Item Search
     
BG-Wiki Search
+ Reply to Thread
Page 1 of 2 1 2 LastLast
Results 1 to 20 of 34
  1. #1
    Custom Title
    Join Date
    Nov 2008
    Posts
    1,066
    BG Level
    6
    FFXI Server
    Diabolos

    Level 99 OA2-4 Distribution

    Went and beat on Mantlets and Fortifications for a whole battle. Not a lot of data, but I can start some guessing. Kicker is to not use Last Resort or gearswaps or anything for delay to be consistent and not screw up KParser's counting.

    Raw data:
    Spoiler: show
    Code:
    Melee Data
    
    Player               # Melee Attacks    # Melee Rounds    Attacks/Round    # Extra Attacks
    Raelia                           501               242                1                259
    
    Player               # +1 Rounds   # +2 Rounds   # +3 Rounds   # +4 Rounds    # >+4 Rounds
    Raelia                        95            40            28             0               0
    
    Player               # MultiAttack Rounds    MultiAttack %     Kills w/Min Attacks    Kills w/<Min Attacks
    Raelia                                163          67.36 %                       0                       0
    
    
    Treat As:
    
    Multi-attacks per attack (2x/3x):
    
    Player               # Double Attacks    DA Rate    Perc. DA     # Triple Attacks    TA Rate    Perc. TA
    Raelia                             95    39.26 %     70.37 %                   40    16.53 %     29.63 %
    
    Multi-attacks per round (Kicks):
    
    Player               Footwork?    # Rounds w/Kicks    Kick Attack Rate
    Raelia                     Yes                  95             54.60 %
    
    Multi-attacks per attack (Zanshin):
    
    Player               # Missed First Attacks    # DA w/Missed First   Possible Zanshin %
    Raelia                                   11                      4              36.36 %
    
    Player               Acc. Rate of First Attacks    Acc. Rate of Second Attacks
    Raelia                                  95.45 %                        75.00 %
    
    
    
    
    
    Uncorrected Details
    
    
    Raelia
      Buckets (Attack intervals at 0.5 sec resolution):
           0      11     179      69       0       1       3      16      11      20
          23      35      57      39      31       2       0       0       0       0
           0       0       0       0       0       0       0       0       0       0
           0       0       1       0       0       0       0       0       1       0
           0       0       0       0       0       0       0       0       0       0
    
