edit: Basic pDIF and fSTR

[Reece]

I'll keep this pretty brief leaving room for discussion...

Due to the overwhelming number of people stacking Heca cap, alkys, etc on weapon skill this is what I have followed. Has anyone done any tests to prove that due to the high caps already set, that just trying to meet the caps with attack is better than stacking STR? I've always been under the impression that STR just raised caps and added a small amount of raw attack, and as attack was added it gradually rose until cap.

Which one out weighs the other when is come down to weapon skill? Attack build? STR build? Mixture of the both? (btw not throwing Acc out the door here that should be on every multi-hit WS build, just noting what should go in extra slots)

[Delekii]

What caps are you talking about? You used the word cap 5 times and I'm not sure what cap you are talking about (except hecatomb cap )

The whole "str to make the cup bigger, attack to fill the cup" is really a bad analogy, it doesn't work like that. STR raises your base damage up until fSTR cap for your weapon, as well as changing WSC if it has a str portion. Attack changes your pDIF until the attack cap is met (2.4 non crit, 3.0 crit, 3.75 ranged crit). Base damage * pDIF = Damage that you see.

Whether or not you will get more benefit from str or attack depends on how much of each, how much of each you already have, what you are fighting, what debuffs it has, what weaponskill you are using at what it's modifiers are, so on so fourth.

[Reece]

I'm sorry, I was VERY vague.

First off:

fSTR is effected by the monster's VIT?
pDIF is effected by the monsters DEF?

2nd:

Are their tiers when it comes to fSTR and pDIF or is it all linear?

3rd:
(#1) if so...
Do monsters VIT and DEF greatly vary from each other. Like do HNM have higher VIT when merit monsters have higher DEF, or is VIT/DEF generally relative to each other in all monsters.

4th:

(This is where my logic might fail again) in DMG equations does fSTR out weight pDIF, or is it just relative to the monsters VIT/DEF?

[Delekii]

I'm sorry, I was VERY vague.

First off:

fSTR is effected by the monster's VIT?
pDIF is effected by the monsters DEF?

Yes and yes

2nd:

Are their tiers when it comes to fSTR and pDIF or is it all linear?

They are both roughly speaking linear, but it depends a little how you define it. STR only increases damage if it increases the WSC or fSTR by 1 which 1 str will not always do. pDIF obviously has a range (since otherwise the damage would always be the same). http://wiki.ffxiclopedia.org/images/thu ... x-PDIF.gif is a graph for pDIF vs attack/defense ratio.

3rd:
(#1) if so...
Do monsters VIT and DEF greatly vary from each other. Like do HNM have higher VIT when merit monsters have higher DEF, or is VIT/DEF generally relative to each other in all monsters.

Oddly worded, and kind of hard to answer.. mob stats are generally determined by their job.. obviously some mobs have much stronger defensive stats overall (eg crabs) while others have very little (eg colibri). If you are asking whether it is common for a mob to have high vit and low def or visa versa? Couldn't tell you. I dont think so.

4th:
(This is where my logic might fail again) in DMG equations does fSTR out weight pDIF, or is it just relative to the monsters VIT/DEF?

Weaponskill damage is calculated from
(WD+fSTR+WSC)*fTP*pDIF
(Weapon Damage + fSTR + Stat modifer)*fTP*pDIF. Its impossible to say 1fSTR=1pDIF or anything like that, it really depends on every other variable.. (edit for ftp )

[aurik]

They are both roughly speaking linear, but it depends a little how you define it. STR only increases damage if it increases the WSC or fSTR by 1 which 1 str will not always do. pDIF obviously has a range (since otherwise the damage would always be the same). http://wiki.ffxiclopedia.org/images/thu ... x-PDIF.gif is a graph for the bounds of pDIF at certain attack/defense ratios.

Just to clarify. The graph shows the min and max and doesn't really go into the frequency distribution of actual pDIF selection between those bounds.

[ringthree]

They are both roughly speaking linear, but it depends a little how you define it. STR only increases damage if it increases the WSC or fSTR by 1 which 1 str will not always do. pDIF obviously has a range (since otherwise the damage would always be the same). http://wiki.ffxiclopedia.org/images/thu ... x-PDIF.gif is a graph for the bounds of pDIF at certain attack/defense ratios.

Just to clarify. The graph shows the min and max and doesn't really go into the frequency distribution of actual pDIF selection between those bounds.

God damn it, its threads like these that make me wish I paid attention in Stats and Econ.

[Bagel]

To start off with, I am not a mathemagician. I will however attempt to answer what I think your basic question is. I think what you're trying to say is something like, "If everyone is so "lolstr rings for tpz, how come STR seems so favored for WS?" So like someone already mentioned, the (physical) WS dmg formula is this:

(( D + fSTR + WSC) * fTP) * PDIF

D is your weapon's damage number thingy, and fTP is dependant on what WS you use and how much TP, neither of which are effected by the rest of your gear. That leaves us with fSTR, WSC and pDIF. Assuming you're using a WS with a STR modifier, which many many physical WSs have, raising your STR will have a direct effect on all three of these variables. Equipping attack gear in a slot instead of STR will probably raise your pdif more than STR, but ignores the other two. This does not make it useless of course, as some WSs (notably Asuran Fists) have relatively low STR mods, or none at all (non-relic THF WSs for example).

