Item Search
     
BG-Wiki Search
+ Reply to Thread
Page 3 of 4 FirstFirst 1 2 3 4 LastLast
Results 41 to 60 of 72

Thread: Magic     submit to reddit submit to twitter

  1. #41
    Hydra
    Join Date
    Jul 2011
    Posts
    128
    BG Level
    3
    FFXI Server
    Ragnarok

    Well I didn't get around to testing Aeroja or Thunder5 more, but here's couple of other things...

    Thundaja -- V = 1005
    Impact -- V = 939, M = ~2.335, INT stat down = -21, and stacks with Burn, the INT stat down doesn't effect the damage of the first casting of Impact.

  2. #42
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    I remember seeing something about impact giving some percentage to -stats. Did you test for that or just got 21 for that particular mob. Might be annoying finding mobs with known decently different ints though lol

  3. #43
    Hydra
    Join Date
    Jul 2011
    Posts
    128
    BG Level
    3
    FFXI Server
    Ragnarok

    It's likely -20% then, as the mobs that were tested on were 105 and 106 INT worms. The INT of the caster didn't seem to affect the amount, and -21 did seem like a weird randomly chosen number, so -20% seems likely.
    I don't have Impact yet as Shinryu is being stingy on the cloak, so I can't do much testing for a more precise M value or on stat reduction amounts.

  4. #44
    Hydra
    Join Date
    Jul 2011
    Posts
    128
    BG Level
    3
    FFXI Server
    Ragnarok

    OK, I tested Thunder 5 earlier on an 81 INT ladybug mob in Abyssea Altepa, using brew.

    So... 999 INT - 81 INT = dINT+918
    MAB = brew + 40 trait + 10 Atma of the Full Moon = x10.5

    Thunder 5 damage = 31342
    If my M value of 9419/4096 was correct, the damage should be = 31332

    Code:
    dINT = +918
    M = 9419/4096
    V = 874
    MAB = 10.5
    
    918 * 9419/4096 = 2110.99658203125
    2110 + 874 = 2984
    2984 * 10.5 = 31332
    So, using a /65536 system...

    150704/65536 = 9419/4096 = too low for dINT+918.
    150705/65536 = both high enough for dINT+918, and low enough for dINT+237.
    150706/65536 = too high for dINT+237.
    150720/65536 = 9420/4096 = 2355/1024 = too high for dINT+944.

    So for Thunder 5, M = 150705/65536 (2.29957580566406)
    ... or otherwise, something even more complicated.


    Edit:
    To pass dINT+918, the M value has to be at least 2.29956427015251
    To pass dINT+237, the M value has to be lower than 2.29957805907173

    If dINT+938 passes with this current M value too, then it should work 100% for any dINT from +1 to +998.
    The actual M value used by the game may well be much simpler than my current value, as this complexity is probably just down to some computer rounding issue.
    Hopefully however, using this M value in a calculator will work 100% though.

  5. #45
    Hydra
    Join Date
    Jul 2011
    Posts
    128
    BG Level
    3
    FFXI Server
    Ragnarok

    OK, a few random things first...

    - Impact, as a dark element spell, is affected by the "Successive use enhances spell potency" bonus of Comet (the +5% bonus).

    - Impact's stat down effect is less effective if cast after elemental debuffs (Burn/Shock/etc) than if cast before.

    Example:
    Mob INT=105, casting Burn with more than 150 INT (INT-13), and Impact stat down = -20%.

    105 - 13 = 92 ... 92 - 20% = 74 INT
    105 - 20% = 84 ... 84 - 13 = 71 INT

    Impact cast after burn gave INT-18, but Impact cast before burn gave INT-21.
    I had to use this to turn a 105 INT worm into 71 INT, because I was having trouble finding one for dINT+928 in the following thing.

    - Now the main thing I want to post...
    150705/65536 doesn't work for Thunder 5's M value either, as it is too high for dINT+938 after all.
    So I changed it to 301409/131072, which is half way between 150704/65536 & 150705/65536...
    This M value was too low for dINT+928, so now I've changed it to 602819/262144, which is half way between 301409/131072 and 301410/131072...
    Now there's just one dINT left that this value may not work for... dINT+701...

    So...
    Code:
    602818/262144	2.29956817626953	Fails dINT+928, too low.
    dINT+928	2.29956896551724	Needs to be this or higher to pass.
    
    602819/262144	2.2995719909668		Passes both dINT+928 & dINT+938.
    
    dINT+701	2.29957203994294	The answer could differ if M isn't high enough for this.
    
    dINT+938	2.29957356076759	Needs to be lower than this to pass.
    602820/262144	2.29957580566406	Fails dINT+938, too high.
    So 602819/262144 is the Thunder 5's M value.. that is until dINT+701 is tested, (which will probably be impossible to test, ever!), where the M value may or may not need to be higher to cover dINT+701.
    M=2.29957 should be accurate enough for every other dINT from +1 to +998 anyway though.

