Item Search
     
BG-Wiki Search
+ Reply to Thread
Page 7 of 47 FirstFirst ... 5 6 7 8 9 17 ... LastLast
Results 121 to 140 of 930

Thread: DPS Spreadsheets     submit to reddit submit to twitter

  1. #121

    Quote Originally Posted by Motenten View Post
    Your formula (the critical % average formula) doesn't appear to be working, pchan.

    What I did:

    First, expanded it out so I could be sure I'm getting all the parentheses in there:

    Code:
    crit%=
    (c+
      (lambda1+1)*
      (-c+a*c-a)/
      (-1+a)+
      1/
      (-1+a)*
      (-
        (-c+a*c-lambda1-1+
          (lambda1+1)*
        a-a)*
        a^(lambda1+1)/
        (-1+a)+
        a^(lambda1+1)*
        (lambda1+1+c)
      )-
      1/
      (-1+a)*
      (-
        (-c+a*c-1)*
        a/
        (-1+a)+
        a*
        (1+c)
      )-
      (-c+a*c-lambda1-1+
        (lambda1+1)*
      a-a)*
      a^(lambda1+1)/
      (-1+a)+
      a^(lambda1+1)*
      cap1+
      (N-2-lambda1)*
      (-
        (-c+a*c-lambda1-1+
          (lambda1+1)*
        a-a)*
        a^(lambda1+1)/
        (-1+a)+
        (-c+a*c-a)/
        (-1+a)+
        a^(lambda1+1)*
      cap1)
    )/N
    Then, using a small chunk of the spreadsheet, entered/calculated values:
    Code:
        AG        AH      AI      AJ
                  base    max     hits to cap
    6   acc       95%     95%     0
    7   crit      12%     62%     50
    8   crit dmg  0%      50%     50
    9   att                       50
    10  hits              144

    Then did a search/replace on the formula you provided, substituting in the appropriate Excel cells:
    Code:
    =
    (ah7+
      (ai7+1)*
      (-ah7+ah6*ah7-ah6)/
      (-1+ah6)+
      1/
      (-1+ah6)*
      (-
        (-ah7+ah6*ah7-ai7-1+
          (ai7+1)*
        ah6-ah6)*
        ah6^(ai7+1)/
        (-1+ah6)+
        ah6^(ai7+1)*
        (ai7+1+ah7)
      )-
      1/
      (-1+ah6)*
      (-
        (-ah7+ah6*ah7-1)*
        ah6/
        (-1+ah6)+
        ah6*
        (1+ah7)
      )-
      (-ah7+ah6*ah7-ai7-1+
        (ai7+1)*
      ah6-ah6)*
      ah6^(ai7+1)/
      (-1+ah6)+
      ah6^(ai7+1)*
      aj7+
      (ai10-2-ai7)*
      (-
        (-ah7+ah6*ah7-ai7-1+
          (ai7+1)*
        ah6-ah6)*
        ah6^(ai7+1)/
        (-1+ah6)+
        (-ah7+ah6*ah7-ah6)/
        (-1+ah6)+
        ah6^(ai7+1)*
      aj7)
    )/ai10

    Then collapsed it all together again to copy into Excel:
    Code:
    =(ah7+(ai7+1)*(-ah7+ah6*ah7-ah6)/(-1+ah6)+1/(-1+ah6)*(-(-ah7+ah6*ah7-ai7-1+(ai7+1)*ah6-ah6)*ah6^(ai7+1)/(-1+ah6)+ah6^(ai7+1)*(ai7+1+ah7))-1/(-1+ah6)*(-(-ah7+ah6*ah7-1)*ah6/(-1+ah6)+ah6*(1+ah7))-(-ah7+ah6*ah7-ai7-1+(ai7+1)*ah6-ah6)*ah6^(ai7+1)/(-1+ah6)+ah6^(ai7+1)*aj7+(ai10-2-ai7)*(-(-ah7+ah6*ah7-ai7-1+(ai7+1)*ah6-ah6)*ah6^(ai7+1)/(-1+ah6)+(-ah7+ah6*ah7-ah6)/(-1+ah6)+ah6^(ai7+1)*aj7))/ai10
    With the above values, the result was 45.53356941.
    You mixed cap1/2 and lambda1/2 I think in your formula. A correct formula would be