    
      12/24/2011 12:20:28 AM -- 2 Attacks    122    243c  
      12/24/2011 12:20:34 AM -- 2 Attacks    126      m   
      12/24/2011 12:20:41 AM -- 1 Attack     134   
      12/24/2011 12:20:48 AM -- 1 Attack     171   
      12/24/2011 12:20:55 AM -- 1 Attack     138   
      12/24/2011 12:21:02 AM -- 1 Attack     223c  
      12/24/2011 12:21:08 AM -- 1 Attack     217c  
      12/24/2011 12:21:16 AM -- 1 Attack     150   
      12/24/2011 12:21:22 AM -- 1 Attack     161   
      12/24/2011 12:21:29 AM -- 2 Attacks    145    223c  
      12/24/2011 12:21:36 AM -- 2 Attacks    163    137   
      12/24/2011 12:21:42 AM -- 1 Attack       m   
      12/24/2011 12:21:49 AM -- 1 Attack     119   
      12/24/2011 12:21:56 AM -- 4 Attacks    210c   211c   126    107   
      12/24/2011 12:22:03 AM -- 2 Attacks    142    218c  
      12/24/2011 12:22:10 AM -- 2 Attacks    137    138   
      12/24/2011 12:22:17 AM -- 1 Attack     168   
      12/24/2011 12:22:24 AM -- 1 Attack     109   
      12/24/2011 12:22:31 AM -- 2 Attacks    214c     m   
      12/24/2011 12:22:37 AM -- 3 Attacks    161    119    118   
      12/24/2011 12:22:44 AM -- 4 Attacks    139    250c   107    253c  
      12/24/2011 12:22:51 AM -- 1 Attack     170   
      12/24/2011 12:22:58 AM -- 1 Attack     143   
      12/24/2011 12:23:04 AM -- 2 Attacks    152    160   
      12/24/2011 12:23:12 AM -- 2 Attacks    153    147   
      12/24/2011 12:23:19 AM -- 1 Attack     114   
      12/24/2011 12:23:25 AM -- 2 Attacks    172    125   
      12/24/2011 12:23:32 AM -- 3 Attacks    158    148    152   
      12/24/2011 12:23:39 AM -- 3 Attacks    137    168    153   
      12/24/2011 12:23:46 AM -- 2 Attacks    153    152   
      12/24/2011 12:23:53 AM -- 2 Attacks    154    131   
      12/24/2011 12:23:59 AM -- 1 Attack     122   
      12/24/2011 12:24:06 AM -- 3 Attacks    116    104    108   
      12/24/2011 12:24:13 AM -- 1 Attack     133   
      12/24/2011 12:24:19 AM -- 3 Attacks    246c   117    111   
      12/24/2011 12:24:27 AM -- 2 Attacks    130    112   
      12/24/2011 12:24:34 AM -- 3 Attacks    157    110    227c  
      12/24/2011 12:24:40 AM -- 1 Attack     233c  
      12/24/2011 12:24:47 AM -- 4 Attacks    257c   248c   164    121   
      12/24/2011 12:24:54 AM -- 2 Attacks    255c   153   
      12/24/2011 12:25:01 AM -- 2 Attacks    150    132   
      12/24/2011 12:25:08 AM -- 2 Attacks    112    131   
      12/24/2011 12:25:14 AM -- 4 Attacks    254c   142    254c   127   
      12/24/2011 12:25:21 AM -- 1 Attack     251c  
      12/24/2011 12:25:28 AM -- 2 Attacks      m      m   
      12/24/2011 12:25:35 AM -- 2 Attacks    124    147   
      12/24/2011 12:25:41 AM -- 2 Attacks    255c   152   
      12/24/2011 12:25:48 AM -- 2 Attacks    126    109   
      12/24/2011 12:25:55 AM -- 2 Attacks    102    122   
      12/24/2011 12:26:02 AM -- 1 Attack     149   
      12/24/2011 12:26:09 AM -- 3 Attacks    163      m    241c  
      12/24/2011 12:26:16 AM -- 4 Attacks    168    252c   160    139   
      12/24/2011 12:26:23 AM -- 2 Attacks    108    167   
      12/24/2011 12:26:29 AM -- 2 Attacks    151    138   
      12/24/2011 12:26:37 AM -- 2 Attacks      m    161   
      12/24/2011 12:26:44 AM -- 1 Attack     153   
      12/24/2011 12:26:50 AM -- 2 Attacks    167    232c  
      12/24/2011 12:26:57 AM -- 4 Attacks      m    158    159    236c  
      12/24/2011 12:27:04 AM -- 4 Attacks      m    121    121      m   
      12/24/2011 12:27:11 AM -- 2 Attacks    154    113   
      12/24/2011 12:27:18 AM -- 1 Attack     251c  
      12/24/2011 12:27:24 AM -- 1 Attack     132   
      12/24/2011 12:27:31 AM -- 1 Attack     124   
      12/24/2011 12:27:38 AM -- 4 Attacks    148    161    123    146   
      12/24/2011 12:27:45 AM -- 1 Attack     157   
      12/24/2011 12:27:52 AM -- 2 Attacks    157    122   
      12/24/2011 12:27:59 AM -- 2 Attacks    257c   128   
      12/24/2011 12:28:05 AM -- 2 Attacks    131    130   
      12/24/2011 12:28:12 AM -- 1 Attack     167   
      12/24/2011 12:28:19 AM -- 3 Attacks      m    227c   114   
      12/24/2011 12:28:26 AM -- 2 Attacks    248c   160   
      12/24/2011 12:28:32 AM -- 2 Attacks    218c   121   
      12/24/2011 12:28:39 AM -- 3 Attacks    110    168    238c  
      12/24/2011 12:29:00 AM -- 2 Attacks    132    127   
      12/24/2011 12:29:06 AM -- 2 Attacks    147    135   
      12/24/2011 12:29:13 AM -- 2 Attacks    139    225c  
      12/24/2011 12:29:20 AM -- 2 Attacks    159    163   
      12/24/2011 12:29:27 AM -- 3 Attacks    132    249c   168   
      12/24/2011 12:29:34 AM -- 2 Attacks    144    169   
      12/24/2011 12:29:41 AM -- 4 Attacks    156    117    204c   104   
      12/24/2011 12:29:48 AM -- 2 Attacks    140    220c  
