Item Search
     
BG-Wiki Search
+ Reply to Thread
Page 2 of 2 FirstFirst 1 2
Results 21 to 36 of 36

Thread: Cursna Testing     submit to reddit submit to twitter

  1. #21
    Impossiblu
    Join Date
    Mar 2010
    Posts
    10,362
    BG Level
    9
    FFXIV Character
    Prothescar Centursa
    FFXIV Server
    Balmung
    FFXI Server
    Valefor

    What else would it be besides a percentage though? You're increasing a chance of an event occurring, not many other ways to express that than with a percentage. Magic accuracy check doesn't make much sense.

  2. #22
    First invited, last in the zone.
    Join Date
    Sep 2008
    Posts
    1,449
    BG Level
    6
    FFXI Server
    Lakshmi

    Quote Originally Posted by Prothescar View Post
    What else would it be besides a percentage though? You're increasing a chance of an event occurring, not many other ways to express that than with a percentage. Magic accuracy check doesn't make much sense.
    Well, supposedly healing magic skill affects it, and healing magic skill isn't a %. So that's at least one thing that isn't a % that (potentially at least) increases the chance of an event occurring. Maybe the WHM legs add "15" to some step in however that % is generated.

    What I meant by "Magic accuracy check" was not literally that it was Macc, or Macc would effect it, but rather that there was potentially a "Cursna accuracy" stat and a "Doom evasion" rating, and the chance of Cursna working was related to the two, with a floor and a ceiling. If that's true, then much like Magic Accuracy has no effect on spell land rate below the floor or above the ceiling, maybe "Enhances Cursna" doesn't either.

    I don't know, I'm just spitballing ideas here in the absence of testing.


    To be absoutely clear, I'm not saying it's *not* a %. I'm just offering other models that are (IMO) analogous to how some other things in the game work that might have explanatory power if a straightforward "Base% * multiplier" doesn't work.

  3. #23
    Old Merits
    Join Date
    Sep 2010
    Posts
    1,094
    BG Level
    6
    FFXI Server
    Ragnarok

    Doooomed! Or maybe not.

    Hmm, this thread certainly got interesting while I wasn't looking.

    Well, I was holding off on posting any of this till I got larger samples and a greater variety of tests. But it doesn't seem like I'm going to get around to it anytime soon. So I'll post what I've got.

    Tests
    I did 4 tests, But can't find any documentation of the first. Well, the sample size was tiny anyway. Cause I kept dying. -.-

    I think on the first test, I had most or all of the enhances cursna gear, but only about ~470 healing skill. And I just kept dying. So I tried pushing my skill up higher, and it appeared to have made quite a large difference. I actually started surviving most doom, and started getting a decent sample.

    So, actual data. Sadly, on this first test, I seem to have neglected to actually record which pieces of cursna gear I was using... I would assume some at least. Also dunno about received gear. :/

    Test1
    Skill 491
    Cursna gear:Unknown
    Received gear:Unknown
    Successes / total casts
    38/66 57.5 Rate.

    Test2
    Skill 500
    Cursna gear: Debilis, Mending, Haomax2, gende feet, Theo. legs +1, Hieros.
    Received gear: Saida x2
    Successes / total casts
    33/44 75.0%

    This test was just awesome. I could indefinitely survive in range of a doom aura. 3 casts was the most tries it ever took to remove doom. Delicious. And still room for improvement. Need HQ Received rings...

    I'm not really sure about the third test... I tend to think chance is messing with me.

    Test 3
    skill 500
    Cursna gear:Debilis, Mending, Haomax2, gende feet, Theo. legs +1, Hieros.
    Received Gear: None
    Successes / total casts
    25/52 48.0%

    I dunno. Less rate than my first test. And 27% less than the second test. With the only difference between test 2 and 3 being the saida rings. I'm inclined to think chance is screwing with me. Considering the sample size.

    Enhances Cursna speculation
    I think someone may have already said it in here, but I think the way cursna gear works is, multiplicative to your current success rate.

    So if you have a base rate of 50%, and add +50% in cursna gear, you get 75% rate. 50*1.5=75.

    I think this is why adding skill+ seemed to have a large impact. Because the success rate gain from adding skill+ was later multiplied by the cursna gear.

    Although, I must note that I really don't have the data to support that thought at all. Considering the small size of that first low skill test(and the subsequent loss of the records.)

    Usability and Usefulness

    Using a cursna set like this would cost a WHM 7 inv slots. Possibly a bit more/less depending on currently carried healing skill gear. And it'll cost the one being cast on 2 slots.

    Is it worth it? Hell yes! 75% doom removal is Powerful. And the Cursna specific pieces can always be satchel'd when fighting non-doom inflicting enemies.

    Going to Yumcax as PLD+WHM has actually been rather entertaining since putting together this cursna set. Everyone else nearby drops dead, but I'm goood. <,< Ard once managed to remove both Martel's and his own doom while weakened(and it wasn't doom that killed him in the first place.) o.o

    Further testing.

    Unfortunately, with the small samples and types of tests, none of this actually proves much. Much less accurately quantifies anything. At best it gives a basic idea of high end cursna potential. Which is not a bad thing at all, but I'd certainly prefer greater accuracy, and understanding of the mechanics at work here. Which means more testing.

