The Sealed Dagger: A Ninja Guide

There isn't a snowball's chance in hell the +2 version of Ambuscade weapons wont have augments. They will, and it will be something absurdly good for the effort. I imagine them recycling the sap line of buffs (da/stp/mab/wsd/etc) onto a weapon, to make some truly unique options. But that's just me.

This is square.... there's a significant chance of anything lol. They could get zero augments... or full cape list of augments... or +3 with aftermaths... or divergence augments... or ranked augments from winning ambuscades with a new point system... or syngerizing evoliths onto them... or skirmish stone augments... or magian trials... literally nothing is off the table.

This is simply wrong, and you're overvaluing attack. Kikoku AM is certainly BENEFICIAL when not pdif capped, but even if you're able to use every bit of that attack it's not normally enough to outweigh the DPS benefits that either Heishi (STP+10/TP Bonus+500) or Kannagi (AGI+70/DEX+20 and the huge benefit of 50%ODT AM) bring to the table.

Plus, Aeonic and Empy's other stats are helpful even when attack capped, whereas Kikoku atk is straight up useless in well buffed situations and obviously falls well behind.


As noted above, you're just wrong if you think 50% ODT isn't better than ~200atk. And, using augmented numbers, the 70 AGI (plus the obviously useful DEX+20) is indeed useful, even more so with current Daken rate (max 79%: 40 base, 25 from +2 neck, 14 from gifts). It's harder to maintain sufficient Racc to match melee accuracy, so an extra ~52.5 Racc from AGI is pretty handy. Actually just about perfect for me to even melee/ranged acc up (excluding Madrigals, which only affect melee acc), since Daken gets a 100racc bonus and my normal non-Kannagi TP sets have about 140~150 more Acc than Racc, while my Kannagi set is almost even with Daken Racc (100~110 more melee Acc than Racc, before accounting for Daken bonus).

FWIW, the contribution to Daken and TP generation is also a big part of the reason Fudo Masamune performs well as an offhand on stuff where you're not capping racc.

And this isn't a question of freeing up the ability to change other TP gear. The rest of your TP gear doesn't really change regardless of weapon: Adhemar+1 head/hands, Kendatsuba+1 3/5 (or Samnuha legs or Herc feet depending on situation), JSE neck/back, same earrings/rings/belts.

I see a lot of people who just can't seem to fathom that Kannagi is any good, based seemingly only on getting subpar results on Blade: Hi with a non-Kannagi weapon that's not good for it (and likely with a subpar WS set), and despite both spreadsheets and real world experience showing otherwise. I have parsed my own DPS many times in same content/buffs with Kannagi and Kikoku, and Kannagi always gives me better DPS than Kikoku outside of situations where AM isn't possible to maintain (very rare). Apex mobs, Ambuscade, Dynamis, you name it. I really only pull the Kikoku out when I want to use Metsu for SC purposes, or if for some weird reason it's not viable to get and keep AM up (super short zerg Ambu fights, weapons and aftermath getting stripped, etc.)

The spreadsheet is known for completely overestimating auto-attack damage in the TP Phase, especially for 1h jobs. You said you parsed but I'd love to see it, because from what I know 45% autoattack damage should not beat 200 attack power.

45% autoattack damage when without store TP roll I do around ~25% of my damage on auto-attacks doesn't make up for 200 attack power I'm pretty sure. Would love for a more math savvy person to answer this.

Relic having attack plus does absolutely nothing unless you're soloing or playing with a party so bad that you may as well be soloing. This is the exact case for Mandau. it's completely ignored because(*) everyone is 100% capped on 100% of the things(*)

I've tested it:

Have you? Your profile shows that you have a Kikoku but not a Kannagi.

No, I have not, as I do not have it. I would like to see the math how 200 attack loses to 45% autoattack damage when autos are only ~25% of your damage.

Edited my post cause missed that part D: Multitasking is too hard.

Any situation where the attack bonus from Kikoku isn't strong enough to matter, then Heishi should be the optimal choice, yes? If you benefit greatly from the attack power(Or skillchain properties) then Kikoku should be better. Kannagi is below both of these weapons, unless niche amnesia/qutrub situations. Is this not correct?

I just don't see how Kannagi can win. You talk up the weapon and say you test and stuff, but I don't see actual proof and evidence. That's all I'm asking for, I'd love to admit to Kannagi being a great weapon but I don't see the data or proof of this claim. Please provide it if possible, I'm asking nicely and not saying I think I'm right, I just would enjoy learning. Knowledge = power.

Personally, I don't think Kikoku is that great of a weapon but it's better than non-rmea katanas. Heishi should beat Kikoku in most situations, and absolutely destroy Kannagi unless it's some amnesia/qutrub type scenario, and for Kikoku to ever be worth it over Heishi you're looking at equally niche situations(skillchaining, solo DD, etc).

All this attack and TP% talk and everyone is leaving off Metsu.

8.4 ftp (at 1k) 80% DEX Darkness / Fragmentation

Howso?

It's a good thing to point out, but you'll never WS at exactly 1k TP every single WS. And TP overflow does zero for Metsu. Absolute. Zero.

Just the way it calculates tp per round, it always assumes the same. You don't always swing 3.3 times when you might only swing 3 times. just the fact that it uses an arithmetic mean for every calculation means it has to either over or undervalue things. It inaccurately gives tp return values since it'll return values you can't even get. Tp phase is overvalued marginally just based on the TP return and always using an average hits to 1k.

Correct but average rounds to WS is all that matters.

