Dexterity's impact on critical hits

[Yugl]

Seems all a bit tl;dr: shit is situational to me for you to keep saying Aurum>Zahaks all the time. With FFXIGear non existant I can't be bothered to show you some more valid examples of Zahaks being better but you are wrong to make the blank statement like you do.

It's actually not that inaccurate of a statement when you consider what I said. In what situations will you not be using Pizza at 75? Colibri and insignificant solos. If you're solo, chances are you're not capping pDIF since you don't have buffs. However, it is possible for you to quickly cap out dDEX while solo. I already addressed Colibri in my last post. Mobs that you would use Pizza will vary, but chances are that it will be way out of reach or so far within reach that Aurum is capable of capping your dDEX. Notice how for Colibri, it was ONLY Elvaan whom I could give Zahaks > Aurum to, and that was only due to the close proximity of the situation. In other words, if Colibri AGI were two less than what it is, then Aurum would out perform Zahaks.

Quite frankly, FFXIGEAR only spit out the simple mathematics of it. You can get the same calculations it does using two sites: http://bellsouthpwp2.net/k/i/kinemat...tatCalculator/ and http://ffxiah.com/. If you want to do that and come to your own conclusions, that's fine.

[Shamaya]

Yes lol you could have Mirax. But we'll be using those mobs to double check for "level correction" (at least I will). Probably not an issue but at this point, probably better safe than sorry. So it wasn't bad that you used the lower mobs, which is where we got our numbers from.

[AoshiZ]

In short, cases where dDEX cap is possible, mobs have shitty DEF, and Aurum instead of Zahaks will result in uncapped DEX, it's possible that Zahaks will outperform Aurum (I only tested one type of mob and one scenario with 500 flat ATT, so I'm not going to blanket statement this). In other cases, you're likely to find Aurum outperform Zahaks.

The main reason for this difference (If this was done correctly since I just quickly fixed the calculator for this) is simply due to the fact that when I compared 15ATT to 2STR and 3% Critical hit rate (i.e. cases where dDEX is capped), it appears that 15ATT wins. However, in cases where it's not capped, it boils down to how much Critical hit % you're getting from that 2DEX. Looking at Masamune's chart and assuming your dDEX is capped with Zahaks and not Aurum (By exactly 2 DEX, this changes if it's only off by 1 DEX) places you with 15 ATT v 2STR and 5% Critical hit rate. However, on mobs with higher AGI, your dDEX is likely to drag down to less than dDEX=40 where differences in DEX yield much less returns (Notice how the change is almost piecewise). This is why I considered DEX to be so important on Colibri for Elvaans where dDEX is likely to be in the 40s while on tougher mobs, I tend to neglect the DEX since you're likely to have dDEX less than the 40s, and consequently, much less returns on DEX investment.

I'm a little confused by your first paragraph (in the quote) versus your second one. Your first paragraph states that gimpy mobs where dDex CAP is possible is where the mobs have shitty defense. Then you go about talking about cases where dDex is capped but PDIF is not or talk about cases where dDex is less than dDEX=40.

The same arguments you make can be countered with your exact arguments if I am not mistaken? If dDEX is capped most likely the 15attack and 2STR from aurum won't help because PDIF will be capped and FSTR will be capped (because the mob is most likely gimpy and is the exact reason with dDex is capped). In cases where you're less than cap it really depends on the mob and buffs doesn't it? If we're talking about pure WS swaps for Critical WS's I would say it would be easy to go either way based on if you have a Cor or Bard in the party and what buffs you are recieving? I would most likely have both gear options available to myself and not do a aurum > zahak or zahak > aurum argument.

[Yugl]

I'm a little confused by your first paragraph (in the quote) versus your second one. Your first paragraph states that gimpy mobs where dDex CAP is possible is where the mobs have shitty defense. Then you go about talking about cases where dDex is capped but PDIF is not or talk about cases where dDex is less than dDEX=40.

The same arguments you make can be countered with your exact arguments if I am not mistaken? If dDEX is capped most likely the 15attack and 2STR from aurum won't help because PDIF will be capped and FSTR will be capped (because the mob is most likely gimpy and is the exact reason with dDex is capped). In cases where you're less than cap it really depends on the mob and buffs doesn't it? If we're talking about pure WS swaps for Critical WS's I would say it would be easy to go either way based on if you have a Cor or Bard in the party and what buffs you are recieving? I would most likely have both gear options available to myself and not do a aurum > zahak or zahak > aurum argument.