    Now that I can reliably survive doom, my biggest test issue is easy continuous access to doom. My current test mob is Tangaroa, for the doom aura it has when it goes into it's shell. With a doom aura, you can constantly remove doom until the aura wears.

    The issue is, that it only has the aura sometimes when it turtles up. I've gone whole fights just getting poisoned and not seeing doom. It's not very time efficient.

    Is there a dooom aura mob that uses doom more regularly? Or for longer periods? Or fulltime? Fulltime would be good. <,<;

  4. #24
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    Quote Originally Posted by Foldypaws View Post
    Well, supposedly healing magic skill affects it, and healing magic skill isn't a %. So that's at least one thing that isn't a % that (potentially at least) increases the chance of an event occurring. Maybe the WHM legs add "15" to some step in however that % is generated.

    To be absoutely clear, I'm not saying it's *not* a %. I'm just offering other models that are (IMO) analogous to how some other things in the game work that might have explanatory power if a straightforward "Base% * multiplier" doesn't work.
    Well if you want to use another fairly similar analogous thing there is tranquil heart. It is based on healing skill but other -emnity is still percentage

  5. #25
    Impossiblu
    Join Date
    Mar 2010
    Posts
    10,362
    BG Level
    9
    FFXIV Character
    Prothescar Centursa
    FFXIV Server
    Balmung
    FFXI Server
    Valefor

    Cursna data

    Tests conducted in Ballista (BLU vs WHM) using Mortal Ray.




    Test 1
    424 + 15 healing magic skill, no potency gear

    Mortal Ray casts: 18
    Successful Cursna casts: 14/81 (+/- 17%)



    Test 2
    424 + 16 healing magic skill, 55% potency (Haoma's ring x2 30%, Theophany pantaloons 15%, Malison medallion 10%)

    Mortal Ray casts: 26
    Successful Cursna casts: 26/80 (+/- 32%)



    Test 3
    424 + 16 healing magic skill, 55% potency (Haoma's ring x2 30%, Theophany pantaloons 15%, Malison medallion 10%) + Unknown % from Gendewitha galoshes and Mending cape

    Mortal Ray casts: 27
    Successful Cursna casts: 27/65 (+/- 41%)






    Need to do a ton more tests and expand sample sizes on these ones, but things I'm looking at as possibilities atm:

    1) Mending Cape or Gendewitha Galoshes having a larger Cursna potency bonus than other pieces. Would gravitate toward the cape having extra.
    2) Healing magic skill works in tiers, and 440 just happened to break a tier in these tests.
    3) Doom is based on stats somehow; INT, MND, or CHR compared between target and caster that helps to determine the base Doom removal chance. Could work for all Dooms, could work for but the BLU one. Hard to say.

    Need a lot more data, but those are some possibilities that I'm tossing around in my head at the moment. All this provides so far is statistical evidence that Cursna potency gear actually does something.

  6. #26
    Old Merits
    Join Date
    Sep 2010
    Posts
    1,094
    BG Level
    6
    FFXI Server
    Ragnarok

    Well, that's certainly a controlled, repeatable test method. Although the ballista time limit could be a bit annoying
    Quote Originally Posted by Prothescar View Post
    3) Doom is based on stats somehow; INT, MND, or CHR compared between target and caster that helps to determine the base Doom removal chance. Could work for all Dooms, could work for but the BLU one. Hard to say.
    It seems odd to me to use a dSTAT comparison for a player targeted support spell. If it worked like most dSTATs, it'd effectively be penalizing the target for having high stats.

    Now this isn't to say that stats can't be involved. But I'd think they'd set it up more like waltz formulas, where highers stats(vit) on the target end improve the results.

  7. #27
    Impossiblu
    Join Date
    Mar 2010
    Posts
    10,362
    BG Level
    9
    FFXIV Character
    Prothescar Centursa
    FFXIV Server
    Balmung
    FFXI Server
    Valefor

    Yeah that's what I mean, it works on some sort of imaginary slider where, if the caster's INT/MND/CHR/whatever is lower than the target's, then Doom has a higher chance of being removed (and vice versa). It would be a ridiculous thing to have to test for (as would healing magic tiers), and I really hope that I don't have to. With luck it's just the cape being amazing.

  8. #28
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    Could it just be sample sizes? Looks like you have some overlapping confidence intervals

  9. #29
    Impossiblu
    Join Date
    Mar 2010
    Posts
    10,362
    BG Level
    9
    FFXIV Character
    Prothescar Centursa
    FFXIV Server
    Balmung
    FFXI Server
    Valefor

    Yes, as I said I'm not finished testing yet and those three items are just theories for what could affect Doom and Cursna. Any of the three could be true at this point, won't know until I do more tests (or someone else does). All my current data says is that there's an enormous disparity between varying levels of cursna potency gear, and potentially between INT/MND/CHR differences (taking off gendewitha boots) compared to the base cursna doom removal rate.