And has anyone figured when the WS bonuses are calculated?

We know it’s after ftp and the 40% and 20% on relic are multiplicative. But is the 40/20 additive or multiplicative with Wsdmg gear? We are probably underestimating R15 relic WSs.

it's multiplicative. i don't know why you think people are underestimating anything when nobody posts any real math

We argue with feelings here Austar.

To be perfectly frank you can't quantify what's better realistically anyway.

There are "kinds" of better. Not even "situations" but just "robotic" better, and "the way you play" better.

Kannagi can be totally better for someone that doesn't immediately ws as fast as a bot would while kikoku is theoretically better while a program ws's exactly perfectly every single time.

If you don't exhibit flawless play, all your numbers go out the window.

It's really not that big a deal they're all good. (except nagi) You're talking literal seconds difference here. Not minutes. Not pass or fail. Not timing out. Not doing zero damage because of some invisible barrier.

The sim DOES do a better job of approximating this, but that's not the gospel either. It's just a guess.

when people get different results using the same program, proper variables should all be shared to find where differences lie. just saying the result doesn't do any good.

And to add another factor, your group will change this too. Generally, I am top of the parse in my LS, by a wide margin. Last night though we had more people on top jobs than usual, and I actually got smoked. Mobs were dead before I could get going, and I had a lot of wasted time. My DPS was down significantly simply because my alliances was much higher. I needed to change my playstyle if I wanted to win that parse.

So yeah, play which mainhand fits how you enjoy playing the job. If you are amazing at WSing asap, Kikoku is possibly your best bet, as that is it's playstyle.

And I feel Metsu

looks like i needto with m Ancient currency... was about to put it down on kikoku

so ignorind aeonics.
Emps are the supperior weapon ?

I'm still not sold on the Kannagi argument. If the theorized impact of AGI was more conclusive (Parry, subtle blow, etc.) then maybe. I don't think Hi is that great of a WS, and would think the double damage and STP+25 Fudo would keep up if not beat Kannagi in a number of circumstances for mainhand (especially if acc and/or att was uncapped). But maybe I just haven't properly geared/buffed Hi to see its benefits.

Kikoku looks like a solid weapon, but I think for how I want to use Nin, Heishi and Ten spam works best. I can see both of those weapons being solid. I can see how Kannagi may work, but I'm not sold on its argument.

i dont know how much i anna +1 this

So often a simple debate turns into a match about who is right rather then listing and share information to find what is the reason of the different and what is the truth.

The spreadsheet on the first page is updated right? If you don't trust, just put some ***in and see what it spits out.

Every bit of gain that kiko gets from r15 kannagi gets as well. 70 modifier (for any ws) is a ton. I feel like a lot of the "kannagi is best" is from 70 agi going into daken.

Kannagi is the only rema that gives a boost to daken at all (right?)

kkk is only melee attack, heishi only got acc/macc and nagi is lolnagi(it also only got acc and macc) ((well heishi has stp for daken))

So the argument is the RACC you get from kannagi's AGI? You get more RACC from another Fudo. And you'd get STP+25 for the Daken throws.

I mean, we're not considering not-rema, so

Also, isnt 70 agi putting some white damage into throws

just ranged crit chance

We should consider whatever could be best, REMA or no.

It would appear to increase the crit rate of Daken, if bgwiki on AGI is to be believed.

I'm interested in what proof there is of the claim that AGI reduced monster TP gain independent on subtle blow. Not simply for NIN, but also for THF. Bgwiki page claims it, but has no source.

The problem with the modifier for Hi is that it's AGI. See, if it was DEX on Kannagi and worked for other WS, it'd be fine and good. At minimum could get to Kannagi and use another WS as needed.

Metsu and Hi have the same fTP and number of hits. Kikoku gets +30% to Metsu from trials, though. And then r15 has a higher multiplier that also STACKS with the +30% to Metsu. So it's (80% dex mod x 5 ftp) x 1.56 vs (80% agi(+70) mod x 5 tp) x 1.1. In all likelihood this means at low gear levels Kannagi and Hi can win out, but eventually Metsu will run away with it. And it's hard to gear to crit for Hi as you want to stack WSD and AGI wherever you can.

Fudo vs Kannagi gets you:

50% double damage vs 50% triple damage (win for Kannagi, but must keep AM3 up)

STP+25 pure win for Fudo

RACC+50 vs RACC+ 51-52 from AGI on Kannagi

ACC+50 vs ACC+15 from DEX

As much as attack+105 from full Shadows on Fudo.

DMG: 161 Fudo, DMG: 153 Kannagi

And a 1.1 mod to a mid-tier WS for Katana.

I'm not seeing that be a clear Kannagi victory by any means.

The page that is currently up on the spreadsheet from the OP has Heishi beating kiko by ~500 dps I'm not gonna dl it and *** with it, but if you want to know why/how kannagi wins (if it does) that will answer your questions.

From a quick look, I can see there are zero buffs to racc, so if you change one to kannagi without changing anything else, it's going to win just based on that.

base acc main/offhand are 1500 daken racc is 1100 (it has torpor turned on*)


This was added... 12/6/2010 apparently


This was a real ***, because blu/rdm used to give 10(0) TP from dia. and the agi change made it 5(0) tp.

So it's official that it does. But...what does it mean. Is it like dDEX on crit rate and overall a rather minor amount that you must meet a certain tier amount to hit? Or is it some linear thing independent on the attributes on the enemy? Considering there's no attribute related to store TP for players, it would make more sense for it to be the latter in some fashion. But to what degree could be significant.