      12/24/2011 12:29:54 AM -- 2 Attacks    147    227c  
      12/24/2011 12:30:02 AM -- 4 Attacks    210c   146    257c   227c  
      12/24/2011 12:30:09 AM -- 2 Attacks    146    133   
      12/24/2011 12:30:15 AM -- 1 Attack     236c  
      12/24/2011 12:30:22 AM -- 1 Attack     154   
      12/24/2011 12:30:29 AM -- 3 Attacks    160    119    152   
      12/24/2011 12:30:36 AM -- 1 Attack     155   
      12/24/2011 12:30:42 AM -- 1 Attack     166   
      12/24/2011 12:30:49 AM -- 2 Attacks    111    108   
      12/24/2011 12:30:56 AM -- 2 Attacks    237c   253c  
      12/24/2011 12:31:03 AM -- 3 Attacks    164    249c   154   
      12/24/2011 12:31:10 AM -- 1 Attack     161   
      12/24/2011 12:31:16 AM -- 4 Attacks    155    236c   146    120   
      12/24/2011 12:31:23 AM -- 3 Attacks    111    146    140   
      12/24/2011 12:31:31 AM -- 1 Attack     255c  
      12/24/2011 12:31:37 AM -- 3 Attacks    105    122    237c  
      12/24/2011 12:31:44 AM -- 2 Attacks    108    156   
      12/24/2011 12:31:51 AM -- 2 Attacks    107    109   
      12/24/2011 12:31:58 AM -- 3 Attacks    130    109    219c  
      12/24/2011 12:32:05 AM -- 1 Attack     200c  
      12/24/2011 12:32:11 AM -- 4 Attacks    154    123    249c   133   
      12/24/2011 12:32:18 AM -- 2 Attacks    158    114   
      12/24/2011 12:32:25 AM -- 2 Attacks    158    113   
      12/24/2011 12:32:31 AM -- 2 Attacks    145    102   
      12/24/2011 12:32:39 AM -- 1 Attack     119   
      12/24/2011 12:32:45 AM -- 2 Attacks    134    145   
      12/24/2011 12:32:52 AM -- 2 Attacks      m    152   
      12/24/2011 12:32:59 AM -- 3 Attacks    134    201c   250c  
      12/24/2011 12:33:06 AM -- 2 Attacks    248c   148   
      12/24/2011 12:33:13 AM -- 2 Attacks    138    117   
      12/24/2011 12:33:20 AM -- 1 Attack     144   
      12/24/2011 12:33:27 AM -- 2 Attacks    120    121   
      12/24/2011 12:33:33 AM -- 1 Attack     135   
      12/24/2011 12:33:40 AM -- 3 Attacks    170    250c   231c  
      12/24/2011 12:33:47 AM -- 4 Attacks    116    164    227c   228c  
      12/24/2011 12:33:54 AM -- 1 Attack     116   
      12/24/2011 12:34:01 AM -- 1 Attack     128   
      12/24/2011 12:34:07 AM -- 2 Attacks    172    163   
      12/24/2011 12:34:14 AM -- 2 Attacks    217c   162   
      12/24/2011 12:34:21 AM -- 1 Attack       m   
      12/24/2011 12:34:28 AM -- 2 Attacks    113    133   
      12/24/2011 12:34:45 AM -- 3 Attacks    144    258c   238c  
      12/24/2011 12:34:52 AM -- 3 Attacks    255c   157    121   
      12/24/2011 12:34:58 AM -- 3 Attacks    168    111    146   
      12/24/2011 12:35:05 AM -- 2 Attacks    129      m   
      12/24/2011 12:35:12 AM -- 2 Attacks    142    249c  
      12/24/2011 12:35:19 AM -- 1 Attack     121   
      12/24/2011 12:35:26 AM -- 1 Attack     134   
      12/24/2011 12:35:32 AM -- 1 Attack     114   
      12/24/2011 12:35:39 AM -- 4 Attacks    118    131    104    209c  
      12/24/2011 12:35:46 AM -- 3 Attacks    122    144    117   
      12/24/2011 12:35:53 AM -- 3 Attacks    163      m    162   
      12/24/2011 12:36:00 AM -- 2 Attacks    103    151   
      12/24/2011 12:36:07 AM -- 4 Attacks    154    151    249c   200c  
      12/24/2011 12:36:13 AM -- 3 Attacks    207c     m    128   
      12/24/2011 12:36:21 AM -- 2 Attacks    111    127   
      12/24/2011 12:36:27 AM -- 2 Attacks    248c   109   
      12/24/2011 12:36:35 AM -- 1 Attack     152   
      12/24/2011 12:36:42 AM -- 1 Attack     236c  
      12/24/2011 12:36:48 AM -- 3 Attacks      m      m    107   
      12/24/2011 12:36:55 AM -- 4 Attacks    161    121    111    227c  
      12/24/2011 12:37:02 AM -- 2 Attacks    108    226c  
      12/24/2011 12:37:09 AM -- 2 Attacks    130      m   
      12/24/2011 12:37:15 AM -- 3 Attacks    210c   246c   132   
      12/24/2011 12:37:22 AM -- 2 Attacks    112    117   
      12/24/2011 12:37:29 AM -- 2 Attacks    129    128   
      12/24/2011 12:37:36 AM -- 2 Attacks    135    123   
      12/24/2011 12:37:49 AM -- 2 Attacks     19     30c  
      12/24/2011 12:37:56 AM -- 2 Attacks     19     21   
      12/24/2011 12:38:03 AM -- 1 Attack      20   
      12/24/2011 12:38:09 AM -- 1 Attack      31c  
      12/24/2011 12:38:16 AM -- 1 Attack      23   
      12/24/2011 12:38:23 AM -- 4 Attacks     23     23     31c    19   
      12/24/2011 12:38:30 AM -- 2 Attacks     22     25   
      12/24/2011 12:38:38 AM -- 3 Attacks     32c    19     32c  
      12/24/2011 12:38:43 AM -- 4 Attacks     20     23     19     22   
      12/24/2011 12:38:50 AM -- 2 Attacks     31c    22   
      12/24/2011 12:38:57 AM -- 1 Attack      18   
      12/24/2011 12:39:04 AM -- 2 Attacks     20     30c  
      12/24/2011 12:39:11 AM -- 4 Attacks     22     19     30c    20   
