Tree of Savior Forum

Solving the Formula : The mystery of ^0.x

General Discussion
#22

I believe i did an error in my earlier math so disregard.

650 x log10((650 / 644 + 1) ^ 0.x + 1)
= 213

for this I am getting x = 0.17

To solve this just requires a little algebraic manipulation and knowledge of logarithmic properties.

First: divide 213 by 650

213/650 = log((650/644 + 1)^x +1)

Second: since its base 10, you set both sides as exponents of 10 and solve, canceling out the log on the right

10^(213/650) =2.12663 = (650/644 + 1)^x +1

Third: Subtract 1 from both sides

1.12663 = (650/644 +1)^x

Fourth: Take the natural log of each side(doesn’t really matter if its base e or 10 or whatever)

ln(1.12663)= ln(650/644 + 1)^x

Since its a log you can bring the exponent down as a factor

ln(1.12663)=x * ln(650/644 + 1)

and simplify

ln(1.12663)/ln(650/644 + 1) = x
0.170868…= x

1 Like
#23

Yeah I thought there was an error earlier, that’s why I was wondering where the numbers came from.

Alright so 0.1708 it is. That’s what I found too…although in a more “bruteforce” way.

cough

2 Likes
#24

Fact is, this proves the “accuracy” of IMC’s released info :expressionless:

1 Like
#25

@TheAlleyCats

I edited my above posts due to error.

However a factor of 0.17 in this experiment is constant. More numbers would be needed to absolutely confirm.

1 Like
#26

You can try with the image linked in the first post

The Attack i had when hitting the mob, the Defense of that mob, and the factor I ended up with. Should all be fairly accurate I believe.

Yea this one was messed up because Imps are Ghost Armour. Despite weaponless attacks having no type, Ghost still infers a penalty on them, unlike Plate, Cloth and Leather having no effect on my damage. So 146 is after ghost penalty.
(which, if you’ve seen some of my other posts about slash/strike/pierce, is hella not constant either lul).

1 Like
#27

Yes, more samples are needed to confirm it, but it’s sure as hell that the 0.9 they used for the damage tables is not a constant, nor the 0.8.

1 Like
#28

Edit: There was an error in the data.
This was fixed in a later post below.

*** I increased the magic attack given by Swell Brain by 1 point (from 109 to 110) and yet the Character Stat display did not increase from 921 to 922 attack. The character sheet or Tooltip for swell brain may of rounded. (going from 10 to 11 SPR)

Now for another mob
Orange Sakmoli’s have 157 MDefense, exactly 400 less then the Kepa magician raider.
Again. Damage dealt on left. Mattack on right.
495 = 696
740 = 922
965 = 1115

Seedmia’s have 57 Mdef. 500 less then the Kepa Magician Raider and 100 less than the Orange Sakmoli’s.
696 = 696
922 = 922
1115 = 1115
Obviously I hit the MIN value in the calculation.

Taking a break.
Anyways hopefully that’s a decent enough dataset to find the trend although I think using a skill would be more accurate since you can see larger changes.

If anyone else wants to take over
I recommend resetting your mage to a Thaumaturge since you can increase M-attack by specific increments using Thauma buffs and SPR.

1 Like
#29

166 x log10((166 / 557 + 1) ^ (-0.8) + 1) = 42
292 x log10((292 / 557 + 1) ^ 0.01 + 1) = 88
523 x log10((523 / 557 + 1) ^ 0.54 + 1) = 201
649 x log10((649 / 557 + 1) ^ 0.644 + 1) = 274
696 x log10((696 / 557 + 1) ^ 0.673 + 1) = 303
802 x log10((802 / 557 + 1) ^ 0.726 + 1) = 372
…
1877 x log10((1877 / 557 + 1) ^ 0.889 + 1) = 1263

You can tell that it’s going to hit ^ 1 eventually, which is when your damage = attack and you hit that MIN limiter in the formula.

1 Like
#30

I think we should use excel sheet to handle the data. Input Attack, Defend, Damage, skill % and use this formula to calculate 0.x
0.x = ln(10^(damage/(attack*skill%))-1)/ln(attack/defend+1)

Maybe we can draw a diagraph with defend and 0.x and see what kind of curve it is.

1 Like
#31

Updated first post with a google docs sheet, sorted by factor, lowest to highest

I can honestly see why the people who post the dossiers would mess up the formula. Assuming the devs didnt give them the exact specifics, with high enough attack and “regular” same level mob defense, the value is going to hover around 0.8 and 0.9.

1 Like
#32

Seem good. Did you try using skill? Is 0.x consistent against the same monster or not?

#33

It remains consistent. This makes it very easy to find other modifiers (refer to my modifier thread) because my skill damage is always my Z attack damage * skill%. I don’t need to go trough this part of the formula to figure those out.

I DO need this part to figure out Armor Break, Shield Push, things like Pommel Beat/Double Slash/Stomping Kick/Shield Charge, since they affect attack or defense within this part.

1 Like
#34

Ah, I see, so the only work remained is solve these data.

1 Like
#35

I am out again, but I can give it another go when I get back in excel. It is wierd that it has so many scaling variables.

But then again, these are the same people that thought DPK was a good idea too.

2 Likes
#36

It’s not actually that complicated.

Defense goes down, Factor goes up.
Attack goes up, Factor goes up.

It’s just the relation between defense/attack that scales, we simply don’t know how yet. ;x;

#37

See what the ratio is why you have exactly Double their Defense or I suppose double Def + 1.

#38

i founded d out the mystry ur is 2 x 2 = 6

stop posting now u cant anymore i solved this mrk this as a solution thx

#39

Did you just have a stroke ?

Also, this post is not constructive(and barely intelligible).

#40

lvl 186.
Zinutes (100 Mdef)
Damage on left. Mattack on right

47 = 133
134 = 259
340 = 485
896 = 965
1063 = 1094
1248 =1248
Around 11-12 times MDef to reach true damage.
Meh.

Something hopefully Cleaner
Eldigo (99 Mdef… so it should end up as 100) Going to try to solve for its “True” damage point.
50 = 133/132 ***
232 = 359
685 =768
823 = 877
977 = 995
983 = 999
993 = 1007
1012 = 1021
1032 = 1036
1042 = 1043
1043 = 1044
1044 = 1045
1046 = 1046

1046 Mattack is the True Damage point for a 99 MDefense enemy (unless there’s any funky rounding going on behind the scenes).
*** Char display has showed both despite 0 equip on. There is probably a rounding discrepancy due to level.

My Kepa Raider data was faulty. Redid it from scratch lvl 186 (enemy 577 Def)

15 = 133
17 =142
48 = 255
50 = 259
53 = 268
108 =407
109 =409
109 = 410
110 = 411
161 = 519
162 = 520
162 = 521
163 = 522
163 =523
168 = 532
169= 533
169 = 534
170 = 535
170 = 536
172 =539
241 = 663
246 = 672
305 =769
311 = 778
416 = 936
417 = 937
418 = 938
419 = 940
420 = 942
421 = 943
422 = 944
422 = 945
423 = 946
492 = 1042
493 = 1043
495 = 1046
496 = 1047
496 =1048
497 = 1049
498 = 1050
498 = 1051
499 =1052
500 = 1053
558 = 1131
648 =1247
655 = 1256
667 - 1271
753 = 1376
754 = 1377
848 = 1489
878 =1524
895 = 1544
897 = 1546
991 = 1652
1066 = 1736
1164 = 1842

#41

I think I found something. in each individual case, I found that Factor is linear with Def/Att, but the coefficient differ in each case. Maybe it is affected by something else. Hope this help.