Blade: Chi Damage Formula

Langues: JP EN FR DE

users online

Forum » FFXI » Jobs » Ninja » Blade: Chi Damage Formula

Blade: Chi Damage Formula

First Page 2

Bahadir

Offline

Posts: 229

By Bahadir 2016-10-01 09:12:38

Link | Citer | R

Verda said: »

Teki is too, both were listed in the wiki as magical rather than hybrid. Also only the wiki seems to list them as elemental attack the in game text says elemental damage.

But your Chi dmg was not too far-off the To dmg. And except for a slightly higher WSC Blade: To should be at a clear disadvantage due to one less base hit (with same fTP). So how come the Chi dmg isnt much higher than the Blade: To dmg?

Verda

Offline

By Verda 2016-10-01 09:35:54

Link | Citer | R

Bahadir said: »

Verda said: »

Teki is too, both were listed in the wiki as magical rather than hybrid. Also only the wiki seems to list them as elemental attack the in game text says elemental damage.

Different sets and buffs. I was using a mab set for the To and also had more attack boosts from a 4 song aeonic bard for the To shot. For the Chi shot I was solo and using more a physical ws set and not even a good one. Also different mob families so lots of uncontrolled variables.

Bahadir said: »

Edited my former post to add more Int+->dmg values and checked as well: Clipper Int is 20. Checked on many of them around the tower and seems like if there are different lvls they all have Int 20.

Checked as well the only magic dmg Doll NM with Bio II and it shows now MDB or MDT. Just to yield remaining info about my post even before that.

Then the damage spikes at a dINT of 95?

Bahadir

Offline

Posts: 229

By Bahadir 2016-10-01 09:51:32

Link | Citer | R

Yes. The first strange jump is at dInt95. But there is a second strange transition i found so far at dInt103 where the dmg increase is 4dmg instead of 3. However, 3->4 is easier to explain than 3->5. Ill keep working on it ^^

[+]

Verda

Offline

By Verda 2016-10-01 10:07:20

Link | Citer | R

Bahadir said: »

556, 559, 562, 565, 568, 571, 574, 577, 580, 585 (yes...checked 5 times or so. its really 585), 588, 591, 594, 597, 600, 603, 606, 610, 613, 616

With clipper INT of 20 (how'd you test that btw ? :D)

dINT 86 = 556
dINT 87 = 559

Bahadir said: »

Dmg = floor(floor((V+dInt+mDmg)*MAB/MDB)*NinjutsuBonus)

with the following values:
V = 69 (for Ni spells)
dInt = 102 (my Int)-X (to find out)
MAB = 1.18 (gifts)
MDB = 1 (assume they dont have MDB)
NinBonus = 2.12 (446 Skill so caped)
mDmg = 20 (JPs have 10 of them)

The real dmg I get is
427 and 444 with Int+7.
I cannot find an Int/MDB combination that gives me exactly that numbers. What am I missing here?

EDIT: 444 dmg was Int+7!! not +9....just 109 in total.

dINT 86: floor(floor(69 + 86 + 20)*1.5)*2.12)=555.44 (Off by one if we floor, this kind of thing is common with estimates it may be because rather than a floor function the game uses some odd rounding function or 1.5 and 2.12 aren't accurate representations of their base 2 numbers).
dINT 87: floor(floor(69 + 87 + 20)*1.5)*2.12)=559.68
dINT 88: floor(floor(177*1.5)*2.12)=561.8
dINT 89: floor(floor(178*1.5)*2.12)=566.04
dINT 90: floor(floor(179*1.5)*2.12)=568.16

So we can see the formula is nearly correct, minus either some base two multipliers being improperly represented or the flooring function actually being an odd rounding function (different computer systems will often treat rounding differently, was actually a problem for reports being penny wise accurate and took special coding to mimic rounding for one company I worked at). Though if I had to measure a guess it'd be just something to do with the numbers being 1024 base or only one flooring step the reason for the occasional off by one discrepancies.

So your equation works :D Good job. Also thanks for doing this :)

[+]

Bahadir

Offline

Posts: 229

By Bahadir 2016-10-01 10:28:56

Link | Citer | R

Verda said: »

With clipper INT of 20 (how'd you test that btw ? :D)

Rightfully sceptical by now ^^ checked with elemental magic. Stone, Water, Aero casts with 50 MAB and the assumption that my Int is at least 50 over the clippers Int yielded exactly 20 Int for all cases and it worked for other spells as well.

