Item Search
     
BG-Wiki Search
+ Reply to Thread
Page 14 of 23 FirstFirst ... 4 12 13 14 15 16 ... LastLast
Results 261 to 280 of 454
  1. #261
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    So did some hot shots. 89 str 100 agi with 6 dmg light crossbow and 1 dmg bolt so that should be definitely capped fstr of 16 and 7 base dmg from weapon and another 70 from wsc for a total of 93 on ws 23 for regular shots. No mab

    Shot some sand hares in the face for 69 dmg each time so definitely capped pdif. The physical first hit is expected to be 279.

    1000 tp- 418 dmg... which fits perfect with the old .5 ftp for the magical
    2000 tp - 711 so ~ 1.55
    3000 tp - 864 so 2.1 ftp

  2. #262
    Impossiblu
    Join Date
    Mar 2010
    Posts
    10,362
    BG Level
    9
    FFXIV Character
    Prothescar Centursa
    FFXIV Server
    Balmung
    FFXI Server
    Valefor

    Dimidiation Attack Bonus/Penalty

    Spoiler: show

    Attack Bonus:
    Target - Lv82 Greater Colibri (67VIT, 327DEF)
    Stats - 33D, 572 Attack, 134STR, 173DEX

    Base Damage - 33 + 141 + 11 = 185
    wRatio = 1.749235474006116
    qRatio UL = 2.124235474006116 * 1.05 = 2.230447247706422
    qRatio LL = 1.271582927178899

    Predicted max randomizer damage - (185 * 3.25) = 601 601 * 2.230447247706422 = 1340
    Predicted lowest randomizer damage - (185 * 3.25) = 601 601 * 1.271582927178899 = 764




    Sample Data
    1102
    1153
    1196
    1278
    1281


    tl;dr it doesn't seem to have either. sample size isn't huge but an attack bonus or penalty would be so easily noticeable that it doesn't really have to be any bigger than this.

    Puts Dimidiation at:

    80% DEX
    2.25fTP @ 1000
    4.50fTP @ 2000
    6.75fTP @ 3000

    fTP Transfer: No.
    0% Attack Bonus/Attack Penalty

  3. #263
    Chram
    Join Date
    Sep 2007
    Posts
    2,526
    BG Level
    7
    FFXI Server
    Fenrir

    Shining Strike

    fTP increased at 1000 TP, so testing at that point.

    Using Warp Cudgel (no iLvl)

    WSC: STR: 40%, MND: 40%

    Code:
    WSC     Dmg
    71      362
    77      372
    87      388
    97      404
    107     420
    +10 WSC => +16 dmg, so clearly the fTP is 1.6 (the 71 WSC is for another point later).

    However my formula does not work for this. (152 + 77) * 1.6 = 366, which leaves no room for dInt vs observed damage of 372 (only 6 points off).

    If I back-calculate things, I find the fLvl component must be 136 (mostly) to get those results:

    Code:
    fLvl    WSC     dInt    Predict     Actual      Diff
    135     71      32      361         362         1
    135     77      32      371         372         1
    135     87      32      387         388         1
    135     97      32      403         404         1
    135     107     32      419         420         1
    -
    136     71      32      363         362         -1
    136     77      32      372         372         0
    136     87      32      388         388         0
    136     97      32      404         404         0
    136     107     32      420         420         0
    Even that, however, is off by 1 at the 71 WSC point.

    I can fix that by using a base /256 fTP instead of a flat 1.6. 1 + 153/256 (1.59765625) yields a perfect match.

    To prove that, I built a set to reach 171 WSC (225 Str, 203 Mnd), exactly +100 over one of the earlier samples. If WSC is less than an exact 1.6, I should get +159 to total damage.

    Code:
    WSC     Dmg
    71      362
    171     524
    Diff == 162

    ...

    212 str, 203 mnd:
    Code:
    WSC     Dmg
    166     516
    (516-362)/(166-71) = 1.62105

    Other comparisons similarly show ratios of about 1.62. Possibly 1.625? No, can't be that because then the +10 WSC samples would increase by +16, +17, +16, etc, and they're always a constant +16.

