Samurai Weaponskills: STR+ versus ATK+

[Jeryhn]

Basically, pick a monster. Now add attack to your setup until it stops checking high defense. At that single point of attack, when added, it stops checking high defense, that is the monster's defense. You've achieved a 1/1 atk/def ratio at that point.

How is this true? Can't the lack of a defense message when checking a monster assume a range of attack/defense difference in the same way the actual check ratings do? For example, a monster that checks Tough to a player can represent that the monster is several levels higher than the player before the mob checks as "Very tough".

[aurik]

There is a range where you get no message (neither high or low defense) when checking a monster.

Your attack is equal to his defense at the very bottom of the range.

[Kawfee]

Thats a great way to determine defense, I never thought of that.
Also, on the topic of pDIF, minus the fact I was ignoring that it caps, that it is a "range" I think is a non-factor. For the sake of comparison, if you assume that it is at either the highest or average, as long as you do it the same for every weapon piece, the relative differences are still valid.

Thanks to Aurik (forgot?) for the Tiamat numbers, those look about right.


Also, for those who look at these calculations and are like, "How can I use these for my own WS's on another job?"

Essentially, it's not hard to adjust them. The part in my equation where its like "LevelAdjVariable * WeaponSkillModifier% * WeaponSkillModifierStat" is just the particular modifier for your level and weapon skill. IE: For NIN's Jin, I think its 30% STR and 30% DEX? (if not, you can figure out w/e):

Rather than mine, which for SAM reads ".83 * .75 * YourSTR" it would be:
.83 * (.3 * YourSTR + .3 * YourDEX)

Then, you have to adjust the TPModifier to be whatever your TP modifier is, its different for every weapon skill at 300% TP. (For SAM its the 2.5 number)

Lastly, if you are using multihit weapon skills, you have multiple sets of the entire equation. However, for all the hits after hit #1, you change the TPModifier to just be 1.0. I think. You can also use this for double attack damage

[archibaldcrane]

Basically, pick a monster. Now add attack to your setup until it stops checking high defense. At that single point of attack, when added, it stops checking high defense, that is the monster's defense. You've achieved a 1/1 atk/def ratio at that point.

How is this true? Can't the lack of a defense message when checking a monster assume a range of attack/defense difference in the same way the actual check ratings do? For example, a monster that checks Tough to a player can represent that the monster is several levels higher than the player before the mob checks as "Very tough".

You're right, the /check system is ONLY for determining a mobs defense for our purposes. It does not determine level. To determine level for mobs above 75, you'll have to extrapolate backwards post-humously from the experience points gained. I'm sure there a chart that would make that relatively painless.

However, if one was to create a list of, say, the DEF range of all the Jnuns in Cadaerva Mire, for instance, I would bet that with proper /checking, you could determine the exact defense, and therefore know the exact level once it's been figured out. (I picked Jnuns because they don't idly buff themselves while laying in the muck, like Imps do - flies would work fine also)

[VZX]

Now, if this is correct, Viper Biting a Ronfaure rabbit at 75, without SA or TA should give you the exact same damage range as a regular swing.

well, if it can go push number to 3.0 without a necessary critical, then yes.

But hey, it's level 1 monster anyway, have your attack doubled or not, its ratio with respect to monster defense already go beyond 2.0.

Maybe you can try on EP mob and see the result.

http://www.killingifrit.com/forums.php?m=posts&q=131514

Thanks

[Kawfee]

Taking into consideration what everyone has said, here is a recalculated table.

Assumptions:

• Player is SAM75/WAR37 (at this point, since SA gives criticals, which changes the formula's pDIF aka Attack/Defense/Level adjustments).[/*:m:b09ea]
• Player has 540 attack. Mob has 300 defense & 75 VIT (For now. I'll do Tiamat too afterwards).[/*:m:b09ea]
• +2 STR also gives +1 Attack[/*:m:b09ea]
• Base player STR from other gear boosts will be 100. Hagun damage = 75.[/*:m:b09ea]

Reposting the "X" Table just in case:

STR-VIT:  then X =
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
   >12         4
 6 ~ 11        6
 1 ~ 5         7
 0 ~ -2        8
-3 ~ -7        9
-8 ~ -15      10
-16 ~ -21     12
  <-22        13