Ok I got that "off by one" a lot and was always eager to find a formula that didnt do that ^^ but if we accept that and blame it on "strange rounding" I guess it works fine.
If the formular was one too low it was usually "ok" for me. But I find it strange that in some cases its one too high. Like
dINT 89: floor(floor(178*1.5)*2.12)=566.04
While the actual dmg is 565. So even with the strictest form of rounding its too high. But yea maybe its a strange 1024 based representation.

Bahadir

Offline

Posts: 229

By Bahadir 2016-10-01 12:59:30

Link | Citer | R

Just a quick heads up. Still working on the exact formulars ^^ but just found out that Blade: To does indeed have a first physical hit and magical dmg is added later. Now I cannot find numbers anymore but using it at mobs full HP compared to a mob that was almost dead dealed like 4 times the dmg.
Ill get more details later.

EDIT: was 305 vs 1428 dmg

Bahadir

Offline

Posts: 229

By Bahadir 2016-10-07 13:54:41

Link | Citer | R

Sooo....went back to slaying maaaany crabs and found some stuff out about hybrid WSs.

I tested with Blade: To (only one hit = easier to test).
Things I found out (about Blade: To. Need to check things like fTP for Chi still):

i) WSCs given on bgWiki are correct
ii) fTP is only applied to magical dmg, NOT physical one (see formula below).
iii) for Blade:To the fTP for phys is just a static 1
iv) for the magic part its fTP=0.5 for 1kTP, fTP=1.5 for 2kTP (educated guess) and fTP=2.5 for 3kTP.
v) The physical WS dmg is as stated on bgWiki with fTP=1 and multi hits like DA or the DW additional hit trigger as usual
vi) All phys hits are evaluated first (including TP return) and if the monster still lives after the last hit magical dmg is added like an add-effect without extra TP gain

Formula: TotalDmg = PhysDmg + (PhysDmg*fTP+mDmg)*MAB/MDB
where PhysDmg is the usual thing with WSCs and fSTR and pDif and so on. fTP is as mentioned in point iv).

Again experimental results if you care about numbers:

Spoiler Alert! Click to view.

We d have to test for the other hybrid WSs if the fTP_phys is always just 1. But in general Id guess the idea is always: evaluate all phys hits and take that as base for the magical add-effect. That explains Verdas dmg spikes (TA/DA procs). I bet Flaming Crush works the same way.

[+]

Verda

Offline

By Verda 2016-10-07 14:12:55

Link | Citer | R

awesome :D

As an addendum, I tested mikage and blade: teki, to and chi and got no huge consistent damage so I'm thinking mikage doesn't apply to ws or something :<

[+]

Bahadir

Offline

Posts: 229

By Bahadir 2016-10-07 14:25:52

Link | Citer | R

Oh good point. Didnt test Mikage. Not sure how that works with WS...

Verda

Offline

By Verda 2016-10-12 12:19:30

Link | Citer | R

https://www.bluegartr.com/threads/121610-Rehauled-Weapon-Skills-tier-lists?p=6143442&viewfull=1#post6143442

That post might interest you it seems to imply (at least to me) what you said about the physical hits not having any fTP associated, for flaming arrow, at least if that was just testing the first hits, it's the only way it'd be 1 ftp every time I can think of.

Bahadir

Offline

Original Post - Cacher

By 2016-09-27 18:56:32

Delete | Edit | Link | Citer | R

I was intrigued by Verdas numbers with Blade: Chi so I ran some tests in order to understand how the dmg is calculated and I have to admit that I am at a loss cause I cannt figure it out. Ill post my test results and maybe one of you guys has an idea.

Here are my base assumptions:
- Blade:Chi has Mods 30% for Int and Str.
- fTP at 3000TP is 1.0 (all WS done during my tests were performed at 3k TP)
- pDif was capped all the time (used trust Dia III + Light Shot to ensure caped pDif)
- fStr was caped at 8 (Rank 0 weapon)

All tests were done with kunai as main hand weapon and other slots were empty (unless stated otherwise). 99 Nin with varing subs depending on situation. Sub and exact Str/Int/MAB stats are given below for the different tests. Unfortunately, I already have some gifts so I have 18 base MAB that I cannot remove.