    Adding one more +10 test:

    Code:
    WSC     Dmg
    77      372
    87      388
    97      404
    107     420
    117     437
    And there we see it going up by more than 16, but only on the fourth iteration. That means it's 1.6225 or less. Might as well use a simple 1.62.

    Looking at the other samples:
    Code:
    WSC     Dmg
    71      362
    121     498 << lvl 113 weapon
    171     524
    Wanted to check an iLvl weapon to see how it fits in, and whether testing using a weapon whose level was well below 99 (ie: Warp Cudgel, lvl 36) was affecting things. Turns out that if you factor out the 2.45 per level from the iLvl portion, the remainder fLvl value is 133. Same (roughly) as the other data.



    So, from all this, all I can conclude is that clubs for some reason use a different fLvl formula. Initial tentative value of (level * 1.33) + 2. fILvl is still the same: (WeaponLevel - 99) * 2.45.

    And fTP at 1000 TP for Shining Strike is 1.62.


    Final test was to see where magic damage fell into the equation.

    Tamaxchi -- lvl 115, +99 m.dmg, +12 MAB
    WSC 121
    Damage = 678

    After factoring out MAB, that's 606. If magic damage is added at the same stage as dInt, the expected total would be 605. I'm not sure where the extra 1 point discrepency is coming from, but the 99 magic damage is clearly added in after fTP.


    Overall formula:

    fLvl =
    -- for clubs: PlayerLevel * 1.33 + 2, or (PlayerLevel + 1) * 1.33
    -- for swords/guns/other?: (PlayerLevel + 1) * 1.5 + 2
    fILvl =
    -- for iLvl weapons: (WeaponLevel - 99) * 2.45
    -- otherwise 0
    dStat = dInt or dAgi or whatever, per weaponskill
    mDmg = Magic Damage
    WSC = WSC for weaponskill

    Damage = (fLvl + fILvl + WSC) * fTP + dStat + mDmg

  4. #264
    Old Merits
    Join Date
    Sep 2010
    Posts
    1,094
    BG Level
    6
    FFXI Server
    Ragnarok

    Quote Originally Posted by Prothescar View Post
    Dimidiation Attack Bonus/Penalty

    Spoiler: show

    Attack Bonus:
    Target - Lv82 Greater Colibri (67VIT, 327DEF)
    Stats - 33D, 572 Attack, 134STR, 173DEX

    Base Damage - 33 + 141 + 11 = 185
    wRatio = 1.749235474006116
    qRatio UL = 2.124235474006116 * 1.05 = 2.230447247706422
    qRatio LL = 1.271582927178899

    Predicted max randomizer damage - (185 * 3.25) = 601 601 * 2.230447247706422 = 1340
    Predicted lowest randomizer damage - (185 * 3.25) = 601 601 * 1.271582927178899 = 764




    Sample Data
    1102
    1153
    1196
    1278
    1281


    tl;dr it doesn't seem to have either. sample size isn't huge but an attack bonus or penalty would be so easily noticeable that it doesn't really have to be any bigger than this.

    Puts Dimidiation at:

    80% DEX
    2.25fTP @ 1000
    4.50fTP @ 2000
    6.75fTP @ 3000

    fTP Transfer: No.
    0% Attack Bonus/Attack Penalty
    Thinking of retesting Drakes atk penalty. I doubt it's changed. but... must know for sure.

    Any particular procedures I need to follow to duplicate this testing with drakesbane?

  5. #265
    Relic Weapons
    Join Date
    Apr 2007
    Posts
    348
    BG Level
    4
    FFXI Server
    Ragnarok

    I did some salvage II for kicks with the ilvl 115 WKR lance with OaT on drg and I found that Drakes damage {comparatively} sucked unless I applied dia 2 and angon. Of course, this is all eyeballing and extremely low sample size and uncontrolled but it's not looking good, my friend.

    Although, It was pretty fun doing self double light without dropping my damage into the trash.

