For The Shinies! A Guide For Thief

Idk what to tell you, my opinion is based on math and your on single anecdotal exeperience. Maybe you was holding tp idk.


Yes it is, with the exception of R15Twashtar/Centovente.

Avg Evisceration at 1000TP with Kaja Knife (at capped pdif) is around 28638. Thats same as main hit on 2250TP Rudra That's 1.7k below Rudra's main hit at 2250TP and far more than 1250TP Rudra. White damage will be higher on Twashtar/xxx but nowhere near to counter much lower damage on WS. Only Centovente can push Rudra far enough (assuming overtping too) but Centovente lowers white damage to equal to 2x ilvl daggers (odt on Twashtar counters loss from Centovente).

Also if you are parse freak on dynamis, remember that Evisceration will do like 15% of full dmg if mob is at 5%hp, while Rudra will do almost full.

its independent of triple attack though.

Right, but one more swing isn't nearly as substantial if STP could potentially be an option. Though I suppose in regards to Raetic+1, that's what I mostly get from it.

Holding TP for higher WS damage vs higher WS frequency is something I cant really calculate that easy (so I wont, I have full time job and kid XD).

I can only do something like this.
Assuming:
- 1250TP (before TP bonuses) and 1300 for Aeneas (because of STP and slightly higher delay)
- 5 hits with each hand
- 3.35 pdif for ilvl daggers and 2.8 for Centovente
- this TPset
- Rudra for Aeneas/Twashtar main and Evisceration for Kaja main
- Perfect none DM augments, DM augments would push Rudra higher, but not perfect none DM augments will also take Rudra's down a little.
- AM3 up for Twashtar main

WhiteDamage/WSDamage/Total:
Twashtar/Cantovente - 7500/32215/39715
Twashtar/Kaja - 9555/25464/35019
Kaja/Twashtar - 6285/29553/35838
Aeneas/Twashtar - 7040/28727/35767 (this combo is also 5% slower than Kaja/Twashtar or Twashtar/Kaja)

Keep in mind I don't really try to calculate full cycle damage often, so there is some margin of error here for sure.

eviscerating also has more variance and generally less favorable skill chain options when just spamming. so an arithmetic mean doesn’t give the full picture, the shorter a fight the “better” rudra will be

What does this model show for Twashtar/Taming? Taming looks stronger on all points except lower delay may get more from Twash white damage.

i have a dual wield tp/round simulator i posted in the red forum. it’s simple enough to use, just add all your totals up. it also compares raetic weapons both nq and hq based on mp values so you can compare the to gain of either path if you need.

Raetic+1 is 2 more tp a round when not in Escha, and 20 more TP while in Escha, per round. Not terrible.

With max samurai roll, Taming actually beats Raetic+1 outside of Escha even with MP up, in Escha it has the 20~ more tp a round advantage.

Though isn't this simulator just showing TP per round, but obviously there are different round speeds so the Raetic+1 has a bit more of an advantage there.

This focus on max DPS is making me rethink my standard Moonlight x2, but I prefer safety over everything and how can you possibly contend with HP+220, DT-10% in those slots, I don't think a few DPS is worth it.

Unrelated question though: Auspice doesn't stack with enlight/endark, is that specific to those buffs, or do Rune Fencer runes also stop the effect?

yeah it’s just per round, since if i cared to compare on a per second basis i’d have to add haste options and it gets closer to the point of just doing a full simulation then. i initially wrote it to have an accurate comparison for raetic and path b options for tp gain.

10250/26533/36783

with 2.5% more hits per round but ~9% slower swinging. I think I would prefer Kaja, but tbh all those ilvl offhands for Twashtar should be really close and have some pros and cons. Kaja is getting further upgrades (or augments) tho.

Yeah, I'm betting next (final?) upgrade will finally replace Taming, but we shall see.

Next upgrade to Ambuscade weapons will be the final direct change to the item itself, then will be followed by some sort of augments.

I've been messing around with twashtar/Kaja alongside Twashtar/Taming and even in its current form I'm liking the kaja knife in the offhand more. It's just so much faster than sari and it's performed very well.


Actually there's more to it than that. The TP formula starts to scale in favor of faster tp gains when you can push your delay below 180. While the TP formula is *mostly* linear across delay ranges it does have a slight curve, and when your delay goes really low the curve starts to give better tp returns. Kaja knife's delay is so low it actually DOES build tp a little bit faster than Taming with capped delay reduction and the sTP pieces in our usual attire, and that's going to reflect in your overall rudra's damage in ways you can't eyeball or easily measure to a reliable degree using a parse. The white damage boost is pretty nice too, especially with aftermath up.

I think it'll all be moot when the final upgrade comes out though. So far we've seen an increase in base damage of 7 --> 5 --> 7 and every step has also added incrementally to accuracy/attack. All of the 4th stage upgrades have dex/int/mnd and the stats follow equalized distribution, so there's a very high likelihood the final upgrade will get 10 str/agi/chr. I suspect there will also be a bump of 5-7 more base damage and another 5 accuracy/attack as well. That's more than enough to surpass a perfect Taming offhand in my book. It wouldn't even need enchantments, the raw stat line is just good enough as is.


