Forum Settings
       
This Forum is Read Only

Melee Stat TestingFollow

#1 Oct 31 2010 at 12:12 AM Rating: Decent
So, over the last day or so, I've done some testing of the melee stats (STR/VIT/DEX). Hoping to find some information, I decided to test some of the stats for myself.

About the tests:
  • This test only dealt with offensive characteristics - no defensive traits were tested.
  • This test only dealt with front-facing hits - left, right, and rear hits were disgarded.
  • Each test in this set was done on the SAME mob. This was done so there would be no variance in mob level, defense, or other stats that may change in a creature.
  • Three tests were done - one for VIT, one for DEX, one for STR. A different mob was used for each, largely due to needing to log off, or people stealing my mob in between resets (some tests had to be performed several times -_-)


Quick Summary, before the details:
  • VIT appears to have no effect on how you fight, offensively. I tested this first so I could increase my vitality for the rest of the tests (survivability). All changes in the VIT test were well within margin of error.
  • DEX has a direct, but diminishing, increase on your accuracy. It decreases both the amount of misses and the amount of evades that occur. It does not seem to affect critical hit rate in a notable way.
  • STR has a prominent effect on increasing your damage. Damage appears to approach a limit, assumably based on level difference. Critical damage reaches this level difference first.


I performed the test by going into the cave in La Noscea at (15,22). I would select a mole, hit it a few times to ensure the level difference was not too great, and then begin recording data. When it approached death, I would disengage and allow it to fully heal, and then repeat until I had approximately 100 hits for each mob.

100 hits per test is not a large sample size. However, the data had to be recorded by hand, since the current parsers do not have any methods to allow differentiation between segments of hitting the same mob. I may look into writing a parser that would allow this.

Some tests were cut slightly short of 100 hits - mostly due to either (1) an accidental crit killing the mob, (2) someone stealing the mob during a reset, or (3) my cat assaulting my modem and disconnecting me (<.<). Any significantly small samples were disregarded and the tests redone.

Most of the data seems very sound. The increases in accuracy from DEX seem appropriate, but seem to diminish as DEX gets higher. I may have hit a maximum accuracy for the level difference between myself and the mob. The same goes for the increased damage from STR.

Interestingly, my critical damage as STR increased appeared to cap long before my regular damage. Also, DEX appeared to increase the range of damage that my hits could do, while only slightly shifting that damage up along the average. It may be interesting to graph out a larger sample size.

The major outliers in these test are Critical Hit Ratio. There seems to be absolutely no correlation between my stats and how often this happened - varying anywhere from 4% to 23%. It seems to be almost strictly dependent on level difference.

Anecdotally, critical hits seemed more common on rear hits. However, I excluded non-front hits from the data, so this is not a tested conclusion.

I have posted the data online, so anyone else that wants to sift through it or run some sort of analysis on it can. In the data, I've included a sheet that lists my gear and other relevant information to the tests.

Hopefully this will help a bit in these early stages of figuring out how the game works.

Data:
FFXIV - Melee Stat Analysis Data (XLS)

Quick Numbers:
DEX25 -> DEX40: +2.6% Acc
DEX40 -> DEX60: +3.6% Acc

STR20 -> STR40: +26% Average Damage
STR40 -> STR60: +18% Average Damage
____________________________
http://ffxivfood.com/ - FFXIV Food - User submitted food effects
Submit your known effects today!
#2 Oct 31 2010 at 6:06 PM Rating: Decent
16 posts
Accuracy could be capped, because it is pretty high even with low DEX. You also got the same hit rat. with different DEX but changed VIT.
This also is my personal impression. your acc seems to be much better with lower lvl mobs.
Also the leve difficulty seems to have a huge impact. Fighting strong enemys seems to lower accuracy drastilcaly.
It seems that the rank difference itself influences taken damage and accuracy a lot.
If I fight enemys with higher rank (Orange/Red) i notice a huge difference between them. A Rank 23 Crap for example hits me everytime for about 250 Damage. A Rank 21-22 Crap for about 110 and i dont think that a Rank 23 Crap has 200 STR more than a Rank 21 one. Also enemys which killed me easily at Rank 20 are with Rank 22 pretty easy to defeat.

About Damage.
If I remember correctly STR boosted the damage range in FFXI.
This could be a reason why your STR increased the damage pretty much.
But it could also be avarage damage boost.
____________________________

#3 Oct 31 2010 at 6:28 PM Rating: Good
***
1,636 posts
you're methodology seems very sound, but as you said, a 100 swing sample size is too small, and until you get it higher, conclusions shouldn't be drawn.

I
____________________________


This forum is read only
This Forum is Read Only!
Recent Visitors: 14 All times are in CST
Anonymous Guests (14)