  6. #266
    Campaign
    Join Date
    Jul 2007
    Posts
    6,633
    BG Level
    8

    Where in this thread are the Katana tests? I must have missed them.
    I only remember a few random posts, nothing extensive like we've seen for other weapon categories.

  7. #267
    Old Merits
    Join Date
    Sep 2010
    Posts
    1,094
    BG Level
    6
    FFXI Server
    Ragnarok

    I'm not really expecting it to have changed. But I need to KNOW it hasn't.

  8. #268
    Sea Torques
    Join Date
    Aug 2008
    Posts
    699
    BG Level
    5
    FFXI Server
    Odin

    Drg got pushed even further into the shitter this update. Might as well save us the trouble and just drop our Ryunos.

  9. #269
    Old Merits
    Join Date
    Sep 2010
    Posts
    1,094
    BG Level
    6
    FFXI Server
    Ragnarok

    What factors would make atk bonus/penalty testing require the least samples?

    Like, would it be best to use a low dmg weapon? I assume ensuring only one hit by WS'ing a nearly dead mob is good. I think I vaguely recall doing that for drakes back when...

    Using a mob with known stats would be useful.

    For drakes, should I go abyssea and cap crit rate? Can't really SA when I need to hit a near dead mob.

    Also, would one go about testing ignores def ion about the same manner?

  10. #270
    First invited, last in the zone.
    Join Date
    Sep 2008
    Posts
    1,449
    BG Level
    6
    FFXI Server
    Lakshmi

    Didn't see it done elsewhere, so very small amount of Impulse Drive testing:

    119 Ryu, 190 STR, SA starting area goblins

    2000 TP: 4433 -> between 2.99 and 3.14 fTP (guessing 3.00)
    3000 TP: 8002 -> between 5.4 and 5.675 fTP (guessing 5.5)

    Useless regardless <_<

  11. #271
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    All right so cloudsplitter. Using ilvl 119 weapon, 201 str/179 mnd no mab on 6 int mobs (though shouldn't matter) so 201 for level factor 152 for wsc.

    1000 tp - 1323 so still 3.75 ftp
    2000 tp - 2364 so 6.6968. 6.7 doesn't work though So let's say 6 + 179/256 for now.
    3000 tp - 3000 so 8.4985. 8.5 works so probably that

    So pretty meh. Previously it's high ftp at higher tps somewhat made up for the lack of dstat... but it hardly gained much compared to others and now several magical ws have dstat AND higher ftps then it so crap got crappier in comparison

  12. #272
    Yoshi P
    Join Date
    Dec 2006
    Posts
    5,360
    BG Level
    8
    WoW Realm
    Arthas

    How about primal rend?

  13. #273
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    Primal rend- First thing I wanted to check was what the heck "The magic accuracy and attack power of the weapon skill itself will also be raised" meant if anything might just have been a typo with the attack but had to make sure.

    First went to check if they changed the dstat. So ws at 108 tp each time, 1779 on 6 int mob that went to 1786 after soundblast so looks like at least the multiplier is the same.

    Using ilvl 119 weapon, 211 dex /205 chr no mab on 6 int mobs so 298 dchr, 201 for level factor and 186 for wsc.

    1000 tp supposedly wasn't changed so I should be getting 1773 (based on bg wiki) but only got 1483... so looks like maybe they got the stat mods wrong or ftp or something or it was wrong before. Switched to ilvl 106 so 169 for level component and dmg went to 1385... that gives an average loss of 3.0625 dmg per level factor so given that looks like ftp really was way lower than previously noted. Bg had it at 3.8125 from studio Gobli and other wiki had it at 4.0 So not sure if everyone else was wrong, it got ninja nerfed awhile back and no one noticed or in their attempt raise the attack power they actually lowered 1000 tps ftp

    Anyways...

