Item Search
     
BG-Wiki Search
+ Reply to Thread
Page 4 of 15 FirstFirst ... 2 3 4 5 6 14 ... LastLast
Results 61 to 80 of 294
  1. #61
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    Yeah ES definitely doesn't have a 100% land rate. I've actually got a 1/8 resist with ES freeze II against a worm in south gustaberg once <.<. Very ironically was showing off dmg at 75 against complete trash and I get a 1/8000 chance resist

  2. #62
    Sea Torques
    Join Date
    Dec 2007
    Posts
    582
    BG Level
    5

    More Subtle Sorcery testing

    See my previous post for Matsui comments on BLM SP2.

    Wind Elemental

    23 Anemohelix
    368 Pyrohelix

    23 is 1/16th of 368.

    Aero damage
    Without Subtle Sorcery: 25
    During Subtle Sorcery: 50

    Seems like removal of (50%) damage reduction but not sure.

    Gnarring Yztarg (level 111; Kamihr Drifts)

    Gnarring Yztarg take 50% earth magic damage reduction and have high earth resistance.

    Stone damage without Subtle Sorcery
    567: 1
    283: 2
    140: 2
    68: 17

    Stone damage during Subtle Sorcery
    1138: 10

    My interpretation
    It seems that Subtle Sorcery removes elemental damage reduction (reduces "damage resistance") and provides a large magic accuracy boost (this can be thought of as removing evasion to magic damage or "accuracy resistance").

    If there is 100% "accuracy" resistance (as with elementals to same-element magic), Subtle Sorcery will have no effect on this. However it will still remove damage reduction or "damage resistance."

    Edit: more data to support my interpretation

    Genbu results with Subtle Sorcery (I have earth and thunder merits)
    Code:
    		Normal		Subtle Sorcery
    Stone		507
    Water		62		124
    Aero		504		1011
    Fire		63		126
    Blizzard	517
    Thunder		1086		1086
    Ionohelix	949		949
    Luminohelix	458		919
    Noctohelix	458
    Genbu seems to take at least a 50% damage cut for all elements except thunder. Subtle Sorcery removes damage reduction but doesn't affect evasion to fire and water damage.

  3. #63
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    So I've been playing around with a theory about magic hit rate (hit rate being the odds that you fully land a spell) and how it's changed since they added immunobreak. Basically my theory is prior hit rate was either determined thru a comparison of meva to macc and then capped to 5% minimum or 95% maximum or it was just set to 0 which is what we used to call immune. And that was your hit rate and then odds of partial resist would be based on that. And all that added together gets you landing rate on debuffs. Basically the info on the wiki pages about this though somewhere else in there the resist traits might factor in needs more testing.

    Then immunobreak was added and some of those mobs that previously were immune would display the message "completely resists the spell" and are now what we refer to as immune. Some would just say resist and sometimes say immunobreak. And that the previous hit rate was no basically a base hit rate after which some amount of hit rate is directly added by the immunobreak and now also stymie allowing you to land said enfeeble but without which you can't. And that it most likely just adds a hit rate to spells that aren't otherwise immune as well.

    Anyways since I have been since early on until wotg a career rdm I've had alot of experience with sleep and had never once landed it on a skeleton so I chose that for my test mob. Found a low level one right inside of gusgen and went on a fully skilled rdm with Bura. Ie should be waaaaaaaaaaaaaaay overcapped as far as macc vs meva goes. Results.

    No stymie no immunobreak: 0/260. Think it's somewhat safe to say that's a flat out 0% hit rate.

    Stymie no immunobreak: 2/2. Obviously this will be the most annoying to get a decent sample size of but looks good so far.

    No stymie 1 immunobreak: 4/29. Rather small sample partly because you have to get alot of resists just to have this chance and I ran into other problems I'll explain later but ~13% for what it's worth.

    No stymie 2 immunobreak: 3/3. Even worse sample size but yay 100% hit rate

    So in conclusion as far as the mobs with what I'll call a soft immunity to something it seems like my theory holds up. Hopefully I or maybe others can add to the size to get an approximate hit rate additions which leads me to a warning as to the problem I ran into. At the start I had about 1/8 resist be an immunobreak. At my 10th immunobreak I landed sleep and was only about 60 or so no stymie or immunobreak resists. I then did not get a single immunobreak in the next 100 or so casts. So I will tentatively say you are allowed a max 10 immunobreaks per mob. It will be however noted that I still landed stymie sleep after that point

    Edit: Bleh can't believe I forgot this but I forgot to check durations on all of them so what I have there is a land rate not a fully unresisted hit rate. I can say for certain that 1 of the stymie was 1 min and the other was 30 seconds and that approximately half of the lands with immunobreak were full duration.