[Kilhart]

I was also interested in this, but in terms of pDIF vs. fSTR2. IIRC, pDIF for ranged attacks caps at 3.0, which is pretty much impossible to reach in merit for most VT-IT mobs, no? So why do RNGs still stack STR gear for Slug Shot, then? Reason for this is because I'm hearing that ratt does more for Slug Shot than STR at any value of PDIF above 1.1, and I'm wondering if this is true.

[theo]

An even mix of STR and Attack will be best in many cases, because boosting the weaker side of the equation will almost always yield the best results. Think 8 x 4 = 32 compared to 6 x 6 = 36. I know the formula isn't nearly that simple, but it illustrates the point fairly well I hope.

But then, Attack is easy to gain with food, meaning you probably don't need as much of it in your WS gear.

[Kilhart]

That's what I'm thinking, but someone gave examples of both when pDIF was close to 1.1 and an extreme value of 1.5 or so, and theorized that ratt would outdo STR in all cases where pDIF > 1.1. Of course, this was using fSTR2 approximations, but still. I'm still looking for that bit of theoretical evidence that matches up with my empirical data.

[Koyangi]

How exactly is a weapon's damage better from 39 than 31 when the Floor formula makes both of them a 30, where is the distinguishing factor that makes 39 a 39 and 31 a 31. Or better question, why does the formula floor the damage in the 1st place, it seems obvious a 39D weapon does more damage than a 31D.

[archibaldcrane]

How exactly is a weapon's damage better from 39 than 31 when the Floor formula makes both of them a 30, where is the distinguishing factor that makes 39 a 39 and 31 a 31. Or better question, why does the formula floor the damage in the 1st place, it seems obvious a 39D weapon does more damage than a 31D.

wut

[Delekii]

How exactly is a weapon's damage better from 39 than 31 when the Floor formula makes both of them a 30, where is the distinguishing factor that makes 39 a 39 and 31 a 31. Or better question, why does the formula floor the damage in the 1st place, it seems obvious a 39D weapon does more damage than a 31D.

???

Both those weapons have the same fSTR cap, but not the same weapon damage... sorry but thats kind of a no shit question..

(WD+fSTR+WSC)*ftp*pDIF=Damage

edit: 39 damage and 31 damage don't even have the same fstr cap..
floor(39/9)+8=12
floor(31/9)+8=11

[Kilhart]

I'm still puzzled as to the solution for the claim that for all values of PDIF > 1.1, ratt trumps STR (all other values held equal). Can anyone help?

[Amele]

I'm still puzzled as to the solution for the claim that for all values of PDIF > 1.1, ratt trumps STR (all other values held equal). Can anyone help?

you're looking for a proof?

I think the answer is 'no'. but that's without doing the math in exhaustive detail. If someone hasn't done it for you by the time I'm done, I'll edit it in here.

here's the edit I promised:

assumption: we'll compare a bellona's ring to a flame ring for weaponskill.

75 Hume M. ranger. relevant merits: 8/8 marksmanship

relevant gear (excluding accuracy): <- this is not an optimal setup! just 'a' setup.

franny/kreigs/hellfire+1/silver bullet
wyvern helm
qiqirin collar
drone / genin earring
osode
s.kotes
amemet+1
r.k.belt+2
Marine M. boots

with base stats of a 75 rng/nin and assuming 1 minuet (to put us at about 1.27 cRatio) vs. a level 81 Greater Colibri (vit 67, def 322, 1.25 piercing bonus)

we have total stats of: 82 Str, 104 Agi, 457 Attack. (no food)
this gives us a cRatio of 1.269, a pdif Max of 1.269 and a pdif Min of 1.177

for expected maximum damage of 1292, and a minimum damage of 1199 on slugshot (agi30%,fTP 5.0)

now, add a flame ring: +5 str (+2.5 ratt) = max 1322 minimum 1228

swap it for a bellona's: +8 ratt = max 1318 minimum 1226

there. in the area of ~1.2 PDIF and fstr2 of ~9. adding 5 strength improves your expected slugshot (max and min) by more than adding 8 ratt.

there's various inflection points where improving pdif will do more than improving fstr2, and so on; but I'm not going to go to the trouble of finding them all for you. (specific questions are of course, fine.)

[Talisien]

http://genomeffxi.livejournal.com/

The first page is a great post on this exact topic.

[Kilhart]

Hrm. I've also heard from some RNGs that there is more to tiering in ratt than simply the "below 0.9, between 0.9 and 1.1, and above 1.1" tiers, most notably at high levels of ratt. Is there any basis to this other than anectodal evidence?

[Reece]

http://genomeffxi.livejournal.com/

The first page is a great post on this exact topic.

awesome, exactly what I was looking for ^^

[Kilhart]

I'd just like to reaffirm a certain scenario, please.

On Greater Colibri, who have 67VIT and 322DEF (I think? Unconfirmed), as a hume 75COR/RNG with 60STR and 75AGI (266 marksmanship skill), which combination is better of the following two situations?

1) +0STR, Double minuet (~125ratt?), Coeurl sub (75ratt)

2) +35STR, Double minuet

Thank you.

[Reece]

I'd just like to reaffirm a certain scenario, please.

On Greater Colibri, who have 67VIT and 322DEF (I think? Unconfirmed), as a hume 75COR/RNG with 60STR and 75AGI (266 marksmanship skill), which combination is better of the following two situations?

1) +0STR, Double minuet (~125ratt?), Coeurl sub (75ratt)

2) +35STR, Double minuet

Thank you.

I've been playing with this DMG cal go ahead and try it... I don't know if it's completely correct but it's a lot better than Human+Paper+Pencil

http://www.killingifrit.com/download...ulator_2.5.zip