With Byrnie+1, STR+5, Attack+25 +2 from STR = +27
-> STR: 105 Attack: 567

=( 75 + ( 105 - 75 + 4) / 2 + .83*.75* 105 ) * 2.5 * ( 567 / 300 - .15 )
=( 75 + ( 34 / 2 ) + .83 * .75 * 105 ) * 2.5 * ( 1.89 - .15 )
=(75 + 17 + 65 ) * 2.5 * ( 1.74 )
= 157 * 2.5 * 1.74
= 682.95 damage with Byrnie+1


With Haubergeon+1, STR+6, Attack+12 +3 from STR = +15
-> STR: 106 Attack: 555

=( 75 + ( 106 - 75 + 4 ) / 2 + .83 * .75 * 106 ) * 2.5 * ( 555/300 - .15 )
=( 75 + ( 35 / 2 ) + .83 * .75 * 106 ) * 2.5 * ( 1.85 - .15 )
=(75 + 17.5 + 65) * 2.5 * ( 1.7 )
= 669.375 damage with Hauby+1


With Byrnie, STR+3, Attack+20 +1 from STR = +21
-> STR: 103 Attack: 561

=( 75 + ( 103 - 75 + 4 ) / 2 + .83 * .75 * 103 ) * 2.5 * ( 561 / 300 - .15 )
=( 75 + ( 32 / 2 ) + 64 ) * 2.5 * ( 1.72 )
= 155 * 2.5 * 1.72
= 666.5 damage with Byrnie


With Kirin's Osode, STR+10, Attack+0 +5 from STR = +5
-> STR: 110 Attack: 545

=( 75 + ( 110 - 75 + 4 ) / 2 + .83 * .75 * 110 ) * 2.5 * ( 545 / 300 - .15 )
=( 75 + ( 39 / 2 ) + 68 ) * 2.5 * ( 1.66666666667 )
= 162.5 * 2.5 * 1.666666667
= 677.08 damage with Kirin's Osode


With Haubergeon, STR+5, Attack+10 +2 from STR = +12
-> STR: 105 Attack: 552

=( 75 + (105 - 75 + 4 ) / 2 + .83 * .75 * 105 ) * 2.5 * ( 552 / 300 - .15 )
=( 75 + ( 34 / 2 ) + 65 ) * 2.5 * ( 1.69 )
= 157 * 2.5 * 1.69
= 663.32 damage with Haubergeon


With Shura Togi, STR+0, Attack+20
-> STR: 100 Attack: 560

=( 75 + ( 100 - 75 + 4 ) / 2 + .83 * .75 * 100 ) * 2.5 * ( 560 / 300 - .15 )
=( 75 + ( 29 / 2 ) + 62 ) * 2.5 * ( 1.716666666666667 )
= 151.5 * 2.5 * 1.716666666667
= 650.18 damage with Shura Togi


With Hachiman Domaru / Usukane Haramaki, STR+8, Attack+0 +4 from STR = +4
-> STR: 108 Attack: 544

=( 75 + ( 108 - 75 + 4 ) / 2 + .83 * .75 * 108 ) * 2.5 * ( 544 / 300 - .15 )
= ( 75 + ( 37 / 2 ) + 67 ) * 2.5 * ( 544 / 300 - .15 )
= 160.5 * 2.5 * 1.6633333333333333333
= 667.41 damage with Hachiman Domaru or Usukane Haramaki



TOTALS:

Byrnie+1      682.95
Osode         677.08
Hauby+1       669.38
Hachiman      667.41
Byrnie        666.50
Hauby         663.32
Shura Togi    650.18

So, in conclusion, in an EXP Party Setup where you can maintain around 550ish attack vs. Sea Puks, Byrnie+1 is the best, then Osode, then Hauby+1, then Hachiman. Which is a rather large reversal of the earlier findings. I think these results should be consistent with an Attack of 700 even.

Submitting this so I don't accidentally delete it, and in the next post will be vs. a Tiamat type mob.

[archibaldcrane]

I think these results should be consistent with an Attack of 700 even.

