Strength Vs. Atk Power For DD

Has anyone tested which is better for DD?
vs Ifrit
vs mobs 50+ (beastman camps)
vs <50 mobs

Can anyone clear up what exactly a point of either stat does as far as affecting damage range? (ex. low end/high end of the damage curve)

If anyone legitimately tested this that can post their findings that would be greatly appreciated. Thanks

i would like to know also.

This:
From this.

I will freely admit that I know nothing about FFXIV's core mechanics, but the post you quoted reminds me of the way people looked at STR/VIT and attack/defense in FFXI back in... oh, I dunno, 2004? I'd dig further, see if you can get closer to ~2007.

This is specifically on R50-52 Zahar'ak Drubber and Halberdier.

322 STR 498 Attack LNC with Ifrit's Harpoon (Base Damage: 151)



285 STR 531 ATK LNC with Ifrit's Harpoon (Base Damage: 151)

It's 33 att vs 37 str, since it is alot easier to obtain 33 att than 37 str (looks like fingerless + att4 vs cobalt mitts + str4 x2), I'd think att is bigger bang for the buck in this situation.

Average damage is statistically equivalent given your sample sizes... In fact, the only outlier on that entire chart is the 8% difference in peak Doomspike II damage. You could easily interpret the results in a variety of ways given the lack of known equations or a baseline test... For instance, if you assume that a percentage increase in attack will equate to a roughly equivalent or slightly greater increase in damage output (consistent with FFXI mechanics), you would expect a 6.6-7% increase in damage with the second set over your nonexistent baseline. If we also assume that your STR is compared to the enemy's VIT such that an increase of ~4 STR will increase a melee attack's base damage by 1 (again consistent with FFXI mechanics), 37 STR would give 9~10 base damage. If your "fSTR" was low in the second test, this could give... up to a 6.6% increase in damage over baseline assuming neutral fSTR in the second test. Hmm. Funny how that works.

I would suggest observing damage peaks on low level monsters with varying STR levels (as low as possible, then increasing it incrementally) if you're interested in testing my suspicions since fSTR would likely be easier to establish initially than the inner workings of the attack/defense equations, plus it may give you a value for capped attack/def function to work with. Would happily do it myself but unfortunately my current system can't run FFXIV.

EDIT: Disclaimer: I am not asserting that the equations must be the same, only that they are likely the same given that it was many of the same people within the same company. I realize that I only provided an example that supports my theory, but it was only put forward with the intent of demonstrating plausibility.

Another way to observe potential correlations would be comparing the difference between these sets on a job with a lower base damage weapon, bearing in mind the possibility that STR could conceivably cap lower in this situation.

Bell curve it, amigo. A bell curve will show the results most accurately.

Information like this is how urban legends get started. The point of outliers is a huge one which is why the bell will be accurate.

Funny story: waaaay back in the day, in pre-abby XI, I had a SAM acquaintance who was decked out. He argued that SAM was the best merit party job and I stated that a monk would wearing Bird Banes would actually be better. The result? His spike damage (outliers) were impressive, but the bell showed the truth: MNK dominated over a 30min period.

I'm aware that averages will most effectively demonstrate differences in most situations, but in testing mechanics it's often simpler to compare maxima if you can be assured that all other variables are held constant since you need only one data point for each test rather than an extremely large sample size.

That is only true if maxima is part of mechanism and not caused by error or external factor such as mob being affected by Dia, or Expose from Maim.
On Ifrit Exposed effect would give 25%~ damage bonus.

so what wins..lol

Id go for the att

Theres some testing on STR on this page towards the bottom: http://www.bluegartr.com/threads/107403-Stats-and-how-they-work./page5 BG is pretty spot on with their information but finding anything is generally a giant gaggle *** with their organization, wish they would put XIV info on their wiki.

STR is the superior stat though, especially on higher end mobs like ifrit.

That data about STR being better for high end mobs like Ifrit is outdated.
And it mostly depends on class really. LNC for example would benefit more from STR, but ARC would benefit more from attack power than STR.

Personally I'd go for whatever stat affects the damage output of that class.
Like LNC would be STR/PIE and ARC DEX/PIE etc

You can't lose with ATT. In my experiences with ARC on moogles, 74 ATT averages much better on light shot damage than 40 DEX, even though controlled tests show a 1:2 ratio with DEX vs ATT.

This wasn't a controlled test, but is pretty consistent and it's good enough for me to make a decision.

Considering you can only add ATT to one piece of gear, except for a few weapons, put ATT in your gloves. All other stats can be melded into other armor pieces. You really can't go wrong with 1 great pair of gloves for all of your DD classes.

As far as weapons go, I have yet to see any crafted weapons outparse Ifrit weapons anyways, in person or otherwise. So no comment there.

Could you try parsing +30 crit rate from LL lance and sipahi vs your attack gloves. I would do it myself but I haven't been able to make a nice pair yet.

I don't own either and have no intentions to acquire them, sorry.

Meh, it's alright, I'm working on a pair of gloves now it's only a matter of time till I get lucky.

good post mowie you just converted me into an attack power glove build for my next forbid project.