    1000 tp - 1483 so 3.0625
    2000 tp - 2556 so 5.8346 or possibly 5+214/256
    3000 tp - 3224 so 7.5607 or probably 7.5625

    Overall from what I've tested like axes kinda got jipped on the ws update in the magical department... saw some physical tested earlier but no numbers put to it but they didn't look impressive either... neither does eyeballing in the field. Ruinator is doing the same like it should and nothing is really surpassing it though Onslaught is pretty close now. Looks like they forgot about bst in the ws update compared to the crazy dmg others are doing now

  14. #274
    Impossiblu
    Join Date
    Mar 2010
    Posts
    10,362
    BG Level
    9
    FFXIV Character
    Prothescar Centursa
    FFXIV Server
    Balmung
    FFXI Server
    Valefor

    Quote Originally Posted by Martel View Post
    Thinking of retesting Drakes atk penalty. I doubt it's changed. but... must know for sure.

    Any particular procedures I need to follow to duplicate this testing with drakesbane?
    Drakesbane is more complex since it has so many hits and is piercing, so it can likely kill the colibri with a non-static number of hits. But you can isolate that more easily if you only land one hit (WS below 10% should do the trick).

    1) Equip non-ilevel weapon
    2) Ensure that your attack does not cap (cap for lv 82 Greater Colibri = 736)
    3) Normally you'd need to control TP so that you get exact amounts, but since Drakesbane's modifier is crit, it doesn't matter for you.
    4) Math the expected High and Low damage for both a crit and a non-crit.
    5) Compare your data to those.

  15. #275
    Old Merits
    Join Date
    Sep 2010
    Posts
    1,094
    BG Level
    6
    FFXI Server
    Ragnarok

    Does it matter if you cap fSTR for this?

  16. #276
    Chram
    Join Date
    Sep 2007
    Posts
    2,526
    BG Level
    7
    FFXI Server
    Fenrir

    More on Shining Strike..... (starting to hate club)

    Ok, did two weaponskills at each TP value that were exactly 50 WSC apart.

    Code:
              71 WSC     121 WSC    Diff  Implied fTP
    1000TP       362         443      81        1.62
    2000TP       669         819     150        3.00
    3000TP      1031        1262     231        4.62
    We can consider those fTP values to be solid, since the only thing that changed is WSC, and we know that WSC is multiplied by fTP.

    However, given that, the remainder of the formula just doesn't add up. It's *stable*, in that the difference between predicted and actual remains constant for each of the WSC values at each fTP level, but they're not constant between TP levels.

    Actually... Hmm. Interesting. If I zero out the dInt modifier, and can put fLvl back at 152, and it -almost- matches up. Everything is either exact, or off by 1.

    The ones that are exact are at 3.0 fTP. The one that are off by 1 are the 1.62 and 4.62 fTPs. Considering those are possibly inexact due to integer math, trying to see if another fTP value will cause things to match up..

    And what do you know? Matches become exact at 1.625 and 4.625 fTP values for 1000 and 3000 TP, respectively. -Including- being off by exactly 99 on the Tamaxchi magic damage test.

    Spreadsheet data:
    Code:
                            Base Str        89                                                                                                                      
                            Base Mnd        76                                                      12 MAB                                                  
            Shining Strike                                                                          99 MDmg                                                 
                                                                            iLvl                    iLvl                                                    
                                                                            113                     115                                                     
            +Str    5       20      23      39      48      60      71      60      123     136     60              12      70      71              12      70
            +Mnd    8       9       31      39      55      69      67      79      127     127     78              1       69      67              1       69
    40%     Str     94      109     112     128     137     149     160     149     212     225     149             101     159     160             101     159
    40%     Mnd     84      85      107     115     131     145     143     155     203     203     154             77      145     143             77      145
            WSC     71      77      87      97      107     117     121     121     166     171     121             71      121     121             71      121
                                                                                                                                                            
    Dmg @1000       362     372     388     404     420     437     443     498     516     524     606     2000 TP 669     819     819     3000 TP 1031    1262
                                                                                                    678                                                     
                                                                                                                                                            
    dStat   0       151.77  151.923 151.77  151.62  151.46  151.92  151.62  185.46  151.54  151.46  251.92          152.00  152.00  152.00          151.92  151.86
    fLvl    152                                                             34                      39                                                 
    fTP     1.625   362     372     388     404     420     437     443     498     516     524     507     3       669     819     819     4.625   1031    1262
    
    Diff            0       0       0       0       0       0       0       0       0       0       99              0       0       0               0       0
    So, solved. Shining Strike (and possibly light-based weaponskills) do not have any dStat value. fLvl and fILvl remain the same.