    Code:
    (ah7+(aj7+1)*(-ah7+ah6*ah7-ah6)/(-1+ah6)+1/(-1+ah6)*(-(-ah7+ah6*ah7-aj7-1+(aj7+1)*ah6-ah6)*ah6^(aj7+1)/(-1+ah6)+ah6^(aj7+1)*(aj7+1+ah7))-1/(-1+ah6)*(-(-ah7+ah6*ah7-1)*ah6/(-1+ah6)+ah6*(1+ah7))-(-ah7+ah6*ah7-aj7-1+(aj7+1)*ah6-ah6)*ah6^(aj7+1)/(-1+ah6)+ah6^(aj7+1)*ai7+(ai10-2-aj7)*(-(-ah7+ah6*ah7-aj7-1+(aj7+1)*ah6-ah6)*ah6^(aj7+1)/(-1+ah6)+(-ah7+ah6*ah7-ah6)/(-1+ah6)+ah6^(aj7+1)*ai7))/ai10
    which returns 27.60982892 with your numbers.

  2. #122

    Anyway, I swinged 10,000 rounds this night

    MNK90/NIN10
    62 STR, 65 dex ( => 10% base crit% with maxed merits)
    6% haste, dual lady bell (delay 216)
    I manually counted the #rounds during impetus that's N=58

    My formula predicts 12.75074922% crit rate bonus during impetus, so 17.65044953% global. I found

    Code:
    Melee Damage
    Player            Melee Dmg  Abs'd.Dmg   Net Dmg   Melee %   Hit/Miss    M.HR %   M.Acc %  M.Low/Hi  M+0.Avg  M-0.Avg
    Pimpchan                  0          0         0    0,00 %  10013/524   95,03 %   95,03 %       0/0     0,00Non Num9rique
    
    
    Melee Crit Damage
    Player                #Crit  C.Low/Hi   C-0.Avg     Crit%
    Pimpchan               1746       0/0      0,00   17,44 %
    not too far.

  3. #123
    Chram
    Join Date
    Sep 2007
    Posts
    2,526
    BG Level
    7
    FFXI Server
    Fenrir

    Quote Originally Posted by pchan View Post
    You mixed cap1/2 and lambda1/2 I think in your formula. A correct formula would be

    Code:
    (ah7+(aj7+1)*(-ah7+ah6*ah7-ah6)/(-1+ah6)+1/(-1+ah6)*(-(-ah7+ah6*ah7-aj7-1+(aj7+1)*ah6-ah6)*ah6^(aj7+1)/(-1+ah6)+ah6^(aj7+1)*(aj7+1+ah7))-1/(-1+ah6)*(-(-ah7+ah6*ah7-1)*ah6/(-1+ah6)+ah6*(1+ah7))-(-ah7+ah6*ah7-aj7-1+(aj7+1)*ah6-ah6)*ah6^(aj7+1)/(-1+ah6)+ah6^(aj7+1)*ai7+(ai10-2-aj7)*(-(-ah7+ah6*ah7-aj7-1+(aj7+1)*ah6-ah6)*ah6^(aj7+1)/(-1+ah6)+(-ah7+ah6*ah7-ah6)/(-1+ah6)+ah6^(aj7+1)*ai7))/ai10
    which returns 27.60982892 with your numbers.
    Re-did the conversion, ended up with a formula identical to yours (verified), however the result of the calculation is now 13.39298598. Not much variance with number of hits, either. 12.1 with 50 hits, 13.95 with 1000 hits.

  4. #124
    New Spam Forum
    Join Date
    Jun 2010
    Posts
    183
    BG Level
    3
    FFXI Server
    Quetzalcoatl

    Quote Originally Posted by pchan View Post
    Anyway, I swinged 10,000 rounds this night

    MNK90/NIN10
    62 STR, 65 dex ( => 10% base crit% with maxed merits)
    6% haste, dual lady bell (delay 216)
    I manually counted the #rounds during impetus that's N=58

    My formula predicts 12.75074922% crit rate bonus during impetus, so 17.65044953% global. I found

