The Sealed Dagger: A Ninja Guide

I'm still not seeing anywhere tp per hit is calculating in daken hits based off the shuriken's delay and not affected by dw reduction, so unless I missed it the spreadsheet will be underestimating ninja tp generation by quite a bit.

Edit: and I hate it when ppl only nitpick and offer nothing else, especially if things lack anything positive and yet here I am doing it. Thank you for the spreadsheet updates, maintenance and hard work Langly I'm sure I'm not the only one who appreciates it.

Ha ha verda, no sweat. I appreciate any and all input.

Edit: I'll brain over the daken TP bit this afternoon.

We already have most of what we want calculated between the data page and the melee page for Dakens contribution to the WS cycle.

Shouldn't be too tough to use that to modify the TP return values.

Edit #2 (because I don't like arbitrarily increasing post count):

As it stands I will have to change a LOT of fields in the Weaponskill page to take into account different TP values through cycles... I am not a fan of this. Ha, more work and I'm casting Tonko: Ni to hide.

Instead, I would like to adjust the TP/Hit field to include a bump. At the moment, the TP/Hit field is applied to all connections we make in a round. It is being used for the hits from Daken as well albeit at a discounted value thanks to the Melee page lumping Daken hits into Hand 1 Avg.Hits. (And perhaps this is okay, as we do not want to be going over 8 hits in a round).

If I take the current TP/Hit and distribute the remainder of the Shuriken TP value (Shuriken TP/Hit - Katana TP/Hit-after DW and StoreTP-) across the actual hit distribution within the melee page, it should bump the TP/Hit up enough to reflect Daken's influence. I'm effectively trying to put Daken's TP gain as a divvied up value on all other hits in the round since it can only happen once in a round.

Tell me if I'm wrong and where, or if I should just scrap that and go with modifying the weaponskill page. x_x

I'm not sure I understand the discussion here. What exactly is the issue with Daken?

I'd look at the sheet to try and follow along, but the version on the last page is throwing up a bunch of reference errors for me at the moment. Looks like some issues with index functions (Data!H9, Data!H10, Data!B22, Data!B23 at a glance).

The TP from Daken appears to be reduced from Dual Wield when it shouldn't be (at least that's what I gathered, idk.)

The index functions work fine for me. I just opened it up in Excel 2013.

The issue arrives in determining the TP required to complete a cycle to WS. Daken is being included in main hand hit distribution on the melee page, and when it comes to deciding necessary Hits > Ready to WS, it's only using the TP/Hit calculation from Data!B230.

So a shuriken hit should be netting something like 144 tp a hit, but it's currently being counted as 119 (The same as a katana hit which is affected by DW) See: Weaponskill!N18 and O18 as examples.

Checking dependencies of Data!B230 will lead you there anyway.

Edit: Those index functions are just pulling a relative value based on Gifts, they look like:


Wordy, I should just use a named range pobably.

Seems Google Sheets doesn't like the way you've written certain parts of those functions, see if this works in Excel:

Data!H8 =IF(D1="War", 10, 0)+SUM('Other Lists'!$B$144:INDEX('Other Lists'!$B$144:$Z$144,1,MATCH(Setup!$B$20,'Other Lists'!$B$143:$Z$143,1)))

Data!H9 =IF(D1="Dnc", 10, 0)+SUM('Other Lists'!$B$145:INDEX('Other Lists'!$B$145:$Z$145,1,MATCH(Setup!$B$20,'Other Lists'!$B$143:$Z$143,1)))

Data!B22 =40%+SUM('Other Lists'!$B$149:INDEX('Other Lists'!$B$149:$Z$149,1,MATCH(Setup!$B$20,'Other Lists'!$B$143:$Z$143,1)))

Data!B23 =SUM('Other Lists'!$B$150:INDEX('Other Lists'!$B$150:$Z$150,1,MATCH(Setup!$B$20,'Other Lists'!$B$143:$Z$143,1)))

Excel didn't mind those lines at all. Updated.

Cool. Looking at the Daken stuff now; I don't think there's an easy solution that is 100% accurate to the effects of differing TP/hit between katanas and shuriken as it pertains to WS thresholds, but I do think there's a slightly simpler/cleaner way to go about the method you've proposed. Going to poke around for a bit, will PM you.