      12/24/2011 12:39:17 AM -- 1 Attack      22   
      12/24/2011 12:39:24 AM -- 3 Attacks     19      m     31c  
      12/24/2011 12:39:31 AM -- 2 Attacks     23     21   
      12/24/2011 12:39:38 AM -- 3 Attacks     21     23     31c  
      12/24/2011 12:39:45 AM -- 1 Attack      23   
      12/24/2011 12:39:51 AM -- 2 Attacks     24     31c  
      12/24/2011 12:39:59 AM -- 4 Attacks     22     20     19     30c  
      12/24/2011 12:40:05 AM -- 2 Attacks     22     19   
      12/24/2011 12:40:12 AM -- 3 Attacks     22     23      m   
      12/24/2011 12:40:19 AM -- 1 Attack      20   
      12/24/2011 12:40:26 AM -- 1 Attack      19   
      12/24/2011 12:40:32 AM -- 3 Attacks     18     30c    31c  
      12/24/2011 12:40:40 AM -- 1 Attack      23   
      12/24/2011 12:40:47 AM -- 1 Attack      24   
      12/24/2011 12:40:53 AM -- 2 Attacks     18      m   
      12/24/2011 12:41:00 AM -- 1 Attack      24   
      12/24/2011 12:41:07 AM -- 1 Attack      23   
      12/24/2011 12:41:13 AM -- 3 Attacks     31c    19      m   
      12/24/2011 12:41:20 AM -- 2 Attacks     24     19   
      12/24/2011 12:41:27 AM -- 3 Attacks     31c    24     22   
      12/24/2011 12:41:34 AM -- 3 Attacks     20     19     23   
      12/24/2011 12:41:42 AM -- 1 Attack      23   
      12/24/2011 12:41:48 AM -- 3 Attacks     18     19     21   
      12/24/2011 12:41:55 AM -- 3 Attacks     18      m     19   
      12/24/2011 12:42:02 AM -- 1 Attack      20   
      12/24/2011 12:42:09 AM -- 1 Attack      21   
      12/24/2011 12:42:15 AM -- 1 Attack      21   
      12/24/2011 12:42:22 AM -- 2 Attacks     31c    19   
      12/24/2011 12:42:29 AM -- 1 Attack      25   
      12/24/2011 12:42:36 AM -- 1 Attack      30c  
      12/24/2011 12:42:43 AM -- 1 Attack      21   
      12/24/2011 12:42:49 AM -- 1 Attack      24   
      12/24/2011 12:42:56 AM -- 2 Attacks     18      m   
      12/24/2011 12:43:03 AM -- 3 Attacks     31c    23     25   
      12/24/2011 12:43:11 AM -- 1 Attack      21   
      12/24/2011 12:43:17 AM -- 2 Attacks     32c    23   
      12/24/2011 12:43:24 AM -- 1 Attack      31c  
      12/24/2011 12:43:31 AM -- 1 Attack      31c  
      12/24/2011 12:43:37 AM -- 2 Attacks     32c    31c  
      12/24/2011 12:43:44 AM -- 4 Attacks     21     21     24     24   
      12/24/2011 12:43:51 AM -- 4 Attacks     22     23     31c    31c  
      12/24/2011 12:43:58 AM -- 1 Attack      19   
      12/24/2011 12:44:04 AM -- 1 Attack      19   
      12/24/2011 12:44:11 AM -- 2 Attacks     24     30c  
      12/24/2011 12:44:18 AM -- 1 Attack      21   
      12/24/2011 12:44:25 AM -- 1 Attack      32c  
      12/24/2011 12:44:32 AM -- 2 Attacks     25     24   
      12/24/2011 12:44:39 AM -- 1 Attack      23   
      12/24/2011 12:44:45 AM -- 1 Attack      31c  
      12/24/2011 12:44:53 AM -- 2 Attacks     25     22   
      12/24/2011 12:44:59 AM -- 3 Attacks     19     30c    25   
      12/24/2011 12:45:06 AM -- 2 Attacks     25     19   
      12/24/2011 12:45:13 AM -- 2 Attacks     29c    18   
      12/24/2011 12:45:20 AM -- 4 Attacks     24     20     20     25   
      12/24/2011 12:45:26 AM -- 4 Attacks     19     20     19     21   
      12/24/2011 12:45:33 AM -- 3 Attacks     22     19     21   
      12/24/2011 12:45:40 AM -- 4 Attacks     24     17     31c    24   
      12/24/2011 12:45:47 AM -- 4 Attacks     30c    24     21     31c  
      12/24/2011 12:45:54 AM -- 2 Attacks     24     19   
      12/24/2011 12:46:01 AM -- 1 Attack      32c  
      12/24/2011 12:46:08 AM -- 1 Attack      20   
      12/24/2011 12:46:14 AM -- 2 Attacks     19     21   
      12/24/2011 12:46:21 AM -- 3 Attacks     23     21     19   
      12/24/2011 12:46:28 AM -- 2 Attacks     23     32c  
      12/24/2011 12:46:35 AM -- 1 Attack      32c  
      12/24/2011 12:46:42 AM -- 2 Attacks     23     23   
      12/24/2011 12:46:48 AM -- 2 Attacks      m     20   
      12/24/2011 12:46:55 AM -- 2 Attacks     32c    21   
      12/24/2011 12:47:02 AM -- 4 Attacks     22     20     19     20   
      12/24/2011 12:47:10 AM -- 1 Attack      18   
      12/24/2011 12:47:16 AM -- 1 Attack      31c  
      12/24/2011 12:47:23 AM -- 2 Attacks     20     31c  
      12/24/2011 12:47:30 AM -- 1 Attack      25   
      12/24/2011 12:47:36 AM -- 1 Attack      31c  
      12/24/2011 12:47:44 AM -- 2 Attacks     30c    21   
      12/24/2011 12:47:50 AM -- 2 Attacks     20     19   
      12/24/2011 12:47:57 AM -- 4 Attacks      m     22     25     20   
      12/24/2011 12:48:04 AM -- 2 Attacks     32c    31c  
      12/24/2011 12:48:11 AM -- 3 Attacks     18     32c    30c  
      12/24/2011 12:48:18 AM -- 2 Attacks     24     30c  
      12/24/2011 12:48:24 AM -- 2 Attacks     23     20


