The Last Dance: Gearing Paradigms For A New Age

Hmmm I don't get it there, how does what Foldy said there has anything to do with Lux Pugio affecting crit of both hands?
It's missing the line "Changing the Emp +2 pants to Lux Pugio results me the same 425 swings with non-crit occuring" or some sort.

I just recently started gearing my dnc so i could try to bring back more excitement with soloing. I own a coruscanti as well as lux,oynos,phurba (soon), and almost done with str/eva daggers. is getting emp dagger a game-breaker for dnc or could i keep my builds for coruscanti? (could i get input for thf as well?)

He was using the dagger when he did the test. If the two hypotheses are:
1) Lux affects both hands : Both hands should have a 100% crit rate
2) Lux only affects its own hand : Main hand should have a 99% crit rate
Then his ~210 hits with the main-hand lets him be about 90% confident that Lux affects both hands. Given the recent Coruscanti test, it seems pretty likely.

He mentioned 96% crit rate because that would change the null hypothesis to "Main hand has a 96% crit rate" which would be a lot easier to detect and would take many fewer hits to reach higher significance. 86 Crits gets you to 95% confidence with a null hypothesis of 96% Crit rate. It takes 303 to reach the same confidence with a null hypothesis of 99%.

Yes, I meant that's the point I'm still skeptic on using it as the basis of conclusion.

How many swings you actually need to prove that the crit rate is 100% instead of 99.x%? (That is assuming the supposed-crit-rate should be higher than 99% if the crit rate only affecting off hand)

yeah someone answer this question lol >.< Is the aftermath that good for dancers?

He didn't have uneven DA/TA between the hands, so you can just divide his sample size by two and subtract a few to conservatively estimate his main hand N (that is why he said 210).

Try this:
* Make an excel spreadsheet where A1:99 is 0.01:0.01:0.99 (0.01 to 0.99, counting by 0.01s)
* Make B1 "=A1^D$1" and fill it down to B99.
* Make C1 "=B1".
* Make C2 "=B2+C1" and fill it down to C99.
* Put your N in D1.

Column C is probability of getting N (D1) Crits in a row with an alternative hypothesis of the value in A to 0% Crit rate. If you want to put in a minimum crit rate, drop a 0 into column C wherever you want that to be. It does not work for low sample size and may not be very reliable for crit values that are far away from the extremes.

Also, this is based on the assumption that there are no non-integer(/100) critical hit rates. It seems very unlikely that there are.

I may be extremely slow at getting the main portion of this thread in order due to school/travelling, but I cover this pretty clearly on the front page.

you need to make an update for coruscanti since these tests show that the crit rate affects both hands, and your cons mention that it only affects the hand you are wielding it in.

yeah, i just finally got back to the internet and fixing it nao 8V

I still can't digest what Byrth said yet, so I'll get back to that later, but here's another question:

Etoile toe shoes +2:
What does Enhance "Closed position" effect do ?

My guess, extra acc and eva, but that is just a guess.

It's not anything that you can see by hitting a Hpemde thousands of time for 0 damage. (Crit rate, Acc, etc.)

Your kitten example does not illustrate diminishing returns. There has to be a productive process associated with adding kittens, the results of which would be the return, not the kitten itself as a comparative percentage of the total number of kittens.

If you have no Triple Attack or Quadruple Attack, Double Attack can be seen as a direct multiplier of your melee DPS (not including WSs). Someone with 20% DA has 20% more melee DPS than someone with 0% DA. Someone with 40% DA only has 16.6666% (1.4/1.2) more melee DPS than someone with 20% DA. Add the same amount of double attack, receive a lower percentage increase in DPS.

Mathematically: (1+X+Y)/(1+X) < (1+Y)/1
(as long as X isn't negative)

Sylow said:
(1+99+1)/(1+99) < (1+1)/1

So he is actually giving an example of diminishing returns.

The mechanics of DA are nice in that they allow you to discretize the whole system for ease of explanation.

(beaten by byrth!)

Although after "Sylow said" your inequality is reversed!

nuh-uh

Thats not how Diminishing Returns is defined. The comparative percentage increase is not the return. Its common sense if you are adding a fixed number to a larger and larger sum then the relative increase over time is a smaller percentage of the whole. I'm not disputing that, its just not how the rest of the world defines diminishing returns.

The point still gets across. No need to argue semantics.

"The law of diminishing returns (also law of diminishing marginal returns or law of increasing relative cost) states that in all productive processes, adding more of one factor of production, while holding all others constant, will at some point yield lower per-unit returns." - lolWikipedia

http://en.wikipedia.org/wiki/Diminishing_returns

This could be debated for pages, as Krizz pointed out its a question of semantics, what are you calling the return? Is it the comparative value of DoT with each tier of DA, or is it the absolute value of DA? In FFXI terms the only thing people care about is the comparative value, that decreasing percentage. In a real textbook economics setting you would never use the comparative values, diminishing returns would be based on the absolute values of DA, which are constant.

At any rate, I don't want to derail this thread and further, lets just agree to disagree.

For FFXI purposes the return on adding any stat to a damage-dealing class is the % increase to your damage.

For DA, kittens, marbles, watermelons, or any other arbitrary discrete object, work because we can simply normalize our attacks with our average damage per attack, and divide that out leaving us a completely discrete system.

In this manner, adding 1% double attack, adding a single attack to a series of 100 attack rounds, and adding a (kitten, marble, watermelon, rapper, penis, taco, hamburger, car, grasshopper) all mean the same thing, allowing us to explain and work with it very simply.

This isn't Econ 201, it's unnecessary to present a technical and textbook definition of diminishing returns, it is however necessary to present something that someone with an obviously weak mathematical background can understand :)

Anyway, I'm going to work on the main page a little bit. So if you have anything you want me to shoot up there IMMEDIATELY, now is the time to speak up.

Dude idk if ur amazing with any other job than DNC but the guide was very nice to read!

So im finally starting to work toward the bigger things, I have a couple nice atma 2 zone wins down 2 more hopefully tonight. Next week i start cracking on my Af3+2 seals but curious as to what's a good head piece to wear till i get my Charis+2 head cuz i'm SO sick of looking at my Optical Hat..

Zelus, Ocelomeh(+1), Charis+1, or Brego Helm.

Just looked up some stuff the Brego Helm seems nice and 300k is definitely affordable for 5% haste, shite if it weren't for the store TP on the charis i might end up using that

Charis+2 has 6% haste. That's the main reason to stick with Charis.