    Code:
    Melee Damage
    Player            Melee Dmg  Abs'd.Dmg   Net Dmg   Melee %   Hit/Miss    M.HR %   M.Acc %  M.Low/Hi  M+0.Avg  M-0.Avg
    Pimpchan                  0          0         0    0,00 %  10013/524   95,03 %   95,03 %       0/0     0,00Non Num9rique
    
    
    Melee Crit Damage
    Player                #Crit  C.Low/Hi   C-0.Avg     Crit%
    Pimpchan               1746       0/0      0,00   17,44 %
    not too far.
    So it appears going from 1swing/rnd>2swings/rnd does stay the same then. Your sample size however does allow a confidence interval of crit during impetus to be as low as 20% or so?? I would expect that @2swings/rnd crit would be between 1-2% less than what 1swing/rnd would be, so idk. Maybe a test with as much multi-hit gear as possible? I think staying /nin you can get something like 4QA,12TA,24DA in gear, but I know you have to get str and dex down low enough, so can't wear everything I suppose.

    With best gear you would have 3.1064swings/rnd, so would be nice the closer to that number as possible. More importantly instead of having minimum swings@2,max@2, it would allow for minimum@2,max@8 which should show a much larger difference from 1swing/rnd if in fact there actually is a difference.

  5. #125

    @motenten

    subsituting your value into the formula :
    Code:
    subs({ah6=0.95,ai10=144,ah7=12,ah8=0,ai7=62,ai8=50,aj7=50,aj8=50},(ah7+(aj7+1)*(-ah7+ah6*ah7-ah6)/(-1+ah6)+1/(-1+ah6)*(-(-ah7+ah6*ah7-aj7-1+(aj7+1)*ah6-ah6)*ah6^(aj7+1)/(-1+ah6)+ah6^(aj7+1)*(aj7+1+ah7))-1/(-1+ah6)*(-(-ah7+ah6*ah7-1)*ah6/(-1+ah6)+ah6*(1+ah7))-(-ah7+ah6*ah7-aj7-1+(aj7+1)*ah6-ah6)*ah6^(aj7+1)/(-1+ah6)+ah6^(aj7+1)*ai7+(ai10-2-aj7)*(-(-ah7+ah6*ah7-aj7-1+(aj7+1)*ah6-ah6)*ah6^(aj7+1)/(-1+ah6)+(-ah7+ah6*ah7-ah6)/(-1+ah6)+ah6^(aj7+1)*ai7))/ai10);
                                 27.60982892
    ah6=0.95 (accuracy rate),
    ai10=144 (N),
    ah7=12 (base crit%),
    ah8=0 (base critdmg%, useless here),
    ai7=62 (crit% cap as an integer),
    ai8=50 (critdmg% cap as an integer),
    aj7=50 (hits cap cap crit%),
    ,aj8=50 (hits to cap critdmg%)

    50 hits => 23.98478180 (+11.98%)
    144 hits => 27.60982892 (+15.60%)
    3000 hits =>29.44549108 (+15.44%)

    Maybe the difference is due to priority on operations ? For maple a/b*c means (a/b)*c

  6. #126
    xXNyteFyreXx420Sharingan
    Join Date
    May 2009
    Posts
    3,710
    BG Level
    7
    FFXI Server
    Fenrir

    Quote Originally Posted by Motenten View Post
    ... That auto-linking is really getting to be obnoxious, isn't it?
    Can opt out here. Going to make a related thread in Stanvanced, keeping that thread tucked away in a corner of BG Tech when it applies exclusively to the FFXI section was probably not the greatest idea.

  7. #127
    Chram
    Join Date
    Sep 2007
    Posts
    2,526
    BG Level
    7
    FFXI Server
    Fenrir

    General update note:

    Well, first attempt at a blu spreadsheet.

    I put in a fair bit of prep work for future spell damage calculations (so I included misc +int or +MAB type gear, for example), but that doesn't work yet; right now it's melee-only.

    Allows you to configure your set spells, and automatically sets appropriate traits if applicable. Don't expect the spells that are set by default to mean anything; I just put in a bunch of random stuff while testing to make sure things worked.

    No guarantees that Sanguine Blade is constructed properly yet; was put in just for setting up gear sets for it.

