ok, it's been bugging me for a little while that there hasn't been any set value for SP's mods. 30-40% for both is nice and all, but I'd like to see something more exact. Thus, I busted out the old numbers I got from bsphil (here) way back in the day, and I slapped them together with StudioGobli's fTP (.75), and I'm pretty happy with the results.
----- Math Parts Starting -----
Damage = ( pDIF ( WD ) ) = ( pDIF ( fTP ( BD + fSTR + WSC ) ) )
Where:
Damage = 197
pDif = 3.15
fTP:
D = 47
fSTR = 10
WSC:
Now, I’ll use the .75 that StudioGobli found. They've been nothing but reliable for years, so I’ll trust they’re either right, or very close, and I’ll leave 197 undivided for the time being, due to how ugly the number is.
197 / 3.15 = .75 ( 47 + 10 + WSC )
(197 / 3.15) / .75 = 57 + WSC
( (197 / 3.15) / .75 ) - 57 = WSC
WSC = 26
26 = ( floor( floor( (54(%mod 1) + 60(%mod 2)) * .83) ) )
From here out, it’s a matter of guess and check, since I don’t know of any way to figure it out otherwise. I’ll replace both mods knowing that .3 ≤ %mod1 ≤ .4, .3 ≤ %mod2 ≤ .4
( floor( floor( (54(.3)+ 60(.3) ) * .83) ) ) = 28
( floor( floor( (54(.4)+ 60(.4) ) * .83) ) ) = 37
This tells me that with the fTP = .75, the minimum threshold is not .3. I think that I may have worked the equation incorrectly. I’ll go back and, just for the sake of trying, use a ceiling function.
ceiling( ceiling(197 / 3.15) / .75 ) - 57 = WSC = 27
With this, if I assume a %mod1 = .3 and %mod2 = .3, the equation comes out at 28. My numbers come out at 27 currently, and the end result is an increase of 3 for damage dealt, which is well within any margin of error by my count.
Now to check, I’ll adjust the STR by the numbers mentioned earlier. Now: S = 75. The value of O I already know to be 217 with this number for S. Nothing else is changed.
217 = 3.15 ( .75 ( 47 + (floor( 21 / 9 ) + 8) + WSC ) ) )
217 / 3.15 = .75 ( 47 + 10 + WSC )
(217 / 3.15) / .75 = 57 + WSC
ceiling( ceiling(217 / 3.15) / .75 ) - 57 = WSC
WSC = 35
( floor( floor( (75*.30 + 60*.30) * .83) ) ) = 33
To low. Margin of error upward is fine for a little, but reverse means that my highest achieved number shouldn’t be possible. Now I’ll try reworking it a bit
.
( floor( floor( (54*.35 + 60*.30) * .83) ) ) = 29 : Damage = 203
( floor( floor( (54*.32 + 60*.32) * .83) ) ) = 29 : Damage = 203
( floor( floor( (75*.35 + 60*.30) * .83) ) ) = 35 : Damage = 217
( floor( floor( (75*.32 + 60*.32) * .83) ) ) = 35 : Damage = 217
203 is still within the margin of error for the first sample, since in that sample, there was no clear cut cap like the +21 str sample (where there are very clearly a large number of 217 results) There aren't any ws that have a 35%/30% split. Actually, none of the 35% mods have a secondary modifier. However, there are 3 ws that have a 32%/32% mod split (Judgment, Calamity, and Weapon Break)
So my conclusion:
Stringing Pummel:
fTP: .75
Str 32% Vit 32%
Of course, I'll do some more extensive testing when I get some time, but for the moment, let me know what ya think.
Edited, Oct 24th 2009 1:12pm by Jinte


. Given roughly equivalent gear, SP and SS should average roughly the same for pup (so will non-vereth VS, for the record), with SP having more variation and higher spikes/lower lows, and SS being more consistent. The higher the target's defense, though, the more the situation favors SS. 