    Problem is I derped it up and did all this in 10% double attack, but I can work around it:

    Had to trim columns from the raw data and let notepad++ count the '1's, '2's, '3's, and '4's. KParser doesn't count quads:

    79 Singles
    95 Doubles
    40 Triples
    28 Quads

    32.6/39.2/16.5/11.5 raw

    Divide singles by 0.9 to correct for conversion to doubles by DA gear, subtract all from 100% to find new Doubles value:

    36.22/35.78/16.5/11.5

    So if I really wanted to take a stab, I'd guess the level 99 OA2-4 to be 36/36/16/12 distribution. To check this:

    501 hits / 242 rounds = 2.070 avg hits/rd

    36/36/16/12 = 2.04 avg hits/rd.

    Now add 10% of singles rate for extra hits from being converted to doubles by DA gear:

    2.04 + 0.036 = 2.076

    Not perfect, not extensive, but definitely a little different from my original 40/30/20/10 guess.

    This is just initial work. Mobs hanging around Pashhow again so I'll have to find a new zone to continue this parse.

  2. #2
    Old Merits
    Join Date
    Sep 2010
    Posts
    1,094
    BG Level
    6
    FFXI Server
    Ragnarok

    If you post the kparser data inside [code] tags it'll maintain the proper spacing on those columns. Makes things much easier to read.

  3. #3
    Custom Title
    Join Date
    Nov 2008
    Posts
    1,066
    BG Level
    6
    FFXI Server
    Diabolos

    Quote Originally Posted by Martel View Post
    If you post the kparser data inside [code] tags it'll maintain the proper spacing on those columns. Makes things much easier to read.
    Blame Kparser for using [pre] instead of [code]. Fixed and thank you.