  6. #46

    It might be better to take all the data you have and see if there isn't possibly a different model for M calculation that fits what you have and conforms with other M values as well, rather than going to increasing powers of 2 to make everything fit, since if you're just looking for a close estimate rather than a model for how these things are calculated server-side, 2.3 seems more than adequate. Also, if it were me (and I understand it's not, so take it for what you will) I'd use QD or Netherblast to confirm the worm MDB thing, because it's deceptively easy to take the numbers and get them to fit with a handful of data points.

    Just my 262144/131072 cents, though. I'll try to butt out now!

  7. #47
    xXNyteFyreXx420Sharingan
    Join Date
    May 2009
    Posts
    3,710
    BG Level
    7
    FFXI Server
    Fenrir

    You could have a friend get the necessary setup for that INT value and get them charmed by a NM in Abyssea if the necessary mob families have popped. Cirein-croin should be available even in Mis Coast's default state I think, if not... not sure, been over a year since mobs weren't popped etc. Poroggo in LTP perhaps. Hell of an effort, kudos to you. Wonder if other things would exhibit similar rounding errors under close scrutiny, seems likely.

  8. #48
    Hydra
    Join Date
    Jul 2011
    Posts
    128
    BG Level
    3
    FFXI Server
    Ragnarok

    @ Suiram
    Yeah, I'm trying figure out a different model, but the first part of doing that was hoping that these tests didn't come down to a precise number that just works for everything. I was hoping there would be conflicting answers where no possible decimal number could be used, but this test has almost ruled that out now for me.

    For worms having MDB, i'm pretty confident in my test in the post at the bottom of the previous page.
    I added 119 INT, with M=1.5 and MAB=1.5, and it added 267 damage.

    However... If for example the worm had the lowest tier of MDB=1.1, then the MAB/MDB term would = 1.5/1.1 (1.35)

    First cast, me with 106 INT...

    480 / 1.35 = 355.5555555556 -- check: 356 * 1.35 = 480.6
    356 - 320 = 36 -- subtracted V
    36 / 1.5 = 24 -- this is the dINT for the first cast, so the worm would have actually had 82 INT, instead of the 106 I believed it to be.

    Then I added 119 INT for the second cast...

    24 + 119 = 143 -- this is what the dINT would be for the second cast
    143 * 1.5 = 214.5 -- dINT * M
    214 + 320 = 534 -- added V
    534 * 1.35 = 720.9 -- D * MAB/MDB

    I got 747 damage when I cast the second spell, whereas this adds up to 720 damage instead (an increase of only +240 damage instead of +267).
    A higher tier than MDB1 would be even worse, and if somehow it was just something like MDB+1 (1.01), then the answer would still be slightly wrong.

    @Nightfyre
    Nice idea. I didn't even think about getting a 298 INT person charmed at all. That sounds like it would work.

  9. #49
    Hydra
    Join Date
    Jul 2011
    Posts
    128
    BG Level
    3
    FFXI Server
    Ragnarok

    OK, I've been trying to look into an idea someone said to me about a few days ago -- that maybe caster INT (cINT) and target INT (tINT) are multiplied by M seperately, and the rounding issue arises from this.

    Hopefully this is not the case, as this would mean that with the same dINT but different cINT vs. tINT, the damage could vary slightly.
    Also, the term "dINT" would become kind of meaningless.

    But, my current M value for Thunder 5 being the actual value the game uses sounds pretty implausible, as it is something like 20 bit value so far.
    So I need to look into different models like this, as Suiram suggested.

    Before that, here's a list of some Thunder 5 casts that I've been working from.
    D-V is the damage before adding the spell's V value, and before including terms like MAB, so it's just dINT*M but without knowing what the value of M is yet.
    Code:
    cINT	tINT	dINT	D-V (aka. dINT*M)
    356	110	246	565
    356	107	249	572
    999	106	893	2053
    999	81	918	2111
    999	71	928	2134
    999	61	938	2156
    999	55	944	2170
    dINT+928 and dINT+938 are the most interesting ones as I've shown earlier, and dINT+701 would be nice to test...
    Code:
    dINT	D-V	(D-V)/dINT	(D-V+1)/dINT
    246	565	2.2967479675	2.3008130081
    249	572	2.297188755	2.3012048193
    893	2053	2.2989921613	2.3001119821
    918	2111	2.2995642702	2.3006535948
    928	2134	2.2995689655	2.3006465517
    938	2156	2.2985074627	2.2995735608
    944	2170	2.2987288136	2.2997881356
    
    For dINT+701, D-V should be one of these two...
    
    701	1611			2.2995720399
    701	1612	2.2995720399	
    
    My current M value for Thunder 5 would result in 1611 for dINT+701...
    