EDIT: Here's a few numbers to back up my point

Twash/Taming gives 53 base TP per hit. The combined delay is 387, so at capped delay reduction you're at 77.4 delay. That's 2790 attack rounds per hour before the 3% TA rate, which for simplicty I'll math using a base 50% TA rate before the dagger. That's pretty much a flat 2% increase in attack rounds (4185 at 50% TA rate versus 4268 at 53% TA rate), raising our previous figure of 2790 to 2845 attack rounds per hour. That equals 301,570 Base TP per hour.

Twash/Kaja gives 51 base TP per hit. The combined delay is 356, so at capped delay reduction you're at 71.2 delay. That's 3033 attack rounds per hour, or 309,366 Base TP per hour.

Feel free to go over my numbers and check for errors, but I'm 99.9% sure I did that right. I went over it multiple times.

So in addition to the added white damage Twashter/Kaja has a 2.58% increase in base TP gain over Twash/Taming when your delay reduction is capped. This ignores all extraneous factors such as accuracy, multi attacks (other than the 3% TA on sari, which I included in the math), and sTP since they're associative and affect tp gain in both builds equally.

Twash/Kaja knife already has a lead over twash/taming in white damage, but when you consider that it ALSO has a lead in TP gain the choice is pretty clear to me. Kaja knife offhand looks a little bit stronger than Taming even in it's current form, but it still has one more upgrade left to go. The final upgrade should widen the gap to the point the difference isn't even questionable.

Allow me. Same test as before. Tax'et, no debuffs that would cause it to take more damage, Honor March, Min5/4/3, Haste2, RCB. None of my weapons are augmented, unfortunately.

Aeneas/Twashtar - 56-57 DPS (5 minutes of rudra'ing)
Kaja/Twashtar - 57-58 DPS (10 minutes of evis'ing)

Of course, SC damage aren't a factor in those numbers. I popped a RCB since you're going to anyways on anything where you'll be using evis as a primary damage source. Basically, it's good, but doesn't blow the standard rudra setup out of the water.

My evis set is pretty standard. Mummu+2 head/hands, Lustratio legs(dex/crit)/feet(str/dex), plunderer+3 body, soil gorget/belt, moonshade/mache earring+1, begrudging/regal ring, yetshila+1, maxed out crit toutatis back.

I went and tried Tax'et but no dps, i have some ws numbers. Twash is r15, sari capped, no buffs but haste, both ws spam close to 1k tp.

Rudra Twash/sari avg 133 ~ max 195
Evi Kaja/twash avg 130 ~ max 213

Added geofury

Rudra Twash/sari avg 212 ~ max 288
Evi Kaja/twash avg 199 ~ max 288

So 2.3% more with rudra with no att buffs, 6.5% with some att.
Add more att, AM3, bigger darkness, etc and is clearly that kaja evi's is no match for rudra.

Scratched all that since evi set was stuck in glitch ws era.

Please stop this stupid tests on 99%DT mobs. Do real test if you want.

I bet there is rounddown going there that breaks everything for multihitWS for example. Like instead of 8x 18.9 for example being 151, you are getting 18.9 rounded down to 18 and you get 144.

What 288 max on Evis suppose to mean? That 28800 Evisceration is highest Evisceration you can get?
Why is max for Rudra and Evisceration the same? Do you want me to believe its coincidence?

You didn't even write what's in your WS sets >.>

EDIT: If you want to make a real practical test:
- Do it on mob with no special damage taken mechanic
- make /echo for TP, when WS is being used and provide that info
- provide info of your WS sets
- provide info about buffs, or important stats like total DEX, STR etc.

Obviously this take much more work than simply SB results from AFK Tax'et with AutoWS, but it's actually can provide some valid info. /rant off

I noticed that too. It's possible it's just a coincidence but I suspect Tax'et is throwing away too much data in the rounding to make it a viable test. Or rather it's throwing away data in uneven intervals that will skewer results.

Hey at least provide a target for the test, even with dt mechanics tax'et is more reliable than w/e those stupid sheets provide.
And yes 288 on both ws is a simple coincidence.

Hey it's not me who want to provide practical test. I wanted to provide theoretical info and did my best.

Apex bat in ra'kaznar, buffs: indi-haste.

Twash/sari 7724 dmg avg rudra
Kaja/twash 13407 dmg avg evi

Dia 2/3 + bog frailty idris.

Twash/sari 33439 dmg avg rudra
Kaja/twash 29683 dmg avg evi

ItemSet 365249
ItemSet 365252

Evi set could use relic+3 body i guess.

Edit1: i wonder is more mummu would get better results since there's a huge gap between avg evi (29683) and max (41007)
Edit2: redo evi test with mummu gamashes, dia2/3 bog frailty, avg 31578 max 39680
Edit3: mummu bonnet over adhemar : avg 30861 max 38168

ItemSet 365229

Thats the set that i was using for calculation.
Relic body+3 is huge improvement, especially if you have abuse free 5% triple attack from behind. Same as 10dex, 5% crit damage herc boots. I dislike tonberry ring, THF is fragile enough without it.