    Also, /DNC, no Zanshin. I freaked out about this for half a second.

  4. #4
    Masamune
    Guest

    from what im reading, am i understanding correctly you "translated" the accidentaly equipped 10% DA into just a convertion of single hits into double hits ?
    If yes, i disagree, since there is no proof on priority order (at least not to my knowledge) between 1/2/3/4.../8 hits check at each round.
    If real OA2-4 distrib is like 30/40/15/15, it could be either :
    - a single random check RAND(0-99) with 0-30=single, 31-70=DA, 71-85=TA, 86-99=4x
    or
    - a priority order rolling a rand() for singles 1st > then another for DA if 1st failed, etc.. until Quad...
    or
    - something else more complciated involving a differenciation between DA/TA traits, and OAx...

    Anyways, i did this mistake also and best and by far easiest safest way i found were simply to redo the test without that accidental DA gear...

    But those results can still be used to compare with results w/o DA, so still can say good job.

    EDIT: also regarding parsing, since you were using a low delay weapon, did you use any haste gear that could make rounds too short for KP to parse correctly? (if i remember correctly, the min time per round for KP must be > 2.5secs?). With testing Quad and even for triple attacks, i think would even need to use hecatomb gears?

  5. #5
    Sea Torques
    Join Date
    Dec 2007
    Posts
    582
    BG Level
    5

    Similarly to Ridill, DA trait may also affect (attenuate) triple and quadruple attacks for Magian multi-attack weapons.

  6. #6
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    I thought there sorta was proof. Maybe not super mapped out but basically how wearing DA gear actuallly lowered your average number of hits while K-club zerging. Which means DA definitely took the place of some 3+ hit rounds that could of been

  7. #7
    Custom Title
    Join Date
    Nov 2008
    Posts
    1,066
    BG Level
    6
    FFXI Server
    Diabolos

    Quote Originally Posted by Masamune View Post
    EDIT: also regarding parsing, since you were using a low delay weapon, did you use any haste gear that could make rounds too short for KP to parse correctly? (if i remember correctly, the min time per round for KP must be > 2.5secs?). With testing Quad and even for triple attacks, i think would even need to use hecatomb gears?
    Uh, I suppose I never said this was my OA2-4 Scythe, lol. Certainly not 'low delay'. 528 even with gear haste is very well suited for KParser to bucket it properly (it just doesn't count quads except in the raw data, which you could have looked at to see my ~7sec delay). The only issue that should be encountered with KParser in this case is changing your delay at all (which is why I blew a few parses by popping Last Resort habitually).

    DA overriding Singles only and DA overriding all, I actually did both and only posted the former:

    Corrected for 10% DA overriding all
    35.86
    33.33
    18.15
    12.65

    Interestingly this gives the same hits/round average, 2.075. This shows well that DA gear has zero net effect on this distribution (and predictably all OA2-4 weapons then) anyway.

    Most of what I've heard about DA gear hurting KC zerg was only speculation based on this thinking. I've never actually seen the alleged data, and even if DA was singles-only no KC DRK really cared for it where you could instead have HP+ or a Souleater piece for 10x or more the change in output. KC is also the only weapon as far as I know that goes significantly past a 2.0 hits/round average, so it'd be the only weapon subject to that effect in anything but hit distribution anyway.

    However, my parse appears consistent enough that I could put this to rest with a second DA-free parse, which I'll be doing (edit) as soon as possible. You know I'm good at this abstract and OOO stuff.

    Edit: Pfft... Sandoria is 6/7, Bastok is 6/7, I'm not going into La Velue or Beadeaux because mobs will never leave the 'last stand', and Windurst isn't clearing mobs for crap as usual and battles will end by the time they do. Gonna have to find something else to beat on I think...

    The Zvahl Fortilaces people AFK skillup on will only get me maybe 35 rounds each. Three or four hours to get another ~250 round parse.

  8. #8
    Chram
    Join Date
    Sep 2007
    Posts
    2,526
    BG Level
    7
    FFXI Server
    Fenrir

    Quote Originally Posted by Raelia
    528 even with gear haste is very well suited for KParser to bucket it properly (it just doesn't count quads except in the raw data
    Er.. yes it does. It's the +3 count (base attack plus 3 additional attacks = 4 attack round). It lists 28, the same as your manually counted amount.

    Anyway, distribution..

    79 Singles
    95 Doubles
    40 Triples
    28 Quads

    Total rounds = 242

    Note: 242 rounds gives us a ~6% margin of error. As such I'm only rounding to the nearest 0.5.