Give it a run, just Byrnie+1 and Osode. I bet 5 gil that Osode wins.

[Kawfee]

Vs. Tiamat:

Assumptions:

• Tiamat's Level = 95[/*:m:42c6e]
• Tiamat's VIT = 132[/*:m:42c6e]
• Tiamat's DEF = 583[/*:m:42c6e]
• You are a SAM75/WAR37, because Sneak Attacking is stupid and Trick Attack has no bearing (unless Chiz chimes in).[/*:m:42c6e]
• Instead of " - .15" in the pDIF part, it will be:
  - .05 * (95 - 75)
  Which = - 1 (ewwww)[/*:m:42c6e]

With Byrnie+1, STR+5, Attack+25 +2 from STR = +27
-> STR: 105 Attack: 567

[b]=( 75 + ( 105 - 132 + 13 ) / 2 + .83 * .75 * 105 ) * 2.5 * ( 567 / 583 - 1 )
=( 75 + -14/2 + 65 ) * 2.5 * ( -.02 )

And now, we have a problem.
You're going to deal negative damage using these numbers. As such, my guess is that Tiamat's VIT is probably fairly accurate, but the defense is not. Because of this, new assumption:

Using http://wiki.ffxiclopedia.org/Defense as a source, I'm calculating Tiamat's Defense as:

Mob defense: [f(level,family defense rank) + 8 + (VIT/2) + job traits] * family multiplier

where f() = 153+[(Lv-50)*5.0]
Tiamat doesn't use Job Traits related to Defense (or if she does and I forgot, we'll say they aren't on <.<)
family multiplier = 1.0

Tiamat's Defense:
= 153 + ( ( 95 - 50 ) * 5.0 ) + 8 + ( 132 / 2 )
= 153 + 225 + 8 + 66
= 452


So, plugging and chugging again ~~~


With Byrnie+1, STR+5, Attack+25 +2 from STR = +27
-> STR: 105 Attack: 567

[b]=( 75 + ( 105 - 132 + 13 ) / 2 + .83 * .75 * 105 ) * 2.5 * ( 567 / 452 - 1 )
=( 75 + -14/2 + 65 ) * 2.5 * ( 0.25442477876106194690265486725664 )
= 84.59

Which can't be right.

I don't think that the equation for defense is correct. I guess the best way would be to work backwards from a known WS value and actual STR and Attack and see what VIT and defense numbers make sense :/

[Kawfee]

I think these results should be consistent with an Attack of 700 even.

Give it a run, just Byrnie+1 and Osode. I bet 5 gil that Osode wins.

Attack base of 700:

With Byrnie+1, STR+5, Attack+25 +2 from STR = +27
-> STR: 105 Attack: 727

=( 75 + ( 105 - 75 + 4) / 2 + .83*.75* 105 ) * 2.5 * ( 727 / 300 - .15 )
=( 75 + ( 34 / 2 ) + .83 * .75 * 105 ) * 2.5 * ( Value over 2.0 )
=(75 + 17 + 65 ) * 2.5 * ( 2.0 )
= 157 * 2.5 * 1.74
= 785 damage with Byrnie+1


With Kirin's Osode, STR+10, Attack+0 +5 from STR = +5
-> STR: 110 Attack: 705

=( 75 + ( 110 - 75 + 4 ) / 2 + .83 * .75 * 110 ) * 2.5 * ( 705 / 300 - .15 )
=( 75 + ( 39 / 2 ) + 68 ) * 2.5 * ( Value over 2.0)
= 162.5 * 2.5 * 2.0
= 812.15 damage with Kirin's Osode


At 700 Attack, Osode > Byrnie+1


Damn that pDIF cap messing with me lol. However, stack that with Sneak Attack, and I think this won't be the case.