    Shining Strike fTP: 1.625 > 3.0 > 4.625



    Cleaned up overall formula:

    fLvl =
    -- (PlayerLevel + 1) * 1.5 + 2
    fILvl =
    -- for iLvl weapons: (WeaponLevel - 99) * 2.45
    -- otherwise 0
    dStat = dInt or dAgi or whatever, per weaponskill; some do not have any dStat
    mDmg = Magic Damage
    WSC = WSC for weaponskill

    Damage = (fLvl + fILvl + WSC) * fTP + dStat + mDmg

  17. #277
    Impossiblu
    Join Date
    Mar 2010
    Posts
    10,362
    BG Level
    9
    FFXIV Character
    Prothescar Centursa
    FFXIV Server
    Balmung
    FFXI Server
    Valefor

    Quote Originally Posted by Martel View Post
    Does it matter if you cap fSTR for this?
    no, just be sure to account for it when you're doing calculations.

  18. #278
    Old Merits
    Join Date
    Sep 2010
    Posts
    1,094
    BG Level
    6
    FFXI Server
    Ragnarok

    Ok, here's some drakes samples to retest the attack penalty. For whoever wants to math.

    Oh, I hope I don't need capped fSTR for this. Cause I'm not capping.
    Spoiler: show

    Greater Colibri level 82 (67 VIT, 327 DEF)
    Daboya 140 dmg
    str 113
    atk 567

    Drakes
    265
    355
    302
    371
    602
    615
    591
    711
    301
    392
    592
    652
    262


    Just looking at the gap between my melee hits and non crit drakes(favors the melee hits)... Gonna say the atk- is still there.

    Probably still -20%, but would like to verify.

  19. #279
    Chram
    Join Date
    Sep 2007
    Posts
    2,526
    BG Level
    7
    FFXI Server
    Fenrir

    With the dStat issue solved, here's data for Seraph Strike:

    Str: 101
    Mnd: 77
    WSC: 71

    1000 TP: 473
    2000 TP: 808
    3000 TP: 1365

    Going back to standard units of /16 for fTP, it comes out to fTP of: 2.125 > 3.625 > 6.125

  20. #280
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    I retested some Atonement stuff:
    https://www.bg-wiki.com/bg/Talk:Atonement

Similar Threads

  1. Weapon Skill setup for nin
    By Starblaster in forum FFXI: Everything
    Replies: 27
    Last Post: 2007-08-08, 10:45
  2. Paralysis and weapon skills
    By Mifaco in forum FFXI: Everything
    Replies: 81
    Last Post: 2007-04-16, 15:11
  3. Question about Weapon Skill Gorget for Relic Noob
    By Hilander in forum FFXI: Everything
    Replies: 6
    Last Post: 2007-03-28, 19:36
  4. Weapon Skill Points
    By Metah in forum FFXI: Everything
    Replies: 21
    Last Post: 2007-03-04, 12:12
  5. Weapon Skill Accuracy
    By Kurp in forum FFXI: Everything
    Replies: 6
    Last Post: 2006-12-30, 18:39
  6. Weapon Skill Chart (Not Skillchain)
    By Kaiseran in forum FFXI: Everything
    Replies: 8
    Last Post: 2006-10-26, 20:11
  7. Weapon skills where Double Attack can proc.
    By Koyangi in forum FFXI: Everything
    Replies: 111
    Last Post: 2006-06-11, 12:58
  8. Question about a Weapon Skill.
    By Twisted Angel in forum FFXI: Everything
    Replies: 9
    Last Post: 2005-10-11, 19:11
  9. Relic Weapon Skills
    By BruceLee in forum FFXI: Everything
    Replies: 5
    Last Post: 2005-07-18, 23:15