    If DA procs first, expect 24 DAs. Distribution after that for OA2-4:

    79 Singles
    71 Doubles
    40 Triples
    28 Quads

    Total: 218

    Giving you distribution of 36%/32.5%/18.5%/13%

    Average of 2.085 hits per round from OAx alone.

    If OA2-4 procs first, then the 79 Singles will be 90% of the source set used for the DA procs, giving us 9 DAs. That leads to an OA2-4 distribution of

    86 Doubles
    40 Triples
    28 Quads

    79+9 Singles+DA

    From the original 242 round total, and a split of: 36.5%/35.5%/16.5%/11.5%

    Average of 2.03 hits per round from OAx alone.


    Actual average hits per round: 2.07.


    Percentages for 2/3/4 hits if OAx procs (assuming proc check is done in two stages):
    DA first: 51%/29%/20.5% (would imply a 64% chance of an OAx proc, and then 50%/30%/20% split)
    DA last: 56%/26%/18% (a rather messy set of numbers that don't lend themselves to any pattern)


    To see whether DA impacts overall hits, going to use a theoretical 50% DA rate.

    DA first (using 64%, 50/30/20 split):
    Out of 1000 rounds, expect:
    No DA: 2088 hits
    W/DA: 2044 hits

    A 2.1% reduction in overall number of hits when adding DA.

    DA last:
    Out of 1000 rounds, expect:
    No DA: 2030 hits
    W/DA: 2212 hits

    A 9% increase in overall number of hits when adding DA.

    Using the 10% DA rate, you get:

    DA first (using 64%, 50/30/20 split), 10% DA rate:
    Out of 1000 rounds, expect:
    No DA: 2088 hits
    W/DA: 2079 hits

    A 0.5% reduction in net number of hits.


    So the effect might be slight, but either way, DA does affect net hits per round when combined with OAx. If DA is checked first, DA will always reduce the average number of hits per round. If DA is checked last, it will always increase the average number of hits per round.

    I'm inclined at the moment towards believing that DA is checked first, and that the OA2-4 uses a ~64% proc check, then a 50/30/20 split for 2/3/4 hits.


    It would be nice if you could get a decent sample both without DA and with DA (getting over 20% should be viable for drk/war). A significant increase in average hits per round (~3.5% increase with 20% DA) would indicate that DA is checked last. If it appears to stay the same, that would be evidence of DA being checked first.

  9. #9
    Custom Title
    Join Date
    Nov 2008
    Posts
    1,066
    BG Level
    6
    FFXI Server
    Diabolos

    Quote Originally Posted by Motenten View Post
    would imply a 64% chance of an OAx proc, and then 50%/30%/20% split
    This I like. Would definitely explain the exclusivity of OAx, Double Attack, and other features.

    36/32/19.2/12.8 is a smidge closer to my earlier parse attempt that gave me the 40/30/20/10 guess. It could be that the OAx rate was bumped from 60% to 64% and the 2x/3x/4x distribution remained the same, and if I remember correctly I thought quads were a little high. With 60% OAx rate it would have looked like 40/30/18/12 and I vaguely remember it being close to that, Singles and Doubles being 40/30 especially, but have since lost the parse data.

    My annotation about DA having 'zero net effect' on OA2-4 weapons was as a worst-case where DA overrides all, just not perfectly clear in carried context. I can apply my 'if I were designing this system' approach and say that DA checking last is the only sensible arrangement, but that'll have to wait until I have a fort to beat on again.

    Edit: As a matter of fact, I'm going to carry that forward instead:

    60% OAx chance, 50/30/20 gives 2.02 hits/rd. 10% Double Attack checked last adds 0.04 for 2.06 hits/rd to approach well to my parse. Very clean and sensible numbers and proc progression, but I definitely need to test with more DA and without DA to be sure.

    With DA checked first, I've got 36/37/16.2/10.8 which is close to the original parse distribution but still lowballs the DA rate and definitely cuts the Quad rate a little low while Triples look okay. But this idea is damned by only averaging 2.018 hits/rd, significantly off from the parse.

  10. #10
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    K-club /___/___/AF3
    AF3+2/AF3/AF3/Brutal
    Versa+1/Dusk+1/Dark/Dark
    Atheling/Anguinus/AF3+2/Hermes
    15 or 17% + 21% + 30% from Atma = 66% or 68% DA rate (I haven't checked if they gave WAR DA II again)

    Hits per round, turning back and forth: 1 2 2 2 2 6 2 4 2 5 2 2 2 2 2 2 2 6 5 2 5 2 5 2 2 2 2 2 4 2 2 2 2 3 5 6 4 5 3 2 2 6 (fatal)

    25 2-hit rounds out of 42 rounds (60% 2-hit rate)

  11. #11
    Nidhogg
    Join Date
    Jul 2008
    Posts
    3,746
    BG Level
    7
    FFXIV Character
    Seraphus Highwynn
    FFXIV Server
    Gilgamesh
    FFXI Server
    Diabolos

    So for OA2 would DA gear reduce OAT or increase your odds of attacking twice per round in general and overall (be it from either source OAT or DA)?