The effect of Sneak Attack is to increase the cap of pDIF to 3.0, yes? Which would yield:


With Byrnie+1, STR+5, Attack+25 +2 from STR = +27
-> STR: 105 Attack: 727

=( 75 + ( 105 - 75 + 4) / 2 + .83*.75* 105 ) * 2.5 * ( 727 / 300 - .15 )
=( 75 + ( 34 / 2 ) + .83 * .75 * 105 ) * 2.5 * ( 2.273333333 )
=(75 + 17 + 65 ) * 2.5 * ( 2.2733333333 )
= 157 * 2.5 * 2.27333333
= 892 damage with Byrnie+1


With Kirin's Osode, STR+10, Attack+0 +5 from STR = +5
-> STR: 110 Attack: 705

=( 75 + ( 110 - 75 + 4 ) / 2 + .83 * .75 * 110 ) * 2.5 * ( 705 / 300 - .15 )
=( 75 + ( 39 / 2 ) + 68 ) * 2.5 * ( 2.2 )
= 162.5 * 2.5 * 2.2
= 893.75 damage with Kirin's Osode


...Okay, so you're right. Barely. Jerk.

But this also confuses me. Perhaps I'm doing something wrong. If you have Byrnie+1, 110 STR and 705 attack and SA+Gekko for 893... you fail at life?

[VZX]

I don't think that the equation for defense is correct. I guess the best way would be to work backwards from a known WS value and actual STR and Attack and see what VIT and defense numbers make sense :/

Actually, I realized this too long ago before I wanted to translate this information. Because I found there's inconsistency between the data they provide on table
and their formula finding.

Look at Lv61 crab(カニ) for comparison
From the table, it's listed that its def is 272.
OK, now assume that crab is the highest class defense enemy (A) and its level is 51 and beyond. so we use this formula:
153+[(Lv-50)*5.0] = 153+10*5.0 = 203.

Even with the VIT=78 (defense+39), its defense (203+39) still doesn't match up on their table

[Kawfee]

That being said, I'm still fairly confident in the numbers generated at 550 and 700 attack vs. a Sea Puk, as those seem at least pretty close to what you should be WSing for with no critical hit and no double attack.

The problem now lies in determining Tiamat's VIT and Defense and what I did wrong (if anything?) in my quick Sneak Attack evaluation.

[archibaldcrane]

The effect of Sneak Attack is to increase the cap of pDIF to 3.0, yes?

Yeeeeeeeeahhhhh kinda.

"Level adjusted melee ratio" gets plugged into VXZ's chart, then you get a range. So all of our "2.0 caps" generate a pDIF value which can be anywhere from 1.6 to 2.4, which is the highest range of a non-crit.

For Crits (which is what SA forces), the "Level adjusted melee ratio" cap is still 2.0. It's just that for crits, the table uses a different min/max for it's pDIF range, and when attack is capped that min/max is between 2.6ish (i'm eyeballing it, you'd have to ask VXZ for the actual #) and 3.0. For our calculations, you can either use 3.0 pDIF anytime the "Level adjusted melee ratio" (which the chart calls atk/def ratio) is capped at 2.0 if you want to determine the maximum value, or use 2.8ish to determine an average value, like we've been doing for non-crits.

It sets up kind of a neat little thingy (for lack of a better word). If your "level adjusted melee ratio" is capped (what people mean when they say "attack is capped", your weakest melee hit will be no weaker than 86.666666% of your strongest hit. That's a pretty narrow range - critting between 86 and 100 onry (if 100 is your highest crit ...you'd be using a low WDMG weapon in this case).

How low you say? Well a D23 weapon in an attack (and STR capped, but let's not go into that) situation will crit between 85-99. So...that low.

edit: Oh, and you owe me 5 gil.

edit#2: Did some random viper bite testing for kicks - target was Sabertooth Tigers in Batallia Downs.

My melee swings hit in a range between 67 and 101. My crits were between 113 and 129. Both of these suggest that my attack (of 358) was capped, as well as my strength. (I actually maybe should've had a low melee hit of 68, it's possible that some of the tigers might have VIT of 33, which would uncap my STR of 72 by enough to hit for 67.) Anyway, I did 6 viper bites with my D32 Magnet knife (not dual wielding). The totals were:

76, 96, 98, 76, 97, 73. All of which are exactly in my melee swing range. If your attack is capped, Viper Bite does no more damage than a regular melee swing.

[Kawfee]

I'm a little confused on what you just said about crits (hey, its 3am, im tired T.T).

