing to in a parse are Empyrean SAM and DRK, even then that's only on the white damage from AM3 procs.





Bottom line though, it takes some work but when you get there a good Ninja is an excellent option for optimising segment runs; as the strong Hybrid Weapon Skills fill in for areas where DPS might be lacking.



Here is a video of one of our segment runs, just shy of 11k, which was great considering it wasn't a perfect run mob-wise.

I didn't double check Rua but did you add the fixes and additional stuff I posted a few weeks/pages ago already?

The initial table you originally posted had like one or two errors and was missing some families.



Also if someone doesn't want to use Altana's Repast or Marine Stewpot, then I think Grape Daifuku +1 is a valid alternative.

Offhand-wise what's best after Kunimitsu? Ambu Katana? Dyna Katana?