Things I tested:
Hypothesis
H1) The first hit follows the phys WS formular
H2) MAB affects the 2nd hit
H3) The player level affects the dmg (as with other magical WS)
H4) There is a dStat term for the dmg of the 2nd hit
H5) The dmg of the first hit influences the dmg of the 2nd.

Short answeres:
Yes, yes, no, no, yes.

Long explanation

Spoiler Alert! Click to view.

Here are some numbers:

Test 1: (Base numbers and check 1st hit)
99Nin/Blu, Escha - Zi'Tah, Target Eschan Obdella, Str 101+18, Int98+18, MAB14+18 (Vorseals included)
Blade: Chi dmg numbers
2409, 2357, 2010 (outlier, no idea how...maybe pDif wasnt caped due to Att debuff or something), 2388, 2227, 2362, 2380, 2349, 2388, 2410
Average: 2363
Dmg of the first hit:
272, 279, 276, 269, 270, 272, 269
Average: 272

Test 2: (Check a different player lvl)
78Nin/Blu, Escha - Zi'Tah, Target Eschan Obdella, Str 84+18+17, Int81+18+17, MAB214+20 (MAB I job trait) couldnt get exactly 18 MAB in gear on 78
2412, 2369, 2457
269 273 270
Stopped after 3 WS each as numbers were almost exactly like the lvl 99 ones (only slighly higher on 2nd hit as was expected due to MAB+2)

Test 3: (just put these numbers cause of the resist!)
With a lvl 85 Kannagi (52 weapon dmg) and slightly different stats I got a very interesting result for complete Chi dmg (forgot to write down exact numbers :/)
4394, 4383, 4356, 2597, 4354
Checked TP return and it was indeed a resist! no miss! But the only one I got during my tests.

Test 4:(Check different MAB values)
99Nin/Blu, Escha - Zi'Tah, Target Eschan Obdella, Str 101+18, Int98+18, MAB14+18 (Vorseals included)
Varied the number of MAB I added with gear but Int and Str remained the same:
Given are both hit numbers as 1st hit only numbers were same as in Test 1.
MAB+10
2515, 2529, 2551, 2539, 2488, 2502, 2478, 2474, 2496, 2488
Average: 2506
MAB+20
2633, 2580, 2605, 2629, 2633, 2657, 2652, 2633, 2676, 2624
Average: 2632
MAB+30
2736, 2761, 2740, 2756, 2747, 2716, 2767, 2751, 2757, 2731
Average: 2746

Test 5: (Increase only first hits dmg)
99Nin/Thf, Qufim Island, Target Clipper (tested and they have roughly 20 Int), Str 102, Int 101, MAB 18
First Hit:
245, 241, 234, 236, 243
Average: 240
Both hits:
1772, 1746, 1743, 1746, 1696,
Average: 1741
With SA:
1998, 2053, 1998, 1910, 1949
Average: 1982
One hit with SA would yield 306-312 (couldnt test but used known formulas).

Test 6: (Change monster stats but keep pDif caped)
99Nin/Thf, Misareaux Coast, Target Big Claw (tested and they have roughly 35 Int), Str 102, Int 101, MAB 18
Average Dmg was almost the same as in Test 5 for first hit and both hits. didnt test SA WS.

I am aware that the tests were quit limited. However, while prob not statistically sufficient they are enough to make some strong guesses about H1-H5)
About H1) All the 1-st Hit numbers exactly match the known phys WS dmg formula: Dmg = (D+WSC+fSTR)*fTP+pDif so first hit seems a normal physical hit.
H2) Test 4 shows that MAB has an impact on (only) the 2nd hit.
H3) Test 1 and 2 show that a reduction of the lvl did not(!) change the dmg, unlike for magical WS.
H4) Tests 5 and 6 show same dmg despite different targets with different stats. So either no dStat term or its capped quickly, which would be strange as dStat terms for magical WSs are usually uncapped, right?
H5) By using SA I increased the pDif of the first and only the first hit. So the dmg in Test 5 increased. However, the final dmg increased more than the dmg of the first hit did. Thus, the 2nd hit depends on the dmg of the 1st hit.

Having said all this...I cannot come up with a formula that explains all numbers. Any1 has an idea?

Top Show Op

First Page 2

From:
To:
Subject:
Body: