Item Search
     
BG-Wiki Search
+ Reply to Thread
Results 1 to 20 of 20

Thread: Aquaveil Testing (2014+)     submit to reddit submit to twitter

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

    Aquaveil Testing (2014+)

    This post isn't really complete, but I accidentally teleported myself and am done for now, so want to get the information recorded somewhere in case I close the text file I'm documenting things in.

    Aquaveil mechanics:
    First, I tried to make Aquaveil wear off on RDM without casting. It doesn't. I took 7000 damage (10 minutes of getting beaten on) and it didn't wear off. You need to be hit while casting to make Aquaveil wear off.

    Next, I switched to NIN/WHM and went out to find a level 100 Chapuli.
    Method:
    Code:
    * Cast Aquaveil.
    * Cast Teleport-Holla and move.
    * Keep track of how many times you're hit during Teleport-Holla until Aquaveil wears off and record it.
    * Repeat x20.
    1 hit - 12 times
    2 hits - 5 times
    3 hits - 3 times
    This works out to be a 40% chance of it wearing off every time I'm hit. I either didn't see a 4-hit case because I got unlucky or because it's capped at 2 extra hits.

    I changed job to SMN/WHM and repeated the test, this time with Vadose Rod, Shedir Seraweels, and Emphatikos Rope. I accidentally teleported myself after 10 attempts, but:
    4 hits - 5 times
    5 hits - 3 times
    6 hits - 2 times
    Approximately the same distribution, but shifted up by 3 hits.

    It is worth noting that in the above tests despite having subjob magic skill I did not have a 100% interruption rate. It is possible that the 1~3 range and 4~6 range are due to non-interruptions, and Aquaveil is just absorbing a static number of interruptions. This should become obvious with a larger sample size for the SMN case, because you wouldn't expect it to always be a 2-hit range as you got to longer durations of aquaveil.

    Next steps would be to:
    1) Complete the SMN test (another 10 trials)
    2) Switch to a job with 500 enhancing skill and see how that affects the distribution without enhancing items (WHM/SCH with Light Arts up casting Klimaform?).
    3) Add Aquaveil+ items and check for a cap.
    4) Maybe test a middle point (like BLM's enhancing skill) to get some idea about the progression.

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

    Also other things to test are approximate damage amounts/mob levels since that seems to make a difference (like going cleaving on blu/rdm and still cast spells with 20 mobs hitting you) though that could just be a case of less interrupts period and like you said it just absorbing so many that would've happened.

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

    Aquaveil Testing

    Introduction: So since the Aquaveil change a few years ago nobody has really went out to verify just how many 'charges' Aquaveil yields per cast. For those unfamiliar, Aquaveil completely prevents an interrupt, X number of times before wearing off.

    Current theory on Bgwiki:
    Makes player uninterruptable for X hits, where X is a function of enhancing skill. After X hits, the effect will wear off.
    It has been untested if you can sustain infinite number of interrupting hits per cast, consuming only 1 aquaveil 'Charge' per cast. Or if every interrupting hit can remove an aquaveil charge.

    >It's currently unknown how many 'charges' aquaveil bestows, or if it's based on Enhancing Magic skill.

    Procedure:
    Borrowing a model from Martel's earlier work on Foil, I used Ra'Dha Scarscute, an NM that allegedly doesn't miss (I learned that that is incorrect. While he has very, very high accuracy, he still misses quite often, at least 70% hit rate with ilvl gear, which is high for a pre-SoA enemy vs ilvl equipped players). I used SCH/SMN with 6 Summoning Magic skill. One assumption of this experiment is that Spell Interruption Rate is a function of player's "Relevant Magic Skill" (skill related to spell that is being cast) vs mob's level(unknown if it's simply mob level or their 'combat skill'). Assumption #2 is that I assume for simplicity that due to my low skill, it should only take 1 hit per cast to interrupt me.

    Step1: Define a baseline of how easily I am interrupted
    I engaged Ra'Dha Scarscute and attempted to cast Carbuncle X # of times to see how easily I am interruped. I do not have Spell Interruption meritted.

    Step 2: Aquaveil Testing
    With data gathered from Step 1, attempt to cast Carbuncle X # of times until Aquaveil wears off. Continue casting until a pattern of attempts is reached. This isn't very scientific but due to lag, it's possible one cast might be sufficiently fast enough to be between Ra'Dha Scarscute's melee attacks, preventing consumption of a 'Charge'. I casted as soon as his headbutt animation began. This was the perfect time to ensure his next headbutt would be about in the middle of my 50% cast and ensured only 1 headbutt per cast. I disregarded any trial where headbutt stunned me, or any trial where he used Skull Smash(he follows up skullsmash with an instant headbutt and confuses my count)

    Step 3: Skill based charges
    Attempt the above procedure, with Aquaveil at various enhancing skill anchor points and then with Shedir Crackows, and Emphatikos Rope.


    Results:
    -Ra'Dha Scarscute interrupted my Summoning casts 40/40 times with his headbutt auto-attacks. As a control, I used 40, 500 skill Cure III and none were interrupted.
    -With 355 Aquaveil up, one headbutt did not interrupt my Carbuncle casts, but a subsequent headbutt did. When timed so that he hits me twice per cast, upon the second cast hitting me, Aquaveil wore off. I did about 5 trials of each:
    - At 355 enhancing skill, Aquaveil prevents 2 interrupting hits.
    - At 378 enhancing skill, Aquaveil prevents 2 interrupting hits.
    - At 433 enhancing skill, Aquaveil prevents 2 interrupting hits.
    - At 458 enhancing skill, Aquaveil prevents 2 interrupting hits.
    - At 433 enhancing skill, Aquaveil prevents 2 interrupting hits.
    - At 506 enhancing skill, Aquaveil prevents 2 interrupting hits.
    - At 506 enhancing skill and Shedir Seraweels, Aquaveil prevents 3 interrupting hits.
    - At 506 enhancing skill and Emphatikos Rope, Aquaveil prevents 3 interrupting hits.
    - At 506 enhancing skill and Shedir Seraweels + Emphatikos Rope, Aquaveil prevents 4 interrupting hits.

    Conclusion - The data seem to show that Aquaveil imparts protection against 2 hits that would have interrupted your cast, regardless of Enhancing Magic skill. Equipment that enhances Aquaveil each add +1 additional charge. I don't have the club from Leviathan, but presumably it also adds +1. Aquaveil *SEEMS* more effective at higher levels though from my own personal and anticdotal observations. For example, when I get CP solo on SCH main with 450+ elemental magic, my tier3/4 nukes almost always get interrupted against VT/IT 123 mobs. With Aquaveil up, I seemingly can cast many more than 2 uninterruptable nukes before it wears off. There could be some kind of hidden funciton that enhances its effectiveness when the difference between your casting skill's level vs enemy level perhaps but this is a hassle to test.

    Carcharian Verve
    -Given that this Aquaveil lasts much, much longer than the White Magic version, and BLU enhancing spells usually are more potent, I decided to test this, why not?
    Repeated above as BLU/SMN, used Carcharian Verve casted at 500 Blue Magic skill.
    -Prevented 10 interrupting hits.
    -Did 5 trials
    -While I didn't bother testing, it may presumably be based on Blue Skill similar to Occultation imparting more Shadow Images, giving 10 at 500 skill.

    Notes:
    Occasionally while testing the White Magic version, sometimes Aquaveil would last once more than predicted. This was rare but it may have been due to
    -Poor timing on my part(if mob hits you within the first 20% of casting, or at 70% you typically do not get interrupted). Sometimes my timing may have been a half second off because the delay on his autoattacks is random, sometimes he does two back to back without delay, sometimes he does them a bit farther apart, but 90% of the time they're consistant and easy to time a Summoning Magic between one headbutt.
    -Perhaps Aquaveil, in addition to guaranteeing protection against X number of interruptable hits, perhaps also grants some "Spell Interruption Rate-X", whereby in addition to guaranteeing at least X hits are prevented from interrupting you, also gives an Enhancing Skill-dependent chance to save you from interruption without consuming a charge?

    Hard to tell honestly, it only happened like 2-3 out of about 100 trials I just discounted them as outliers, and only counted the minimum number of protection charges for each Tx group.

  4. #4
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    Byrth did some similar testing awhile back. Though iirc it was just counting how many hits till it wore off not interrupted but could just add 1 to that I suppose. He didn't make sure his spells would get interrupted so as such he had a small range but combined with yours it seems to indicate that only ones that would've otherwise interrupted will consume a charge. And rod was 1 more.

    As far as the effectiveness would be hard to judge without exact numbers with a decent sample size. Even a 90% interrupt rate could still result in a long lasting aquaveil

  5. #5
    Nidhogg
    Join Date
    Jul 2008
    Posts
    3,746
    BG Level
    7
    FFXIV Character
    Seraphus Highwynn
    FFXIV Server
    Gilgamesh
    FFXI Server
    Diabolos

    I just saw his post now(after I did my testing). In my tests, in about 110 casts of Summoning Magic with Aquaveil+Carcharian combined, the predicted number of hits always removed Aquaveil, with maybe 2 outliers which could be potentially due to lag, or perhaps a different mechanic all together.

    From my experience, when I casted Carbuncle and interrupted myself and then he hit me, it didn't count as an 'interrupting hit' because I guaranteed my interruption by moving. So like when testing Carcharian verve when I was on my 9th charge, if I casted Carbuncle and moved and he headbutt me, Aquaveil didn't wear. On my next cast when I casted Carbuncle and didn't move, Aquaveil was stripped as it was the 10th charge. I did Carcharian Verve testing 5 times, and all 5 lasted for exactly 10 squarely landed Headbutts without moving, ones where I moved before his headbutt was readied, the game didn't count as an interrupting hit.

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

    Weird cause his testing all of them he moved and he still got it to wear... maybe if it's only if you are hit while moving?

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

    Odd. I moved before the hit and never saw it wear but i'll retest it tomorrow. Interesting he got 4 hits with it wearing, i never got that to occur but i wouldnt rule it out like so a third-eye type mechanic. Something like first 2 casts are guaranteed and and any additional is a %. Could explain why i saw 2/110 casts last for 3 but idk, seems unlikely given 500+ skill and no sig diff between 400 and 500 skill.

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

    Did you check if +aquaveil stuff works on verve?

  9. #9
    Nidhogg
    Join Date
    Jul 2008
    Posts
    3,746
    BG Level
    7
    FFXIV Character
    Seraphus Highwynn
    FFXIV Server
    Gilgamesh
    FFXI Server
    Diabolos

    I didn't because it's a 5min check between verves. I would presume it does, doesn't +stoneskin stuff work on diamondhide? Pretty much no blu ever uses verve for aquaveil but maybe some will now if cleaving.

  10. #10
    Yoshi P
    Join Date
    Dec 2006
    Posts
    5,360
    BG Level
    8
    WoW Realm
    Arthas

    I use verve for the aquaveil all the time since it lasts for like 20 minutes. I have the pants and belt macroed in but I have no idea if they do anything.

    +Stoneskin doesn't work diamondhide or metallic body, unless they changed it at some point over the years.

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

    I can test it at some point this week. Didn't think many BLU's would have the inventory for it but I'll try it.

  12. #12
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    Quote Originally Posted by Ophannus View Post
    . Interesting he got 4 hits with it wearing, i never got that to occur but i wouldnt rule it out like so a third-eye type mechanic. Something like first 2 casts are guaranteed and and any additional is a %. Could explain why i saw 2/110 casts last for 3 but idk, seems unlikely given 500+ skill and no sig diff between 400 and 500 skill.
    Well given the huge difference in skill between your test and byrths it would make it look like those "extra" hits were due normal chances to not be interrupted. There is probably some floor on interruption so your 100% might actually be more like 95%. So a very small chance for you not to be interrupted could easily on a rare occasion get an extra and on a very rare occasion maybe even 2. But with higher skill the odds become more meaningful which might explain why it seems to work so good against high level stuff. Even something as low as say 40% chance to cast uninterrupted would increase your average number of hits before losing aqua

  13. #13
    Nidhogg
    Join Date
    Jul 2008
    Posts
    3,746
    BG Level
    7
    FFXIV Character
    Seraphus Highwynn
    FFXIV Server
    Gilgamesh
    FFXI Server
    Diabolos

    Wasn't he NIN/WHM casting teleports though? I'd imagine the difference between 6 summoning skill vs a level 75ish NM would be about the same as 140ish skill(assuming he was capped at /49 vs a 101 Chapuli?

    And you're right. There probably is a floor but lord knows I don't want to test that. There may even be other stupid factors that contribute to overall spell interruption rate, could be a dStat like dMND or dINT that gives a baseline or it could strictly be some function of casting skill vs attacker's level or casting skill vs attacker's attacking skill. Mobs presumably have a combat skill associated with their "weapon" and that's "skill" might be compared to our casting skill to create a spell interruption rate. If i recall, some NMs have been shown to be easily interruptable from straight melee attacks that otherwise shouldn't be, compared to comparable mobs of their level. I can't recall exactly which NM's but i do remember hearing of some NMs that cast that people say can easily be interrupted by a player of roughly the same or lower level and they hypothesized that this mob has low "skill" in that magic type or has the job as a subjob with gimped skill(note: most mobs get 2 main jobs like MNK99/WAR99, but they conjectured this mob's subjob was lower than its main, making it easier to interrupt than usual).

  14. #14
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    Well mainhand weapon might have helped with that not sure. And yeah your mob seems rather skilled at interrupting. Like I dual box avatars a lot lately and while it is high my thf/nin still can get utsesmi thru a hit often enough to be noticeable

  15. #15
    Old Merits
    Join Date
    Nov 2009
    Posts
    1,246
    BG Level
    6
    FFXI Server
    Shiva

    Quote Originally Posted by dasva View Post
    There is probably some floor on interruption so your 100% might actually be more like 95%.
    Pretty sure spell interrupt rate was tested to cap at 100%, you just need to slightly overgear to actually get 100%.

  16. #16
    Nidhogg
    Join Date
    Jul 2008
    Posts
    3,746
    BG Level
    7
    FFXIV Character
    Seraphus Highwynn
    FFXIV Server
    Gilgamesh
    FFXI Server
    Diabolos

    i was weaponless incase of any ilvl shenanigans, also my back was facing away

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

    Quote Originally Posted by Rooney View Post
    Pretty sure spell interrupt rate was tested to cap at 100%, you just need to slightly overgear to actually get 100%.
    Yeah I know there is a cap that goes all the way to 100% I was guessing that there might be a floor which I don't think has been tested

    Quote Originally Posted by Ophannus View Post
    i was weaponless incase of any ilvl shenanigans, also my back was facing away
    Yeah figured you probably were especially with that almost 0 rate but not sure about byrths test

  18. #18
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    I used Terra staff on nin/whm, so no iLevel there either. Preliminarily, it looks like there is at least some skill component that caps at 2 interruptions, because I got some single hit Aquaveils. If Ophannus' testing was super consistently 2 hits (and it looks like it was), then I'm willing to bet that Aquaveil is a static number of interruptions rather than some probabilistic thing stacked on top of the already-probabilistic "interrupted or not" check.

    So we can say that Aquaveil gives a static number of interruption preventions with some skill relationship:
    139 skill - 1
    140-354 - 1 or 2
    355-506 - 2 hits

    And that Aquaveil+1 gives +1 non-interruption to Aquaveil.

  19. #19
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    So the last piece of information we need is where the switch is between 1 and 2 interruptions. This will really only be relevant if it can be hit by /RDM or /SCH with enhancing gear, so as long as we can show it's above ~200 we probably wouldn't need to go any further.

  20. #20
    Nidhogg
    Join Date
    Jul 2008
    Posts
    3,746
    BG Level
    7
    FFXIV Character
    Seraphus Highwynn
    FFXIV Server
    Gilgamesh
    FFXI Server
    Diabolos

    I'm not objecting but why would this need its own thread vs leaving it in random magic testing? Not like aquaveil mechanics are gamechanging, though as a main blu, carcharian verve is quote potent(10min duration, amazing for blu/nin solos or when you need /nin for anything and might get interrupted).
    I'd also be curious if Embolden works with it but my RUN is 49. Also curious if multiple attacks per round work, might be faster to test on mandies since they hit twice per round and likely more accurate than Scarscute. It wasnt a problem on sch, but on blu i evaded too much and it took like 8min to get solid hits because if i removed ilvl gear to lower my evasion, then stun would proc on headbutt and that would be confounding.