  4. #64
    Sea Torques
    Join Date
    Dec 2007
    Posts
    582
    BG Level
    5

    Based on people's reports I think that Stymie may just effectively cap out magic hit rate at 95% since Matsui said Stymie won't wear off if the enfeeble resists (?). (Then again Stymie not wearing off may just account for those enfeebles that "completely resist") If you had 4 out of 5 Stymie+Sleep at 60 seconds that's still consistent with 95%. 3 of 5, less so.

    (Speculation on mechanisms that are practically impossible to verify) The mechanism of how Stymie effectively caps out magic hit rate may be either a very large magic accuracy bonus that's overkill even for Delve bosses (dispelling Dakuwaqa through the shield or landing Paralyze II on Tojil) OR that it simply sets the magic hit rate at 95% OR that it drops magic evasion to 0 (hard to tell either way). Let's say magic evasion to Sleep for skeletons was set at 999 (impossible to determine; this is just a hypothetical). Then it would be practically impossible to land that spell even with the newer gear.

    Perhaps meva is just set to 0 for Stymie since Matsui said Stymie will "guarantee the spell to land ignoring resistance" and ES is clearly insufficient for some situations (sleeping Tojil, at least for me, and paralyzing Uptala). (BTW is Tojil even susceptible to dark sleep?) Then again... this is inconsistent with Sleep having a 0% land rate on skeletons (the floor "should" be 5% for the usual macc/meva check). I would just go with simply capping magic hit rate at 95%.

    Also this confirms that Immunobreak itself is subject to "resistance" that is distinct from the resistance to the particular spell (which is 100% for sleep on skeletons normally). This is sort of consistent with the statement "If the [Immunobreak] effect is continuously applied, the enemy will slowly gain more resistance, and eventually Immunobreaks will no longer occur." (This is from the version update notes.)

    Immunobreak may just modify meva and/or alter hit rate directly (impossible to tell) since adding magic accuracy appears to be ineffective for something known to be 100% resistant to sleep.

    I'm quoting the version update details below as a refresher

    http://forum.square-enix.com/ffxi/th...Version-Update

    A new system has been added in which using enfeebling magic multiple times will lower the target enemy's resistance. This system is known as "Immunobreak."
    • When an Immunobreak occurs after the enemy has resisted a particular enfeebling magic, the resistance of the enemy to that effect will decrease by a certain amount.
    • Immunobreaks can stack and continually lower an enemy's resistance.
    • The more times the enfeebling magic is resisted, the greater the likelihood that an Immunobreak will occur.
    • When the enfeebling effect is successfully applied, the resistance will reset.
    • If the effect is continuously applied, the enemy will slowly gain more resistance, and eventually Immunobreaks will no longer occur.
    • Note that some monsters have complete resistance to enfeebling effects. (Immunobreaks will not occur on enemies which have complete resistance.)
    • Immunobreaks only apply to enfeebling effects (magic abilities based on Enfeebling Magic skill level).

  5. #65
    Relic Horn
    Join Date
    Aug 2008
    Posts
    3,131
    BG Level
    7
    FFXI Server
    Sylph

    According to Matsui (I think it was him), Stymie does not protect against partial resists, which fits with my experience; I've been playing around with Stymie against Naakuals, and when I use it with Saboteur + Sleep II (which should have a duration of about 2:20), I get wildly varying duration times; everywhere from about 30 seconds to a bit over a minute.

    Dasva's data (getting a 30 second duration on sleep) seems to agree, and further implies that specific element MDT doesn't affect duration under Stymie, which means that the Naakuals are all heavily resisting Sleep duration.

  6. #66
    Sea Torques
    Join Date
    Dec 2007
    Posts
    582
    BG Level
    5

    I thought Saboteur+Sleep I would result in duration at least 120 seconds (base 60 seconds). With AF3+2 hands duration goes up to 60*2.2 = 132 seconds or 2:12

    Sleep duration also has only 3 states normally afaik (60 seconds, 30 seconds, full resist). It sounds like you are getting more than 3 states so that is something else that I didn't account for.