  10. #30
    Ridill
    Join Date
    Apr 2011
    Posts
    23,736
    BG Level
    10
    FFXI Server
    Bahamut

    Shame Mortal ray is 91... 1 level lower and you could check stats easy in monstrosity. Shame ballista still requires points to do your own or 6 to do a regular one. Makes testing so much harder

  11. #31
    Impossiblu
    Join Date
    Mar 2010
    Posts
    10,362
    BG Level
    9
    FFXIV Character
    Prothescar Centursa
    FFXIV Server
    Balmung
    FFXI Server
    Valefor

    I've got a metric fuckton of points, not really prohibitive for me. Biggest obstacle is building up the desire to bother finishing this since it's so tedious.

  12. #32
    Impossiblu
    Join Date
    Mar 2010
    Posts
    10,362
    BG Level
    9
    FFXIV Character
    Prothescar Centursa
    FFXIV Server
    Balmung
    FFXI Server
    Valefor

    Cursna data part deux

    Tests conducted in Ballista (BLU vs WHM) using Mortal Ray.




    Test 1
    424 + 15 healing magic skill, no potency gear

    Mortal Ray casts: 30
    Successful Cursna casts: 26/116 (+/- 22%)



    Test 2
    424 + 16 healing magic skill, 55% potency (Haoma's ring x2 30%, Theophany pantaloons 15%, Malison medallion 10%)

    Mortal Ray casts: 26
    Successful Cursna casts: 26/80 (+/- 33%)



    Test 3
    424 + 16 healing magic skill, 55% potency (Haoma's ring x2 30%, Theophany pantaloons 15%, Malison medallion 10%) + Unknown % from Gendewitha galoshes and Mending cape

    Mortal Ray casts: 27
    Successful Cursna casts: 27/65 (+/- 42%)



    Test 4
    424 + 68 healing magic skill, no potency gear

    Mortal Ray casts: 32
    Successful Cursna casts: 28/115 (+/- 24%)



    Seems skill makes very, very little difference in terms of overall success rate. This does, however, further refine the base doom removal rate and makes the third test look a little less derpy; cape+feet together being 30% most likely, meaning 15% each which falls into expected values. Overall an 85% or so boost from all that gear, which also falls in line. Pretty obvious that gear doesn't cap at 45% and that it is likely just multiplicative with base cursna doom removal rate, which is determined by healing magic skill and probably some base floor. I haven't got enough gear to determine whether or not cursna removal rate itself caps at 45% or not, nor the patience to get a sample size high enough to say for certain. However, this data is sufficient to draw basic conclusions on how cursna, and cursna enhancing gear, works. Only other test I plan to do in the immediate future is /WHM to try to get more of a feel for how skill affects removal rate; other than that, just expanding sample sizes when I can be bothered.

  13. #33
    Old Merits
    Join Date
    Sep 2010
    Posts
    1,094
    BG Level
    6
    FFXI Server
    Ragnarok

    So, looking at this, my 75% was probably a 60% that was hitting high within it's range.

    +85% cursna gear plus hieros for another +10%. =95% cursna gear
    then saidax2 for +30% cursna received.
    using 24% base removal. Although since i had 500 skill, it could have been higher.

    Assuming cursna received is multiplicative in a different stage than normal cursna gear(cure potency/received-esque.)

    24*1.95=46.8
    46.8*1.3=60.84% removal rate.

    With my piddly 33/44 sample, 60.84% is juuust barely within the low end of the range. Although, if my base had been 25% from having 500 skill(a very neat, logical number) then my final removal rate would be 63.37%. Which doesn't swing quite so close to the edge of the range.

    It's disappointing to think the the actually values are lower than I thought. But it makes good sense, and deals with the differences between our tests.

    My test 3 matches up rather neatly too. If I had 25% base with +95% cursna gear, 48.75% rate. Test 3 result: 48.0%.

    With all this in mind..... I need my second eshmun's ring and shabti legs +1! lol.

    500 skill
    +95% cursna
    +55% cursna received
    25% base

    25*1.95=48.75
    48.75*1.55=75.56%

    So 75% doom removal rate should be the current upper limit assuming perfect caster and target sets.

  14. #34
    Relic Horn
    Join Date
    Aug 2008
    Posts
    3,131
    BG Level
    7
    FFXI Server
    Sylph

    Didn't SE say that base Cursna chance caps at 30% st 500 skill?

    Also, you forgot 121 Gambanteinn for +100% to Cursna. The BG wiki cursna page lists enough gear for Cursna +110 (+210 with Gambanteinn) and enough to get Cursna Received to +65, which, if it's possible to get 500 skill with it all on, should give:

    30% * 2.1/3.1 * 1.65 = 103.95%/133.95%

  15. #35
    BG Content
    Join Date
    Jul 2007
    Posts
    21,104
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    Well, his post was from February of 2014 so 121 Gamb wasn't released yet. Now it should be possible to get a 100% removal rate, however.

  16. #36
    Relic Horn
    Join Date
    Aug 2008
    Posts
    3,131
    BG Level
    7
    FFXI Server
    Sylph

    Oh, whoops; I saw the recently updated graphic on the topic, but missed Songen's post and thought Mariel's was the recent one.