    I'm sure I missed plenty of gear options, so let me know if you want stuff added.

  8. #128
    xXNyteFyreXx420Sharingan
    Join Date
    May 2009
    Posts
    3,710
    BG Level
    7
    FFXI Server
    Fenrir

    Awesome, really appreciate this. What is the "Other TP Gain/Loss" field for?

    March field should cap at +5 rather than +4 (99 ghorn is +4).

    Gear: Epona's Ring, White Tathlum, Romanus Cape, Kaikias' Cape, Potentia Cape, Vigilance Cape +1, Ire Torque +1, and the multiattack Khadgas.

  9. #129
    Chram
    Join Date
    Sep 2007
    Posts
    2,526
    BG Level
    7
    FFXI Server
    Fenrir

    Quote Originally Posted by Nightfyre View Post
    Awesome, really appreciate this. What is the "Other TP Gain/Loss" field for?

    March field should cap at +5 rather than +4 (99 ghorn is +4).

    Gear: Epona's Ring, White Tathlum, Romanus Cape, Kaikias' Cape, Potentia Cape, Vigilance Cape +1, Ire Torque +1, and the multiattack Khadgas.
    "Other TP Gain/Loss" is for accounting for Step/Samba use, mostly. It's other TP you spent during any given weaponskill cycle that slows down the weaponskill rate.

    March cap actually hadn't been updated on most of the spreadsheets, so fixed all of them, as well as an update to acc from Stalwart's.

    Gear added.

    Uploading new versions now.

  10. #130

    nvm

  11. #131
    Theory Fighter
    Join Date
    Jan 2011
    Posts
    1,427
    BG Level
    6
    FFXIV Character
    Frejan Schultz
    FFXIV Server
    Ragnarok
    FFXI Server
    Ragnarok

    How could I simulate Kenkonken 99 30% boost to Stringing Pummel? If I just make a weapon with KKK 99 stats and 30% WSDmg, it won't work, because it will only apply to the first hit.

  12. #132
    Chram
    Join Date
    Sep 2007
    Posts
    2,526
    BG Level
    7
    FFXI Server
    Fenrir

    By reminding me that it's missing from pup spreadsheet. Fixed and uploaded.

  13. #133
    Theory Fighter
    Join Date
    Jan 2011
    Posts
    1,427
    BG Level
    6
    FFXIV Character
    Frejan Schultz
    FFXIV Server
    Ragnarok
    FFXI Server
    Ragnarok

    Thank you!

  14. #134
    CoP Dynamis
    Join Date
    Jun 2010
    Posts
    252
    BG Level
    4
    FFXI Server
    Siren

    nvm

  15. #135
    Chram
    Join Date
    Sep 2007
    Posts
    2,526
    BG Level
    7
    FFXI Server
    Fenrir

    Quote Originally Posted by pchan
    Maybe the difference is due to priority on operations ? For maple a/b*c means (a/b)*c
    Double checked, and Excel follows standard operation order. a/b*c is calculated as (a/b)*c. a*b^c is a*(b^c). b^c*a is (b^c)*a.



    Certain terms used frequently throughout the formula:
    (-1+AH6) == a-1
    (AJ7+1) == lambda1+1
    (ah6*ah7) == a * c == acc * baseCrit

    Going to use
    b = a - 1
    m = lambda1 + 1
    d = a * c

    Consolidated the result down:

    Code:
    crit%=
    (
      c+
      m*(d-a-c)/b+
      1/b*
      (-
        (m*a+d-lambda1-a-c-1)*
        a^m/b+a^m*(m+c)
      )-
      1/b*
      (-
        (d-c-1)*
        a/b+a+d
      )-
      (-
        d+c-lambda1-1+m*a-a
      )*
      a^m/(-1+a)+
      a^m*cap1+
      (N-2-lambda1)*
      (-
        (d-c-lambda1-1+m*a-a)*a^m/b+
        (d-c-a)/b+
        a^m*cap1
      )
    )/N
    Code:
    crit%=
    (
      AH7+
      AK7*(AL7-AH6-AH7)/AK6+
      1/AK6*
      (-
        (AK7*AH6+AL7-AJ7-AH6-AH7-1)*
        AH6^AK7/AK6+AH6^AK7*(AK7+AH7)
      )-
      1/AK6*
      (-
        (AL7-AH7-1)*
        AH6/AK6+AH6+AL7
      )-
      (-
        AL7+AH7-AJ7-1+AK7*AH6-AH6
      )*
      AH6^AK7/(-1+AH6)+
      AH6^AK7*AI7+
      (AI10-2-AJ7)*
      (-
        (AL7-AH7-AJ7-1+AK7*AH6-AH6)*AH6^AK7/AK6+
        (AL7-AH7-AH6)/AK6+
        AH6^AK7*AI7
      )
    )/AI10

