Tree of Savior Forum

[kTest[Ktos]] Class rework thread

Critical Attack does get multiplied by skill factor. I tried it on ktest…

damage = %factor(skill) x attack x log10(attack / (defense + 1) + 1)^0,9 + additional damage

Full formula is more likely to be:

non-crit attack = %factor(skill) x attack x log10(attack / (defense + 1) + 1)^0,9 + additional damage

crit attack = %factor(skill) x [(attack * 1,5) + critical attack] x log10([(attack * 1,5) + critical attack] / (defense + 1) + 1)^0,9 + additional damage

If we’d like to put it into a single formula, we must fill in an “IF” value.

damage = %factor(skill) x attack x log10(attack / (defense + 1) + 1)^0,9 + additional damage

attack = [IF Critical = True, return [(Physical Attack * 1,5) + Critical Attack] else [Physical Attack]

2 Likes

Difficult to compare, because it will depend on gear.
Shinobi build damage is nice. You must have clones activated to have optimal performance.
Clones copy swordie double slash, bash, thrust, pommel beat. And Highlander crown, moulinet.

Dop3 is very popular. Sword3 high3 build is very rare. Recommend you to test the builds when we get reset.

@FONTOURAO Have few dragoon vids in first post, maybe you could get some ideas for builds after watching some.
Some I have seen
Sword3 Cata3 drag2
Sword3 Hop3 drag2
Pelt hop3 dop drag2
Sword3 barb3 drag2
Sword3 high3 drag2
Maybe you could try non-spear too during reset, you could like it. This is a good opportunity.

1 Like

Critical Attack was never multiplied by anything universal. I think the only classes that have something like that is Highlander 2h mastery (+25% of the stat on new patch) and Fencer Epee Garde (+50% of the whole critical damage on new patch).

The only advantage of focusing on Critical Attack is actually its disadvantage as well:
By not being inside the formula, it can’t be reduced by normal defenses.

Therefore, for those who do not have crit dmg skill modifiers, it will be like using an arde dagger (its damage can’t be reduced normally as well) with the exception that it only procs when you crit and can be raised with stat and items.

Except as I said above (With video you can watch for evidence), Crit Attack is NOT factored into the log calculation.
Neither does your attack get 1.5x on crit for the log part, it’s only your attack on the stat sheet, nothing else.

Feel free to present your own proof otherwise, though that’d mean IMC change the formula again in the short time between that video and your testing.

No longer true in the new calculation. +property attack on the other hand does ignore defense (on the flip side it does not benefit from skill factor)

This is almost everyone before and definitely after the revamp hits:

Edit: I removed a redundant [and]. :confounded:

3 Likes

EDIT: Ok I jumped a little and now I read what you all said.

What makes you think that critical damage is jut not only the old formula added in the end like it is now? Or like a property attack for that matter?


If you’re talking about this formula:

crit attack = %factor(skill) x [(attack * 1,5) + critical attack] x log10([(attack * 1,5) + critical attack] / (defense + 1) + 1)^0,9 + additional damage

I would say not quite, because you have two parts, one that is reduced:

crit attack = %factor(skill) x [(attack * 1,5) + critical attack] x log10([(attack * 1,5) + critical attack] / (defense + 1) + 1)^0,9 + additional damage

and another which isn’t

crit attack = %factor(skill) x [(attack * 1,5) + critical attack] x log10([(attack * 1,5) + critical attack] / (defense + 1) + 1)^0,9 + additional damage

If it is another, please tell me because I did not find the formula on dev blogs nor in any other source, which is something that interest me.

I have a question with the new patch coming, the wiz2 and wiz3 still necessary for a good elementalist and if not Cryo3-Ele3-RC​ could be good for solo?

So, let’s take into account the guy is hitting Lv291 Vilkas: https://tos-ktest.neet.tv/npcs/58592

DEF = 1123

STR Video: 1648~1675 P.Atk, 20 C. Atk
DEX Video: 1087~1114 P.Atk, 1612 C.Atk

Now I’m not sure what level Pouncing he has or Attribute, but let’s consider it Lv5 with no attribute: 144%

STR case would result in 1564 Damage while DEX 1572 Damage.

So yeah, rofldat is correct about the log part (it’s only “attack” multiplied there), while Skill Factor is applied to “Attack x 1.5 + Critical Attack”.