  7. #67
    First invited, last in the zone.
    Join Date
    Sep 2008
    Posts
    1,449
    BG Level
    6
    FFXI Server
    Lakshmi

    Quote Originally Posted by CDF View Post
    I thought Saboteur+Sleep I would result in duration at least 120 seconds (base 60 seconds). With AF3+2 hands duration goes up to 60*2.2 = 132 seconds or 2:12

    Sleep duration also has only 3 states normally afaik (60 seconds, 30 seconds, full resist). It sounds like you are getting more than 3 states so that is something else that I didn't account for.
    most
    saboteur is nerfed on most NMs to only add 25% duration/potency which rises to 45% with gloves. 90*1.45=130.5

  8. #68
    Relic Horn
    Join Date
    Aug 2008
    Posts
    3,131
    BG Level
    7
    FFXI Server
    Sylph

    You sure? I thought both Saboteur and the gloves were halved in potency on NMs, not that Saboteur was quartered and the gloves were full potency.

  9. #69
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    I would lean more to direct hit rate for immunobreak and stymie. Mostly because if it was macc or -meva you'd still have at least 5% chance to land it fully and another almost 5% to land with half duration. Though I wouldn't discount maybe adding some macc as well.

    As far as protecting from partials iirc the note on that was kind of vague and confusing since half resists are calculated off of full hit rate. Might have just been a funny way of saying you wont have a 100% hit rate idk

  10. #70
    Sea Torques
    Join Date
    Dec 2007
    Posts
    582
    BG Level
    5

    If you keep getting partial durations this could be a sign that Stymie imposes a hit rate cap less than 95% for spells that have soft immunity. But I would just let the data lead the hypothesis generation instead of me continuing to speculate.

  11. #71
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    Yeah I'm working on it. Needless to say it will take awhile with 1 hour recasts. I will say this I just got another 1/2 resist. And while a sample size of 3 is pretty horrible but with the real high land rate but low hit rate I'm starting to wonder if maybe this does something reallly really odd. Like instead of modifying hit rate and then having partial resists based on that rate like it normally is it might it just say add 50% to full land rate and 50% to partial (not necessarily those amount but something that would add up to at least 100%) or something weird like that. idk gunna see if I can automate it somewhat to get more samples today

    Edit: another half resist

  12. #72
    Sea Torques
    Join Date
    Dec 2007
    Posts
    582
    BG Level
    5

    More on Subtle Sorcery

    Hurkan results with Subtle Sorcery - maximum damage observed

    Code:
    		Normal		Subtle Sorcery
    Noctohelix	666 (day proc)	1117
    Luminohelix	67		1369 (day proc)
    Ionohelix	78		159
    Geohelix	1255		1287
    Pyrohelix	606		1245
    Anemohelix	75		154
    Hydrohelix	606		1213
    From the standpoint of magic damage (not accuracy), Naakuals are weak to two elements and strong to the opposite of those elements. For example, Hurkan is weak to earth and ice damage (and incidentally strong to thunder and wind damage).

    It is also known that Naakuals take a 50% damage reduction for nukes corresponding to elements to which they are "not weak", and further that nukes corresponding to elements to which they are strong will always inflict 1/8 resist damage (so in effect this the 1/16 resist that people observe - 1/8 resist damage and 50% reduction of that damage).

    This is consistent with previous results showing that Subtle Sorcery removes elemental damage reduction but otherwise has no effect on "inherent" resistance to a given element.

    Edit: all Naakuals seem to be weak to two elements (meaning you can get full damage without Subtle Sorcery) but not necessarily strong to the opposite elements. E.g., Kumhau does not even appear to have "100%" resistance (meaning 100% 1/8 resist) to ice damage (I still get full ice damage with Subtle Sorcery) but is weak to fire and dark damage. Yumcax seems resistant to water, earth, and thunder but is weak to wind and fire damage.

    Again this is for magic damage, not accuracy. Kumhau has higher magic evasion to ice magic in general than to fire damage.

  13. #73
    Sea Torques
    Join Date
    Dec 2007
    Posts
    582
    BG Level
    5

    Helix damage - ceiling of dINT dependence of base damage

    Background
    Apparently it is known that the M multiplier for helix base damage is 0.5 after dINT = 78 such that the base damage formula is

    25 + dINT, 0 ≤ dINT ≤ 78
    103 + 0.5*(dINT-78), dINT > 78

    This makes it easier to determine whether there is a ceiling or cap on the dINT contribution to base damage (since MAB shouldn't be a confounder), as well as whether contributions from the magic damage attribute are independent of this ceiling.

    I already showed (reconfirmed?) that V remains 0.5 for dINT up to 210.

    Conclusion

    It appears that ceiling of helix base damage is 181, which occurs when dINT = 234 or dINT = 235. However, magic damage permits helix base damage to exceed 181 (at least up to magic damage 205).