    Still getting the exact same results. Line-by-line (treating paren segments with indented contents as a single line), the above formula gives values of:

    12%
    975.12
    -2000%
    -1.388856804
    -20
    -18.05
    -3.494
    -1.46195453
    0.04532059
    92
    14.03970801
    144


    Final bonus value: 13.393

    Was wondering if any of those individual segments come out differently for you. Haven't been able to figure out where that difference is coming from.

    Or perhaps if you could try converting your own code to an Excel formula and see if you have the same problem I have.

  16. #136

    ok, that's because your spreadsheet takes 0.12 for c=AH7 and 0.62 for AI7=cap1. So reiteraring the parameters of the formula like in my post 125 :

    ah6=0.95 (accuracy rate),
    ai10=144 (N),
    ah7=12 (base crit%),
    ah8=0 (base critdmg%, useless here),
    ai7=62 (crit% cap as an integer),
    ai8=50 (critdmg% cap as an integer),
    aj7=50 (hits cap cap crit%),
    ,aj8=50 (hits to cap critdmg%)
    if you take 0,12 and 0,50 you find 13,39 but you should take 12 and 62 instead.

    also some buffs suggestion to add :

    minuets (daurdabla allows 4 songs).
    boostdex / str, the minimum is +5 stats and the maximum is +25, last 5 minutes.
    Ifrit's enfire tends to increase a lot damage too. Parser says it averages to 39-40 per melee hit.

  17. #137
    Chram
    Join Date
    Sep 2007
    Posts
    2,526
    BG Level
    7
    FFXI Server
    Fenrir

    Ah, dumb mistake, of course.

    Ok, implemented your formulas for crit rate and crit damage. Also adapted the crit rate formula for accuracy and attack. Accuracy doesn't feed back on itself, but it is unlikely to be off by more than 1 point of acc, so not going to worry about it. Crit rate, crit damage and attack are based on the hit rate after the Impetus accuracy value is added in.

    With these changes, it looks like Tantra+2 body gets a pretty nice boost as a TP piece. A few quick checks had it approaching or surpassing Thaumas body, depending on conditions (difficulty of mob, hit rate, etc), though I didn't put any time into optimizing gear sets for the comparison.

    New version with Impetus changes has been uploaded.

    I'll look into the other buffs later. Ifrit's enfire is likely skill based as well, though, as I've seen it at 33 damage per hit.

  18. #138
    Relic Shield
    Join Date
    Jun 2007
    Posts
    1,627
    BG Level
    6

    in any case the new data strengthens the choice of moonshade (TP bonus ) and Thaumas head for victory smite, during impetus (but not oustide). Although, i have cases (at ~55%critical hit rate, ie capped dDEX and impetus up) where augmented shura (with WSD str and att) is on par/slightly better.

  19. #139
    xXNyteFyreXx420Sharingan
    Join Date
    May 2009
    Posts
    3,710
    BG Level
    7
    FFXI Server
    Fenrir

    BLU sheet doesn't recognize setting STP II trait (4 spells), stat totals for the "Total 2" column don't include the values for the K column, and Fencer doesn't appear to be implemented for single-wield /WAR.

    Any plans to add BRD AF3+2 set bonus, Boost-STR, and Boost-DEX?

    EDIT: Phorcys Korazin has 5% WSD on the DRK spreadsheet instead of 7%.

  20. #140
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    I tried to ghetto-add Single Wielding to the DNC sheet after doing AM3 and it totally didn't work. Does it already have the support and I just don't realize it?

+ Reply to Thread
Page 7 of 47 FirstFirst ... 5 6 7 8 9 17 ... LastLast