As I recall, evis hits stop if it's already enough to kill the mob, which is why your evis damage is well under your rudra damage. That's why you need a mob that you can beat on for an extended amount of time for this kind of test.

If you want to calculate damage the mathy way, we can do that too.

total damage = ((dmg + fstr + wsc) * fTP) * pdif

For Rudra (Aeneas/Twashtar)

Main hand
= (140 + 23 + (.8*444)) * 8.75
= 4532
= 4532 * 1.32 (+wsdmg)
= 5982
Offhand
= (140 + 23 + (.8*444))
= 518
Total
= 6500
= 6500 * 3.85 (pdif)
= 25025

Evis (Kaja/Twashtar)

Mainhand
= 117 + 21 + (449 * .5) * 1.45
= 525.62
= 525.62 * 5 (5 hits)
= 2625
Offhand
= 117 + 21 + (449 * .5)
= 362
Total
= 2987

= (2987 * 4.85 * 1.47 * .63) (crit) + (2987 * 3.85 * 1 * .53) (non-crit)
= 13416 + 6094
= 19510
= 19510 * 1.5 (kaja bonus)
= 29265

Don't take that as set in stone. I can wake up tomorrow, do the same calculations again and get slightly different answers. Though, I've done this calculations a few times and I always get the same outcome. Evis was never close to Rudra pre-Kaja. It always bugged me to hear people say that, since neither the math or numerous tests I've done showed that. Evis does pull ahead now though, with the new dagger, and both my math and my practical test (on Tax'et) shows that. Aeneas is a better dagger though, with higher dmg and store tp+10, so that closes the gap between them.

It's nice when practical tests and theoretical calculations line up like this.

For those of us that don't have Twash offhand, or any REMA... would Skinflayer with Crit rate or Crit Damage and 10-15 DEX beat Taming Sari for offhand with Kaja main?

I know Taming Sari offhand has been the gold standard for a long time, but if the goal is to boost Evis damage, does Skinflayer make more sense?

Crit rate is more likely to have a higher impact than crit damage. Path D shijo is actually a notch ahead of skinflayer though, and it came up in discussion a couple pages back. If we're excluding Taming Sari in addition to REMA then Kaja Knife's ideal offhand choices are Path D Shijo --> Skinflayer --> Everything else. The difference between shijo and skinflayer isn't very much though, so if you already have skinflayer but not shijo and manage to snag a full crit rate augment with some other good stats I'd probably keep that and call it a day.

So Taming Sari would still be the best OH?


*edit:
Also, has anyone considered augmented Sandung? Triple attack and Crit Rate +5% Ignore this

That would depend on how good of an augment your sari has. I believe back when taming first came out we determined that a perfect taming sari is better than shijo, but if it's a weak augment shijo is the better option. It's been so long though I can't remember what the cutoff point was anymore. Maybe someone else can clarify for me.

But that was then.. i'm strictly asking if it's still the best for Evisceration and Kaja Knife main.

this is a bit off. 63 and 53 don't equal 100, I assume those should be your rates, at least. also, where are you getting 3.85 pdif from? it's certainly not an average, it's almost max roll with neck and it looks like you're using fotia in the previous bit.

The math should still hold up. Kaja Knife in the main hand is going to apply the same 50% bonus to evisceration regardless which dagger is in your offhand. It's associative, so if a perfect taming sari beat a shijo back then, it would still do so now. I'm certain we determined you only needed a mid range taming augment for it to pull ahead of shijo.

The order was mid range or higher Taming --> Shijo Path D --> Skinflayer crit rate --> everything else

That's what I'd aim for as a non rema option for Kaja offhand. A perfect taming sari augment is still the best non REMA Kaja knife offhand, but getting that aug may or may not be more hassle than it's worth. I got mine before they changed sinister reign, so maybe with the recent changes from several patches ago the process to get a flawless taming aug is a bit less grueling than it used to be. After what I went through the first time I have no intention of going back to find out....

EDIT Found the old general guide I was looking for waaaay back on page 127. Jeanpaul gave a really detailed writeup on daggers at the time. The changes to REMA since then have altered their statuses, but the hierarchy for non REMA daggers is still valid today. The post is 5 down from the top.

https://www.ffxiah.com/forum/topic/36654/for-the-shinies-a-guide-for-thief/127/

I already knew the heiarchy from before, but I was assuming that all of that was calculated using Rudra's Storm as the main WS, not Evisceration. In theory, Evis wouldn't gain as much from having Taming Sari offhand as it would with something that has similar Dex but more crit rate/damage, with a little bit of Triple Attack. That was my question, not what weapon was better than the others in the past.

The math shouldn't still hold up, since the focus is on Evis not Rudra's Storm.

If the math does still hold up, i'd like to know. That's why i'm asking. Has anyone done the math using Evisceration as the primary WS, not Rudra's Storm?