    602819/262144	2.2995719909668
    The (D-V)/dINT column shows the minimum the M value needs to be to give the correct D-V answer.
    The (D-V+1)/dINT column shows at what value M would be too high and cause D-V to be +1 higher than it should be.

    Well so far I've testing the idea by using the following:
    Code:
    FLOOR(cINT * M) - FLOOR(tINT * M)
    I may not have fully understood the idea correctly, but if that's correct then if I use just my dINT+928 and +938 D-V values, the M value can't be anything/2048.

    My own idea at the moment is something along the lines of what I was trying to describe about recast time calculation in the "Cast and Recast" thread.
    The greater values take up more bits leaving less precision for the fraction part of the value.
    I think if it's something like that, then just finding the M value needed for the most precise situation (ie. between dINT+928, 701 & 938 in Thunder 5's case) should work for all the dINT which don't need such precision.

  10. #50
    Radsourceful

    Join Date
    Jul 2007
    Posts
    1,964
    BG Level
    6
    FFXI Server
    Bismarck

    Unrelated to current discussion, if you have a fool's drink active, utsusemi shadows are not wiped by monster's aga magic.

  11. #51
    Hydra
    Join Date
    Jul 2011
    Posts
    128
    BG Level
    3
    FFXI Server
    Ragnarok

    Well this is annoying... doing solo skillchains and magic bursts on SAM99/sch49 was nice and simple.
    Now with SCH99/sam49, it turns out I have a Magic Burst Bonus trait... took me 10 minutes to realise, lol...

    I assume the trait is simply added in the standard MB Bonus term?
    So for a 2 WS skillchain the MB Bonus term is *1.3, but with MB Bonus I trait also, the MB Bonus term would instead be *1.35 ?

    Wiki says SCH has MB Bonus II @ SCH89, and otherwiki only mentions SCH getting this trait once, just MB Bonus I @ SCH79.
    Both sites currently have the following:
    Code:
    MB Bonus I = +5%  (SCH79)
    MB Bonus II = +7%  (SCH89 [on BG wiki only])
    MB Bonus III = +9%
    MB Bonus IV = +11%
    MB Bonus V = +13%  [only BG wiki has this tier, at current]
    The only one that works for my SCH99/sam49 so far, if any of them, is +11%, supposedly from MB Bonus IV ?

    Well, if MB Bonus IV = +11% is correct for SCH99, then Klimaform is calculated after MB Bonus:
    Code:
    Resist > MB Bonus > Klimaform bonus > Day + Weather + Twilight Cape > MAB

  12. #52
    BG Content
    Join Date
    Jul 2007
    Posts
    21,103
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    I would have expected Scholar to get one more tier (99), but not two more. Perhaps they changed something in the last update (same time WAR lost its crit bonus III).

  13. #53
    Hydra
    Join Date
    Jul 2011
    Posts
    128
    BG Level
    3
    FFXI Server
    Ragnarok

    I wonder if they lowered the level SCH gets the first tier, or if SCH ever gets tier III or just jumps straight up to IV...
    I'll try to test SCH MB tiers somewhen, say SCH/sam lv78-79, 88, 89, and 90+.
    It's pretty borring waiting on Sekkanoki every time though, so I may never get around to it...

    I got my +11% MB bonus results while testing Resist > MB > Klimaform, with Elemental Skill = 0
    My MB's were with 1/2 and 1/8 resists, where 1.40 or 1.42 MB bonus gave wrong answers, and only 1.41 (1.3 + 0.11) would work.

    I've been playing around with Ul'hpemde and BLM AF3+2 Goetia set bonus...
    The Ul'hpemde at Al'Taieu in the area where you come in from Sealion's Den have INT = 53, 54 and 55.

    I'm only casting while their mouths are open, I've been getting a lot of conserve MP procs, but zero Goetia +2 set bonus procs so far...
    I've had procs with just two pieces before on different mobs, and I've done half my casts with 2 pieces, half with 3 pieces.
    But even though I'm getting conserve MP procs quite regularly, I still haven't had a single Goetia +2 set proc since I got there...

    I'm starting to wonder if Goetia set +2 and whatever makes Hpemde take double damage are part of the same term, and maybe there's a cap on it ?
    Or maybe they changed something in an update, and now you need to wear all 5 pieces for the set bonus?

  14. #54
    BG Content
    Join Date
    Jul 2007
    Posts
    21,103
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    You can also get Icarus Wings or use the scholar skillchain JA. I tended to do the latter.

  15. #55
    Hydra
    Join Date
    Jul 2011
    Posts
    128
    BG Level
    3
    FFXI Server
    Ragnarok

    Icarus Wing was my first idea, but they have a 2 hour medication effect... so you'd have to keep running backwards and forwards to Lower Jeuno.
    You're the second person to mention using a SCH skillchain JA... so I guess something like that really does exist... :/

  16. #56
    BG Content
    Join Date
    Jul 2007
    Posts
    21,103
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    It's honestly not very convenient because it restricts (me at least) to pretty low tier spells.