Gimpy mobs is a terrible term to use, but I used it for due to time constraints. I'm basically putting on the argument that in solo situations where you're not tackling something crazy (And thus not using food), there's a greater chance of you capping out dDEX than you would pDIF. My argument for this lies in the fact that Elvaan (The worst DEX of the races) fighting Colibris (Which isn't necessarily one of those "Gimpy" mobs I'm referring to even though DRG can solo them) will only find himself 2DEX short of the dDEX cap using Aurum Body (Which is why I pointed out earlier that it is a VERY particular case) in a very solid Drakesbane build (Will post the set up in mind soon and even critic it). ATT on the other hand caps on Colibri at around 600-700 ATT if I remember correctly (Maybe even higher?). While it may not be that high soloing, I'm sure you're not going to cap it out at 475 ATT, which is VERY generous estimate for ATT while soloing without food. On the other hand, DEX cap is something reachable by some races, and closely reached by Elvaan. Due to these circumstances, I'm convinced that it's easier to cap dDEX than it is to cap ATT.

Note that the goal of capping ATT is so that the critical hit from Zahaks outweighs that 15ATT. This means you cannot use gear to achieve this since you'll end up tossing loads of STR/DEX in the process, and since we're considering solo situations, gear and food are the only common means of raising your ATT. The former is taken by STR/DEX demands for WS. The latter I explicitly omitted for in my general argument, and quite frankly, doubt many use while soloing.

As for the technical issue here, I'm convinced that if your dDEX isn't [39-48] (Possibly 49, but I'm too lazy to set up the excel again) and pDIF isn't capped, then Aurum will triumph over Zahaks. The general idea is that between 39 and 48, you're gaining roughly 1% Critical hit rate per dDEX, which in practice comes out to 5% Critical hit and 2STR v 15ATT. From what I've looked at using this WS Set on Elvaan, and assuming a 500ATT base, the 5% Critical hit and 2STR will outperform the 15ATT. However, in situations where (dDEX<39) or (dDEX>48), it seems that the 15ATT will outperform 3/4% Critical hit and 2 STR. To be more precise, that 4% DEX comes in when you're 1dDEX from cap or hitting a new between <dDEX<39>. The latter is a onerous situation to even consider since you would need precise details on each and every mob.

What this comes down to, as you can see, is that you need very specific information about the mob you're fighting in order to make such a judgment call. That's why I've been using broad statements that relay my opinion of what types of mobs would go within each class. However, I'm unfamiliar with Studiogobli, so until someone give exact stat values for each and every mob encountered in Merits and EG, I'm standing by what I've said. Moreover, there is also the issue of how much initial Critical hit rate Drakesbane receives. I'm unsure if that will alter the comparison (Someone who has given it more thought or is a statistics person can enlighten you on this), but it's something to consider. At this point, it's clear that more information is necessary to settle the indefinitely (Since new mobs will appear), but from what we have, the general statements made earlier seem to be a reasonable argument. The problem now is that people are quick to demand that people go test stuff and demand new information, but seldom do they think to obtain such information themselves.

Edit: A very strong critic of the build I placed is that it does not utilize WS Gorgets nor does it have capped ACC on Pizzaless mobs (I think?), which places more support for Zahaks in Colibri situations. Keep in mind, however, that this build is directed towards Elvaan. Other races will cap out dDEX much earlier.

[Masamune]

Moreover, there is also the issue of how much initial Critical hit rate Drakesbane receives.

Not only for Drakesbane, but also crit based WSs... Personnally, i'm on RagingRush but i can't do alone. Same should go for other WSs.

The problem now is that people are quick to demand that people go test stuff and demand new information, but seldom do they think to obtain such information themselves.

so true...

So far, i'm still on checking mobs stats to fill in the Mobs Families stats that JPs left unfinished, but should have a good portion of it filled/verified in next weeks (you can check mobs i already /checked i posted their stats calculations in their own wiki discussion pages).

Next, i plan to verify with Motenten(if he ok) the relation between [time fought] and [# hits] displayed by Kparser to continue further with DA/TA/Zanshin combinations tests, using method used here.

Without those infos (+ crit% for WSs), i doubt highly any math poeple can even answer Starr accurately. I mean in which situations (mob and buffs) ArmorA will be better than ArmorB, when we we don't know not even half the accurate conditions of said situations.

[Mirax]

If you can design a test for me I'd be totally willing to put in some hard work to find the new pdif values and damage equation. I've been looking since the 2 hander update. Since the pdif varies at cap now still, even with crit, I have no idea how id go about doing this.