Am I to understand then, that if you "cap" your pDIF at 2.0, the formula for a crit will actually generate a value of, on average, 2.8, which is substituted for the 2.0?

Which means in my previous, example, the 893 damage would instead be: 1137.5

Which seems much closer to a true WS number (granted still off by ~200 maybe, but its in a ballpark I'm willing to accept).

That being said, now all we need is Taimat WS numbers <.<

I know I have some screenshots somewhere or other from my WS's, but it would take me hours to find them, and... {No thanks} lol

[archibaldcrane]

I'm a little confused on what you just said about crits (hey, its 3am, im tired T.T).

Am I to understand then, that if you "cap" your pDIF at 2.0, the formula for a crit will actually generate a value of, on average, 2.8, which is substituted for the 2.0?

Basically, but you meant "cap your level adjusted melee ratio at 2.0" - a # which provides you the pDIF range, but is not = pDIF. That 2.0 #, which you enter in the horizontal axis of the chart, will be replaced by a # that is somewhere between 2.6(ish) and 3.0 on the vertical axis of the chart by the game's randomizer. That 2.6-3.0 # is your pDIF (or between 1.6 and 2.4 if it's not a crit - we were just using 2.0 as an average before.

[Kawfee]

Right, I understand its a range, but for simplicity's sake, I'm using a single number to compare averages.
edit: Unless you think a Min, Median, and Max value would change the outcome at the extremes? But I generally don't like to count on extremes as case studies, and don't think that they will prove different anyways >.>

[archibaldcrane]

Right, I understand its a range, but for simplicity's sake, I'm using a single number to compare averages.
edit: Unless you think a Min, Median, and Max value would change the outcome at the extremes? But I generally don't like to count on extremes as case studies, and don't think that they will prove different anyways >.>

If youre always talking about capped attack scenarios, it won't matter. However, if you aren't, your min/median/max values can get wonky. At certain places in the chart, most notably for non-crits in the 0.5 to 0.8 range, your lows keep rising as you approach 0.8 "level adjusted melee ratio" from 0.1 to 0.5 pDIF, but your highs stay right at 1.0 pDIF. Shit's weird, just keep it in mind.

[VZX]

edit#2: Did some random viper bite testing for kicks - target was Sabertooth Tigers in Batallia Downs.

My melee swings hit in a range between 67 and 101. My crits were between 113 and 129. Both of these suggest that my attack (of 358) was capped, as well as my strength. (I actually maybe should've had a low melee hit of 68, it's possible that some of the tigers might have VIT of 33, which would uncap my STR of 72 by enough to hit for 67.) Anyway, I did 6 viper bites with my D32 Magnet knife (not dual wielding). The totals were:

76, 96, 98, 76, 97, 73. All of which are exactly in my melee swing range. If your attack is capped, Viper Bite does no more damage than a regular melee swing.

well, indeed, your fSTR and pDIF are at cap, given those result numbers. (slightly skewed from the real "formula" though, but oh well... SG themselves said their formula still have some minimal errors)

anyway, It's still doesn't show anything. your atk/def is already capped in the first place. If VB really does double your attack, you should show the numbers from fighting higher level enemy. (i.e. your corrected atk/def is about 1.0)

oh and this is an interesting thing you guys can test:
from the range 1.6~2.4, what's the most frequent number that pop in that range. I think it doesn't follow normal distribution graph (Gaussian). I've heard it's rather skewed to the left (lower number sides), so your average may not be sitting on 2.0, but rather (slightly) below it

[Mysterio]

Archibald seems similar to Genome.

[TB]

Vs. Tiamat:

Assumptions:

• You are a SAM75/WAR37, because Sneak Attacking is stupid and Trick Attack has no bearing (unless Chiz chimes in).[/*:m:aef7f]

Why? Get some good tanks. >_>

[Zanz L. Barre The 3rd]

Vs. Tiamat:

Assumptions:

• You are a SAM75/WAR37, because Sneak Attacking is stupid and Trick Attack has no bearing (unless Chiz chimes in).[/*:m:f3e6e]

Why? Get some good tanks. >_>

Think he was joking. No way people are going to waste their time doing 300-500 dmg WS when they can be doing 1-1.6k W