Sunny, you for your DET testing data you posted a few pages back, what were the Determination Values you used?
Printable View
Sunny, you for your DET testing data you posted a few pages back, what were the Determination Values you used?
These are the data points I've got so far.
(I don't trust myself not to mislabel things, though.)
(Potency/100)*(((WD*STR*0.0031499)+(STR*0.1006780))*(DET*0.0007328+1))
Could people try do some min/max parses and test that this formula is working somewhat? Still collecting datapoints, but it's accurate so far.
EDIT: I'll add, it only works for the Dragoon AND MAYBE the Bard. For other jobs, you'll need to divide the total DMG by 0.9-1.2 (Depends on the Job). The main thing is that I want to make sure that the actual format of the formula is working as intended. If it's accurate for a lot of parses for many damage ranges, then I'll work on readjusting it to account for job Coefficients.
EDIT 2: DET has a weighting of around 0.387 with this formula of mine.
Hey guys, do you know if a Longinus build with 38 Det, 51 Crit and 2 Acc is possible? If it is then I am thinking of going for the following build:
http://ffxiv.ariyala.com/Q7YG
However, the Ariyala BIS solver seems to think that this one would be better...
http://ffxiv.ariyala.com/Q7YF
Hmm. I don't like losing Crit for SS - even if I do end up gaining Str. Choices, choices...
you can get 38 det/50 crit/2 acc; not 51 crit though. i'm currently using the 2nd gearset you linked, though i'm adding sksp to get 9 skill bfb whenever i can be assed to grind soldiery
anyway, 2 str and 1 det is like 2.333 str, while 15 crit is worth 3.495~ (the difference between the 1st and 2nd set is -2 str +15 crit, -1 det, -15 sksp). 1.165 str difference before you account for 15 skillspeed, so it's definitely worth it
edit: oh god why is it so hard to find the most recent drg stat weights
dervy pls update reddit thread
They're all on my blog now. CBA with reddit half the time. Too much of a circle jerk there lmao.
You'll need to click the "3 lines" widget thing in the corner to open up my Menu.
Having said that, DET is looking to be valued a lot more than we originally thought... thanks to sunny that is.
oh ok, cool. didn't see those squiggly lines at first
Consistently overestimating values with low WD and high STR and high DET (e.g.).
XXX*(A*DET + B) isn't consistent with my DET findings. The trends would have to converge at dmg = 0 for this to be the form.
I'll add some of your Data points and solve it again. That is odd through, I solved with high STR and low WD values
I was thinking XX +(WD*DRT*A) + (STR*DTR*B) as well, but that just doesn't look streamlined at all. (Most probably is)
I want to start collecting data for DET, but I need more right-side accessories
EDIT: I've done multiple new solves of (WD*STR*A)+(STR*B)*(DTR*C*+D) and (WD*STR*A)+(STR*B)+(WD*DTR*C)+(STR*DTR*D)... In each solve, (WD*STR*0.0036x)+(STR*0.08) (rounded) is very stable. The issue herein lies with Determination, for sure.
EDIT 2: Try this out
(Potency/100)*(((WD*STR*0.0034553)+(STR*0.0870452))+(WD*DTR*0.0017074)+(STR*DTR*0.0000449))
I've tested it with multiple 5-22 WD parses (with high and low DTR/STR ranges), as well as max ilvl/stats HT + FT. 12 datapoints, each one has a variance between 0.87 and -0.56. Only one of the, has a variance of -1.37, but I suspect this is due to extremely heavy in-game rounding.
GUYS HE DID IT
http://puu.sh/gYq3A/3f96f4891a.jpg