    It is impossible to tell whether it's 234 or 235 via truncation as 103 + 0.5*(234-78) = 181 and 103 + 0.5*(235-78) = 181.5. My guess is that it's really 234 since 25+78+0.5*(156) = 181

    Results (target Tiny Mandragora, INT = 6)
    Spoiler: show
    Code:
    INT	dINT	MAB	MDMG	Actual	Predicted
    220	214	126	0	386	386
    223	217	126	0	388	388
    227	221	109	0	363	363
    135	129	39	10	191	191
    262	256	190	205	1119	1151
    247	241	165	10	506	514
    220	214	126	0	386	386
    230	224	126	0	397	397
    233	227	126	0	400	400
    241	235	126	0	409	409
    247	241	103	0	367	373
    243	237	103	0	367	369
    240	234	103	0	367	367
    244	238	86	0	336	340
    240	234	86	0	336	336
    243	237	86	0	336	338
    241	235	86	0	336	336
    233	227	134	195	870	870
    241	235	134	195	879	879

  14. #74
    Sea Torques
    Join Date
    Dec 2007
    Posts
    582
    BG Level
    5

    Magic - base damage calculation beyond dINT 200

    The July 2013 version update notes describe the base damage formula changes for some nukes but don't describe what happens for dINT > 199.

    • Tier II single-target nukes have dINT ceiling 200, after which M = 0.
    • For Tier III single-target nukes M = 1 for dINT >= 200
    • For Stone IV, M = 2 is possible for dINT >= 200. For Thunder IV, M = 1.95 is possible for dINT >= 200 (subject to further verification).
    • For Stone V, M = 3 is possible for dINT >= 200. For Thunder IV, M = 2.95 is possible for dINT >= 200 (subject to further verification).
    • There does not appear to be a global base damage ceiling accounting for the magic damage attribute (if one exists it hasn't been encountered yet)


    This is a baseline to predict what the M will be for AoE nukes, ancient magic and Comet for dINT between 200 and 300 (can play around with M values until expected equals observed across the range of elements), as well as what M would be for tier III nukes beyond dINT 300 (probably M = 0).

    1) Magic damage 0, MAB 126, dINT 207:

    Stone II
    Expected damage: TRUNC((1+126/100)*TRUNC(350 + 1*(207-100))) = 1032
    Observed damage: 1017
    Expected damage if M = 0 for dINT >= 200: TRUNC((1+126/100)*(350 + 100)) = 1017

    Thunder II
    Expected damage if M = 0 for dINT >= 200: TRUNC((1+126/100)*(375 + 100)) = 1073
    Observed damage: 1073

    2) Magic damage 205, MAB 190, dINT 249:

    Thunder II
    Expected damage if M = 0 for dINT >= 200: TRUNC((1+190/100)*(205 + 375 + 100)) = 1972
    Observed damage: 1972

    Stone III
    Predicted damage if M = 1 for dINT >= 200: (educated guess): TRUNC((1+190/100)*(205 + 550 + 200 + 1*(249-200))) = 2911
    Observed damage: 2911

    Thunder III
    Predicted damage if M = 1 for dINT >= 200: TRUNC((1+190/100)*(205 + 600 + 175 + 1*(249-200))) = 2984
    Observed damage: 2984

    Thunder IV
    Predicted damage if M = 1 for dINT >= 200: TRUNC((1+190/100)*TRUNC(205 + 955 + 275 + 1*(249-200)))) = 4303
    Observed damage: 4437 (M adjusted to 1.95 gives 4437)

    Stone IV
    Predicted damage if M = 1.95 for dINT >= 200: TRUNC((1+190/100)*TRUNC(205 + 850 + 300 + 1.95*(249-200)))) = 4205
    Observed damage: 4213 (M adjusted to 2 gives 4213)

    Thunder V
    Predicted damage if M = 2 for dINT >= 200: 5698
    Observed damage: 5831 (M adjusted to 2.95 gives 5831)

    Stone V
    Predicted damage if M = 2.95 for dINT >= 200: 5652
    Observed damage: 5660 (M adjusted to 3 gives 5660)

  15. #75
    Impossiblu
    Join Date
    Mar 2010
    Posts
    10,362
    BG Level
    9
    FFXIV Character
    Prothescar Centursa
    FFXIV Server
    Balmung
    FFXI Server
    Valefor

    Using Blazing Bound, Geo-Fend = +45MDB @ 781 skill. Comes out to around a 30% reduction in magic damage without any other MDB. Pretty substantial, surprised it hasn't been tested before.