Incidentally, have we established where Daken falls with regards to the 8 hits/round cap? For instance, if I get a Daken proc and a double QA proc would I lose the Daken proc or an offhand hit? Have we actually established that it counts towards the cap? This would also have ramifications for the method of implementation.

We have not. It was something I mused about shortly with Sechs in a PM, but there has been no concrete testing to determine where it falls in the multi-attack hierarchy.

This is why I was hesitant to get into the melee page and really mess about with it or the WS threshold column. It's ~already~ being counted, just discounted. That's why I figured I could make up the difference of that value within TP/Hit. Within that field, it would be after any store tp effects to the katana tp return so it's fairly unobtrusive. The probability of it being included as a hit already exists within the weaponskill page. I was going to just toss the remainder of the TP value to the TP/Hit field, except as a function of its inclusion in the hit distribution.

This, this is why I wanted input. lol!

I think that solution is good. A lot of the sheet tries to average things. In my mind at least I only see this problem with it but I'm not sure if it could easily be solved, I just retested with 2 STP (used seki shurikien sorry >.>) and no DW gear other than the amount listed:

So hopefully that will help match it up with models on the wiki.

In my mind say you get to ws sometimes and not others depending on daken procing enough or not. An average could aim low if you said 66 tp per hit with 50% proc rate is 33 tp per hit boost because then it would project you aren't getting to ws one round earlier ever, when in reality sometimes you do and sometimes you don't, though Nightfire seems the expert on that stuff so I'll let you guys handle it thank you :)

Edit, without Seki:

No test was ever performed about it.
I remember a test about Kick Attacks on MNK many, many years ago (hinting Kick Attack could not break the 8 hit per attack turn cap, but it was an inconclusive test in the end) but no test for Daken.
Conceptually Daken is similar to Kick Attack I think, an additional attack that can proc once per attack turn.

I don't think there is anything in game that can break the 8 hits per turn cap -correct me if I'm wrong- so it's safe to assume that Daken likely falls under the same limitation as well but of course without testing we can't know.

Before delving deeper into this topic though (there's really nothing more we can say that hasn't been said already without some testing) let me ask something.
Is it really that relevant?
I mean, a situation where you have a QA proc on MH plus QA proc on OH plus Daken proc, all in the same attack turn, doesn't seem particularly frequent to me?

I mean to say that even if the numbers produced by the spreadsheet won't be perfectly accurate because the sheet is unable to correctly address this specific situation, at the same time I seriously doubt the final numbers would be too far off, given how rare of an occurrence this is?
I'd bother with other tweaks in the ***first, before starting to think about such a rare situation.

Just my two cents of course, if someone wants to do some extensive test about it be my guest but it's gonna be a very tricky thing to test with lots and lots and lots of data required I'm afraid.

About the TP per hit, I think MNK has the same issue. Nobody is mantaining the MNK spreadsheet updated, I know, but the same thing happens.
MNK's kick attacks, while footwork is up, should generate a fixed TP as if your delay was a fixed one of 480.
This only happens to kick attack procs, and only when Footwork is up.

Pretty similar situation to NIN's Daken and Shuriken's delay.

not that hard to test, use a kraken club and merc dagger with minimal multihit gear. use merit ja and super revita and should take under 10 min

I don't think we can quite model it this way. Almost, but not quite.

In my eye, I see a Tp/Hit that's being applied to every hit we avg in a round. So in this case, if we add on the remainder at any function that doesn't distribute the tp across all hits (or the avg of our multi-hit distribution) then we're going to inflate the TP gain from daken.

Look at it this way, if we bump the tp/hit up based on the remainder * by the proc rate, then when you swing 6 times from two triple attacks, you're going to be over inflating the TP gain from Daken. Even if you don't hit 6 times, but two. You're still over inflating the tp gain from daken. Since you're applying it to 'all hits in the round.'

I see it as... we're going to need to add in the remainder of that TP as a function based on the current hit distribution in an attack round to spread the additional TP over all potential hits. The daken's acc and proc rate is already taken care of in the rounds > next weaponskill field since it is included in the melee calculations as an extra hit already (based on acc and proc rate). Would like to hear Night's input.