  12. #12
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    In the case of K-club, it looks like it would reduce it. In the case of the OA2-4 weapon, it seems unlikely that it would actually matter given the samples we have gotten so far.

  13. #13
    Custom Title
    Join Date
    Nov 2008
    Posts
    1,066
    BG Level
    6
    FFXI Server
    Diabolos

    Quote Originally Posted by Ophannus View Post
    So for OA2 would DA gear reduce OAT or increase your odds of attacking twice per round in general and overall (be it from either source OAT or DA)?
    Dependent on the average hits/round of the weapon as it already stands. As OAT is around 1.4, DA will always increase it no matter if it comes first or last (just varying in effectiveness).

    DA coming first is good for <2.0 hits/rd weapons, while DA coming last is better for >2.0 hits/rd weapons. Only Kraken Club is truly subject to the latter, while OA2-4 magians are effectively neutral

    Still need to test OA2-4, since that flaw may be specific to Kraken Club and not to the augment effect.

  14. #14
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    It was also tested to be the case with Ridill, but I have an OA2-4 dagger I could try with using the same setup, I guess.

  15. #15
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    Also thought the same held true for quint spear...

  16. #16
    Custom Title
    Join Date
    Nov 2008
    Posts
    1,066
    BG Level
    6
    FFXI Server
    Diabolos

    Ridill is still only 1.9 hits/round, so DA is always an increase. Quint spear might be a smidge over 2.0 though.

    I don't doubt that augment versions don't function any differently from hardcoded really, but it'd be worth a look.

  17. #17
    Sea Torques
    Join Date
    Dec 2007
    Posts
    582
    BG Level
    5

    A distinct OAx rate cannot be proven, only inferred from the presence of "unusual" probability estimates, but obviously the estimates are so imprecise at this point that 40/30/20/10 is still admissible.

  18. #18
    Custom Title
    Join Date
    Nov 2008
    Posts
    1,066
    BG Level
    6
    FFXI Server
    Diabolos

    Quote Originally Posted by CDF View Post
    A distinct OAx rate cannot be proven, only inferred from the presence of "unusual" probability estimates, but obviously the estimates are so imprecise at this point that 40/30/20/10 is still admissible.
    Except Moten's assessment that all multihit capability of the weapon could fall under an initial 60% or 64% chance, and then provides a sensible distribution afterwards, is very compelling for why DA would override even superior hitcount rounds. 40/30/20/10 was based on the assumption that kparser was flawed and counting some triples as quads and trying overly to 'fit' the observed distribution to whole multiples of ten.

    Evidentialism is a plague around here.

  19. #19
    Sea Torques
    Join Date
    Dec 2007
    Posts
    582
    BG Level
    5

    Quote Originally Posted by Raelia View Post
    Except Moten's assessment that all multihit capability of the weapon could fall under an initial 60% or 64% chance, and then provides a sensible distribution afterwards, is very compelling for why DA would override even superior hitcount rounds. 40/30/20/10 was based on the assumption that kparser was flawed and counting some triples as quads and trying overly to 'fit' the observed distribution to whole multiples of ten.

    Evidentialism is a plague around here.
    DA has been shown to override triple attacks for Ridill without reference to multiple checks for OA2-3 so appeals to sensibility ring hollow.

    This is an issue of sampling variability, not "evidentialism." You cited 40/30/20/10, not me, but it's clear that your lack of data cannot even rule out your original 40/30/20/10 hypothesis. This doesn't mean accept 40/30/20/10 or something "nice-looking," but gather or cite more data.

  20. #20
    Custom Title
    Join Date
    Nov 2008
    Posts
    1,066
    BG Level
    6
    FFXI Server
    Diabolos

    I rescind that estimate for sake of assumptions I know I made for it particularly in light of more recent parsing producing the same slightly higher Quad result and affirmation of KParser's ability to distinguish between Triple+Single and a Quad.

    I'm allowed to say "No, that estimate was crap because I made half of it up" aren't I?

    Evidentialism because you want evidence that the estimate is invalid, while I instead point out that there was no evidence of it being entirely valid in the first place.

    But if you really want that, Moten gave about a 6% margin of error for the parse volume. 10% quad rate falls well outside this (11.5% * .094 = 10.8% minimum) since we know DA rate will only decrease it.