  16. #76
    Old Odin
    Join Date
    Oct 2006
    Posts
    6,198
    BG Level
    8
    FFXI Server
    Asura

    Quote Originally Posted by Prothescar View Post
    Using Blazing Bound, Geo-Fend = +45MDB @ 781 skill. Comes out to around a 30% reduction in magic damage without any other MDB. Pretty substantial, surprised it hasn't been tested before.
    this is definitly nice to know for when hurkan hits the delve rounds... isnt one of his AoE attacks some kind of high magic dmg?

  17. #77
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    Makes me wonder how strong the -mab one is....

  18. #78
    Impossiblu
    Join Date
    Mar 2010
    Posts
    10,362
    BG Level
    9
    FFXIV Character
    Prothescar Centursa
    FFXIV Server
    Balmung
    FFXI Server
    Valefor

    -17 with 766 skill

  19. #79
    Sea Torques
    Join Date
    Dec 2007
    Posts
    582
    BG Level
    5

    Paralyze I rate and MND dependence

    Background



    TBH I don't remember doing any exploratory analysis with this data

    Code:
    Call:
    glm(formula = blah ~ X3, family = binomial(link = "identity"), 
        data = shite, weights = X2)
    
    Deviance Residuals: 
          1        2        3        4  
     0.1656   0.1488  -0.6711   0.5374  
    
    Coefficients:
                Estimate Std. Error z value Pr(>|z|)    
    (Intercept) 0.135564   0.009831  13.789  < 2e-16 ***
    X3          0.003606   0.001130   3.192  0.00141 ** 
    ---
    Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
    
    (Dispersion parameter for binomial family taken to be 1)
    
        Null deviance: 10.91747  on 3  degrees of freedom
    Residual deviance:  0.78865  on 2  degrees of freedom
    AIC: 31.468
    95% CI for intercept term (dMND = 0) is (.1163, .1548). 15% paralyze rate at dMND = 0 is consistent with the data.

    95% CI for dMND is (.0014, .0058). Paralyze rate increasing by +1% for every 2 MND (.005) or 4 MND (.0025) is consistent with this data.

    Also based on the above data, since Paralyze I seems to have a 25% rate cap, this might occur at dMND = 20 (.15 + .05*20) or dMND = 40 (.15 + .0025*40) given the above data.

    Level 62 Chigoes in Caedarva Mire seem to have 37 MND, while level 64 Chigoes have 35 MND (based on Banish I and Banish II damage back-calculations) so theoretically these can be used as test subjects to figure out how much dMND is necessary to cap paralyze rate.

  20. #80
    Relic Shield
    Join Date
    Jun 2005
    Posts
    1,906
    BG Level
    6
    FFXI Server
    Asura

    Geomancer's Cardinal Chant

    So, originally when SE announced Geomancer and its cardinal chant trait, the directions were mislabeled and 2 of the directions had unknown effects since they were not readily testable. Image here: http://i.imgur.com/DJBjT.jpg

    The supposed "shortens recast reduction" direction did not seem to exist whatsoever. However, I came across the chart on ffxiah in the Geo forum where a direction was labeled as Magic Burst Bonus which would explain why the effects of the direction were not apparent. I went out and did a simple test with Fiv in Kamihr Drifts using Raptors.

    Naked Tarutaru GEO/RDM w/ Ionis and Venabulum Staff.
    Stone 1 cast with mob positioned West of me did 340 damage.
    On Magic Burst, casting from the same position: 540 damage.
    Immediately ran to position the mob south of me and MB'd the same skillchain: 442 damage.

    442/340 = 1.3 which fits in line with the standard magic burst bonus.
    540/340 = 1.588 which clearly shows bonus MB damage.

    Assuming I have not made a mistake, this suggests that casting with the mob positioned west does indeed grant a magic burst bonus.

+ Reply to Thread
Page 4 of 15 FirstFirst ... 2 3 4 5 6 14 ... LastLast

Similar Threads

  1. Random Facts Thread: Abilities
    By Yugl in forum FFXI: Advanced Math
    Replies: 201
    Last Post: 2024-01-14, 15:22
  2. Random Facts Thread: Traits and Stats (Player and Monster)
    By Yugl in forum FFXI: Advanced Math
    Replies: 502
    Last Post: 2023-09-07, 20:41
  3. Random Facts Thread: Other
    By Yugl in forum FFXI: Advanced Math
    Replies: 490
    Last Post: 2022-07-06, 17:54
  4. Random Facts Thread
    By Yugl in forum FFXI: Advanced Math
    Replies: 157
    Last Post: 2011-12-08, 18:49