  17. #57
    Old Merits
    Join Date
    Sep 2010
    Posts
    1,094
    BG Level
    6
    FFXI Server
    Ragnarok

    Cleric's Pantaloons' +2 "Enhances Shellra V"

    I suppose this would go here.

    I just finished my alt's Cleric's Pantanloon's +2 and got the "Enhances Shellra V" augment. So I went out to the Ectozoon's in abyssea Ule to test.

    99 PLDWAR. No MDT/DT gear. 67 INT. 5/5 Shellra Merits

    Base test. No buffs.
    Code:
      Spells
        Quake
    +^      959:    4
        Stone IV
    +^      684:    5
        Stonega III
            576:    1
    +^      721:    5
    With Relic enhanced Shellra V.
    Code:
    Martel
      Spells
        Quake
    +^      678:    4
        Stone IV
    +^      483:    4
        Stonega III
            253:    1
    +^      509:    4
    So...
    678/959 = 29.30%
    483/684 = 29.38%
    509/721 = 29.40%

    I'd tend to go by quake's numbers. Being the largest they should be more accurate.

    So, a 29.3% reduction. About a 2% increase from the +2 legs. I think that'd be 75/256. 5/256 being from the legs.

  18. #58
    Hydra
    Join Date
    Jul 2011
    Posts
    128
    BG Level
    3
    FFXI Server
    Ragnarok

    Here's some interesting stuff I have for the initial "D" term, which is what I want to test once I finish testing the calculation order:

    - ALL tier I and II black magic elemental nukes inflect halfway to their cap values. This inflection point is also known by some people as the "half-life".
    - Stone III, however, doesn't have an inflection point.
    - Stone III's cap for "D" is 681. Taking away it's "V" (210) gives 471, then remove "M" (1.5) too and you are left with 314.

    I tested Stone III with dINT=320 (cINT=362, tINT=42)
    dINT=320 -- D=681
    dINT=315 -- D=681
    dINT=314 -- D=681
    dINT=313 -- D=679

    I think for all black and white magic spells where the M value is 1.0, they probably all have a half-life where their M halves to 0.5.
    If you brew Thunder II and Stone III, with cINT=999, the difference between them is similar to between Blizzard II and Thunder II...
    However, between Thunder III and Stone IV, there's a big difference in damage of over +33%, probably because it's likely neither spell has an inflection point.

    - All tier III and IV single target black magic nukes cap below dINT=879
    - AM I and II all seem to cap at the same dINT value, dINT=879.

    I tried to test if Stone 5 has a cap, but it was hard finding out my dINT with the methods I used to reach higher dINT.
    The best method I've found so far is: Brew + Impact + Absorb-INT, as BLM99/drk49 vs. Pachypodiums (which all start out as tINT=55).
    It seemed like Stone 5 may have capped at dINT=973, as I can hit at least dINT=979 on them if I cast spells quick enough after Absorb-INT...

    There's interesting patterns and similarities with the cap and half-life values of BLM AoE nukes too.
    I'll try and write up a full table for the "D" term soon...

    ---

    My test for MB Bonus > Klimaform might not be finished though...

    SCH99 MB Bonus trait isn't 11%... or 5%, 7%, 9% or 13%...

    I'm still assuming this % is simply added to the regular MB Bonus 30% for a 2-WS skill chain, so with MB Bonus I trait would be +35%, or *1.35 ?
    I'm also assuming the regular MB Bonus without job traits is definitely exactly 30% (*1.3000000) for the 2-WS SC...

    If that's correct...
    - SCH99 MB Bonus is higher than 11% but lower than 12%
    - MB Bonus requires more accuracy than 3 decimal places
    - Using a /1024 number would not be accurate enough, as the MB term for SCH99 is a number between 1+426/1024 and 1+427/1024...

    After just two casts so far, the simplest binary system number that would work is 1+853/2048 (1.4165...)
    Hopefully I'm just doing this completely wrong...
    Is the MB Bonus trait supposed to be used as a separate term from the normal MB Bonus?
    Is it supposed to instead be something like: floor(floor(...*1.3)*trait) ?

  19. #59
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    Actually I think it might. Not sure if it still does but other wiki has had MB bonus gear applied seperately the natural dmg bonus. Even had them seperate from each other too.

  20. #60
    Hydra
    Join Date
    Jul 2011
    Posts
    128
    BG Level
    3
    FFXI Server
    Ragnarok

    You probably mean this page right?
    http://wiki.ffxiclopedia.org/wiki/Us...n#Magic_Bursts

    I wonder how the current 5, 7, 9, 11, 13 % trait values were determined. I haven't seen where it was tested...