Sorry if it was unclear I wasn't actually suggesting that method but pointing out what would be wrong with it, I'm not actually sure how the spreadsheet would best accomplish this outside having the TP column on the weaponskill sheet have the tp from daken added in or not added in at all a certain percentage of the time.

currently it is:
For example could become something like (untested and pseudocode for the most part):

The issue is I don't think the hits till ws etc is calcuated from that currently but I could be wrong this is an area of the spreadsheet I purposely never dove much into because it's so complicated.

If you could go this route you'd want to remove daken calc from hits per round and also add the condition if total hits >= 8 to zero out daken or leave it in depending how you wish to model it.

Hey I have been using the recently posted spreedhseet trying to tweak my shun set for when I get aeonic.

I had heard everyone always talking bad about Jokushu Haidate on these forums. I am getting them on paar (4 dps win) with Samnuha at low buffs (zerk+aggressor+dia+chaos roll) and at higher buffs (geo) getting them to do 300 more. Same stuff happens with metsu (but using wsd herc pants).

Safe to say you cant go wrong with these pants for ws if acc isnt a huge issue, and with shun gear its normally not!

quick question, the OP says to use lugra earrings for WS's, but since the stats arnt active at day time what are people switching to during that time? because I'm certain the earrings arnt in the sets for the DA since we can get more from other sources like brutal.

edit
also if you dont have a spare herc body for ws, would abnoba kaftan be a good swap?

It would depend on the weaponskill.

Ten - Ishvara/Moonshade
Metsu - Ishvara/Mache Earring +1
Hi - Ishvara/Moonshade
Shun - Brutal/Moonshade

You could swap out Ishvara for Ten and Metsu and maybe Moonshade for Shun. I haven't done any spreadsheet for this though. Suppose I need to revise my gear for NIN.

and instead of moonshade at 3k tp? im assuming mache earring?

what augments should I be looking to get on my kanaria

3 TA, 15 DEX, 19-20 DMG 20 accuracy, 20 attack would be the best augments possible using Taupe stones. Fern doesn't increase the Triple Attack cap at all and 5 DEX is much better than 5 attack/accuracy, so Taupe is hands down the best stone for Kanaria. You'll probably end up settling for 3 TA 10+ DEX and some other stats.

so aim for ta3 with dex10 atleast for settling augments ill keep spaming stones but just wanna get a ball park of what I should be looking for in the short term

I have thousands of taupe into my kanarai and have yet to see 3TA an 10 dex together.... so you will drop a good amount of gil on avg for that.

if you can get 80% of max values you are doing golden.

Also if its for offhand dont worry about the dmg aug, it wont be a real impact overall.

sorry if its been asnwerd before but whats a good replacement piece for samhuha tights for tp and shun

Theres a lot its hard to repond directly. I mean you can use herc, which is a good choice? Jokashu isnt terrible for shun, and can beat samnuha even in right circumstances anyhow. I would atleast aim for herc legs at the lowest. They are extremely easy to get. Anyone can have them in 2 hrs~

So got my aeonic tonight finally!

Have been playing around on spreedsheet and seems Heishi/taka at best combo at max buffs, however Heishi/kik is best for trust play, was odd to me i figured that kanarai would have beat kik offhand.

heish/kik and kik/heish seem to be eachothers best setup which I didnt expect the spredsheets to show, but they also have dang near the same dps. Actually solo fighting selkit too tonight (10~) pops, both combos held a very consistent 2100 dps in the actual game.

Aeonic obviously shines when you can multi step solo chains. But between the 2 I dont see any large to matter dps difference. This is both nice and kind of disheartening haha.

So basically if you have access to aeonic, dont worry about kik unless you want the best offhand for tier 135~ stuff. Even then its only doing about 50 dps over the other options.

aeonic users, what SC 2-3-4-5-6 stpes are you actively finding give best dps results?

Relics as offhands are pretty bad. Your spreadsheet must be incorrectly accounting for the attack on it.

i went in and manually took off the attack for that same reason. Still got it winning, dont understand it!