I achieved those results with this formula:

damage = %factor(skill) x attack x log10(attack / (defense + 1) + 1)^0,9 + additional damage

attack = [IF Critical = True, return [(Physical Attack * 1,5) + Critical Attack] else [Physical Attack]

The guy in the video might have a Lv30 attribute on that though, we can’t check that.

I can’t say whether this is the exact formula or not, but about Crit Attack multiplying with Skill Factor, it does, I’m 100% sure about it. On kTest I tried both full STR and full DEX archer classes (especially those with +(Critical Chance)% skills). I can say that Appraiser’s Blindside full DEX was hitting like a truck. It wouldn’t if Crit. Attack was just summed as additional damage.

2 Likes

Vid
Some monk TBL PVP

Monk double punch destroying dummy [here] (https://www.youtube.com/watch?v=DDGUrVqQ1r8)

2 Likes

May i know if this is full dex? the punch speed seems really fast.

1 Like

Yes, high dex monk.
Double punch speed is affected by dex (and connection).

Monstrance boost +30% dex, and +10 dex. A lot of speed.

Example of dex monk vs mobs [here] (https://www.youtube.com/watch?v=akVSro2Iyeo&t=506)


vid

Earthtower

  1. Wizard c1, Cryomancer c3, Chronomancer c3, Sage c1
  2. Wizard c2, Linker c3, Chronomancer c3
  3. Archer c2, Sapper c2, Fletcher c3, Mergen c1
  4. Swordman c1, Highlander c3, Barbarian c1, Doppelsoeldner c3
  5. Cleric c3, Priest c3, Miko c1, Daoshi c1

Former fantasy library
Chrono3
chaplain taoist
highlander3 doppel
falcon3
sapper3 appraiser cannon

Sword3 barb3 goon2

And

Sword3 hop3 goon2
Sword3 pain barrier recommended for hoplite3 because your evasion is very low with finestra buff.
Bosses will knock you a lot.

1 Like

Fking hate the colour scheme of this forum, anyway. (Grey on white for creating a new message. Plus mild dyslexia means I’m not good with black text on white background anyway. All my papers are done in yellow. (Colour overlays on pc’s are horrible))

Ok eff this crap. Messed up multiple things in my calculation.
Dex is useless other than evasion and attack speed.
In terms of damage Str far outweighs Dex. (1 I forgot to take into account 1.5x attack, 2 I forgot to take into account the damage from not critting to calculate and average damage)
With 1k attack against a 200 def enemy. << Str is from Wep or something. (Not stat based.)
Pure Str does 1.9k Average
Pure Dex does 1.7k Average
50/50 does 1.8k Average.
So on the assumption you have a 60% Crit rate and you benefit from the attack speed boost:
Dex > Str
However if you have neither of those
Str > Dex
One thing to notice though is that that’s on average 100 less damage for pure dex but you get an additional 80 evasion. Which assuming your evasion was already equal to the enemy accuracy, that’s a 17% evasion chance obtained from lowering your average damage by 100. (Your crit hits will be about 300 higher, but your none crits will be about 500 lower.)
Again though, remove the high crit rate and Str > Dex.

Can someone test Priest3 Stone skin? I really want to know its efficiency after the change.

Stone skin: Reduces Slash/Pierce/Bow/Pistol Damage dealt to you and your party members. 20% Reduction at level 5

I see, but what about the total damage reduction?
Reduces damage to yourself, your party members, stabbing, bowing, and total attack

1 Like

It will only reduce damage of slash, pierce, arrow, bullet type attacks.

Theory is nice and all, but i guess it will be better to just wait a few days where ppl can actually record the tests to see how crit atk and magic amp act with the new formula, and then no one can refute.

Well, at KtoS general Thread this crit damage discussion actually took place while ktest first received the rebalcing patch and the very first “new damage formula” was released by IMC.

Some dude playing on ktest with a Ranger3RogueHackka provided a lot of data hitting low def mobs then. I still hd access to only an incomplete version of damage formula (heck we were even trying to deduce what the 0.x meant in the formula).

But results were pretty consistent with damage on critical taking transforming the ATTACK part of damage formula to (1.5*ATK+CritATK), but not affecting the log part. Though, to be honest, I don’t have that spreadsheet anymore to test how it’d go with 0.x=0.8, the min argument used there and the +1 on both def and outside the atk/def ratio…

What is sure though is that crit atk is affected by both Skill Factors and target’s def. I’m just not sure anymore whether it also changes log part or not.

PvP vid
Monk at 2v2 (several matches)

3 Likes