[Masamune]

I'm interested too into pDIF testing, but :
- kinda offtopic here, and the method used here is based on doing 0dmg so not applicable,
- I've tested pDIF caps for 1H & 2H weapons on rabbits ronfaure (check wiki discussion page for pDIF)
- but since pDIF caps tests have been done under capped fSTR, testing under non-capped conditions would have to account this 2ndary parameter, in parallel with your attack.

My conclusion is that, for the moment, i see pDIF testing like a 3-D function with a format like DMG = f(fSTR;cRatio), which makes testing lot more tedious than crit% test done here. That means would have to build a 2-D curve DMG=g(cRatio) @ each fSTR values.
The methodology itself i think is kinda simple:
-just smack a mob from which you know exactly its VIT and Defense (that's why i'm checking mobs stats atm) to get your maximum and minimum normal/crit dmg values.
-then make divisions :

pDIFmax= DMGmax/BaseDMG
pDIFmin= DMGmin/BaseDMG
CRITpdDIFmax = CritDMGmax/BaseDMG
CRITpdDIFmin = CritDMGmin/BaseDMG
...to obtain finally a range for each pDIF normal & crit, for the weapon you used during test.

- repeat test with several weapon ranks, until you end up with a single value just by overlapping the different ranges obtained with each weapon.

You can also ask Shamaya who kindly directed me @ some japanese blog (same blogger TTaru) to a pDIF testing test, but since i didnot understood at all like 80% of their methodology, i'm not convinced about the result(check curve green and red at bottom) they obtained, at all.
On top of that, their pDIF caps don't seem to match the values i got myself (unless using a cRatio capped @ 2.5 maybe?)

Hope that help. but i'll stop here it's offtopic anyways.

[JnnnnnL]

I know it's a pretty old thread but I'm wondering something, ok it's cool this thread shows you can get 24% crit hit rate on birds pretty easily (especially if you are a mithra) but I'm wondering if you can get 50 dDEX in dyna/ein ect for example or if there is any data about other mobs' AGI or if someone (mithra) ever tried to get 24% crit on another mob than Gcolibri ?

[Masamune]

Honestly, since i didnot test barely 10% of different mob families' stats in that game, i doubt someone can pull you exact agi numbers out their ass without tests, unless specified mob have been already tested.

The mobs i already tested have their calculated stats posted on their respective wiki discussion page (when not on main page).

But as a general rule, mobs above lvl75 have an agi ~70 if not more (very roughly), so you can guess how much DEX you would need for mobs at the events you mentioned.

[pchan]

Considering this agi range, you can expect dex build to work on mobs lower than even match only, which excludes most of endgame events. I know for a fact that it won't work in salvage for instance. I don't have a specific dex build, but my gear gives me 12% crit on colibris and 9% only inside salvage (low sample size etc), which is floor. So if you check the chart, you need 15 DEX or more to have the same crit% in salvage as in colibri camp, which imo is unrealistic, unless for WS damage maybe.

[JnnnnnL]

well basically I asked that question because I usually get 12,5%~ on birds with 101dex and the other day with 101dex I got 11,70% in dyna-windy (i'm elvaan) so I was wondering if for a mithra it was possible to break the 50 dDEX on some event mobs, also if it's possible to do it on only certain mobs you could use a DEX macro

[Yugl]

well basically I asked that question because I usually get 12,5%~ on birds with 101dex and the other day with 101dex I got 11,70% in dyna-windy (i'm elvaan) so I was wondering if for a mithra it was possible to break the 50 dDEX on some event mobs, also if it's possible to do it on only certain mobs you could use a DEX macro

It's difficult to calculate the dDEX with those values for two reasons:

1. Sample size probably isn't enough to ensure those figures are accurate.
2. Even if they were accurate, the piece-wise function (I think that's what it's called) makes it difficult to differentiate between dDEX values prior to 39.

[Shamaya]

I'm gonna say you likely can. And if you wanna go that route, just keep your parser up and you'll be able to get some verification. Might as well try if it's something you're interested in. I believe back in the day, Aurik was trying to find out dynamis mob stats (or so I heard; maybe it was just a long-running joke that wooshed me), but he has long since left the building.

[Yugl]

If he is going to do it, he's going to need to keep his dDEX values high enough to ensure that it is greater than dDEX=39 since it will be difficult to know the precise value otherwise. For example, the values he posted could be anywhere from dDEX=14-20 and dDEX=20-30.