Magic Accuracy; Dagger With Skill Or Staff?

言語: JP EN DE FR
2010-06-21
New Items
users online
フォーラム » FFXI » Jobs » Bard » Magic accuracy; Dagger with skill or staff?
Magic accuracy; Dagger with skill or staff?
 Ragnarok.Imitole
Offline
サーバ: Ragnarok
Game: FFXI
user: Guniuni
Posts: 3
By Ragnarok.Imitole 2014-11-26 09:10:06  
I have been looking around to getting my bard back up to speed with gear and such. In the process I came across the question of which weapon to use for debuff landing. I currently own a 119 Mandau that has magic accuracy skill +215. I also used to have a Chatoyant staff.

Is the chatoyant staff outdated? Would a dagger with magic skill be more useful? I am not very good at understanding the mathy parts of the game. I tried reading the wiki on the subject beforehand, so I apologize in advance if this really is a stupid question. Thank you.
 Bismarck.Ihina
Offline
サーバ: Bismarck
Game: FFXI
user: Ihina
Posts: 3187
By Bismarck.Ihina 2014-11-26 09:20:26  
Staff, but not that one, Chatoyant is well outdated.

Mythic aside, I think the best ones are either Lehbrailg +2 with magic accuracy augment or Twebuliij with magic accuracy augment. Mandau is a ...decent option for now, until you get either of those.
[+]
 Siren.Sieha
Offline
サーバ: Siren
Game: FFXI
user: Sieha1
Posts: 503
By Siren.Sieha 2014-11-26 09:25:13  
magic accuracy skill ~= .9 magic accuracy. Same for wind, string and singing skill.

Hard content, 1 chr = 1 magic accuracy. weak content, its more like 2 chr = 1 magic accuracy.

Wind instruments have a higher chance to land debuffs then string but string has much higher range.

Maths not much hard just need to check the wiki for the break down.
[+]
 Siren.Bruno
Offline
サーバ: Siren
Game: FFXI
user: BruHouse
Posts: 398
By Siren.Bruno 2014-11-26 18:43:18  
Siren.Sieha said: »
magic accuracy skill ~= .9 magic accuracy. Same for wind, string and singing skill.

Source? As for the intruments/singing skill, BGWiki's Magic Acc page says "1 Skill = 1 Magic Accuracy", however the Song page does also say its unclear how it is exactly calculated. This has also been questioned in the past. but generally I think people tend to add it up as just 1:1, link suggests it may be more complex than that though.
[+]
 Asura.Zheta
Offline
サーバ: Asura
Game: FFXI
user: Zhizi
Posts: 5
By Asura.Zheta 2014-11-28 06:31:26  
Siren.Bruno said: »
Siren.Sieha said: »
magic accuracy skill ~= .9 magic accuracy. Same for wind, string and singing skill.

Source? As for the intruments/singing skill, BGWiki's Magic Acc page says "1 Skill = 1 Magic Accuracy", but the Song page does also say its unclear how it is exactly calculated. This has also been questioned in the past. but generally I think people tend to add it up as just 1:1, link suggests it may be more complex than that though.

I don't know the maths, but they're suggesting that 'magic accuracy skill' is not the same as 'magic accuracy' or 'singing/wind/string skill.'
 Sylph.Namonaki
Offline
サーバ: Sylph
Game: FFXI
user: Namonaki
Posts: 54
By Sylph.Namonaki 2014-11-28 07:09:26  
Grab yourself a Lehbralig(however its spelled) and augment it with some m-acc =]. I prefer staff over dagger for that m-acc grip.
 Lakshmi.Saevel
Offline
サーバ: Lakshmi
Game: FFXI
Posts: 2228
By Lakshmi.Saevel 2014-11-28 07:12:37  
Siren.Bruno said: »
Siren.Sieha said: »
magic accuracy skill ~= .9 magic accuracy. Same for wind, string and singing skill.

Source? As for the intruments/singing skill, BGWiki's Magic Acc page says "1 Skill = 1 Magic Accuracy", but the Song page does also say its unclear how it is exactly calculated. This has also been questioned in the past. but generally I think people tend to add it up as just 1:1, link suggests it may be more complex than that though.

They are confusing melee accuracy from skill and magic accuracy. SE implemented a sliding formula after 200 skill such that 1 skill = .9 melee accuracy. No evidence has ever been presented that magic accuracy works that way, especially since melee attack is still 1 skill = 1 attack. SE did some really weird stuff years ago with their math, level scaling Alpha on WS's being another example.
 Lakshmi.Byrth
VIP
Offline
サーバ: Lakshmi
Game: FFXI
user: Byrthnoth
Posts: 6137
By Lakshmi.Byrth 2014-11-28 08:21:11  
To answer the original question, the only really competitive MAcc dagger is Carnwenhan. If you aren't going to get that, you should augment a Lehbrialg +2.


Lehbrailg +2/Benthos Grip is capable of being superior even to Carnwenhan/Genbu's Shield for pure MAcc if you get lucky enough with augments (228 skill, 33 MAcc, and 11 CHR max vs. 228 skill, 31 MAcc). Carnwenhan has superior duration, of course, but it's technically 11 CHR and 2 MAcc behind in pure MAcc.
[+]
 Bismarck.Ihina
Offline
サーバ: Bismarck
Game: FFXI
user: Ihina
Posts: 3187
By Bismarck.Ihina 2014-11-28 08:53:06  
Siren.Bruno said: »
Siren.Sieha said: »
magic accuracy skill ~= .9 magic accuracy. Same for wind, string and singing skill.

Source? As for the intruments/singing skill, BGWiki's Magic Acc page says "1 Skill = 1 Magic Accuracy", but the Song page does also say its unclear how it is exactly calculated. This has also been questioned in the past. but generally I think people tend to add it up as just 1:1, link suggests it may be more complex than that though.

There is no source. This is just one of those things repeated enough times that people think is true.
 Asura.Sechs
Offline
サーバ: Asura
Game: FFXI
user: Akumasama
Posts: 9876
By Asura.Sechs 2014-11-28 10:28:42  
What Byrth said!
Carns also has AM1 though, if you can use it.
Saldy BRD atm has no good shield option, once a nice shield comes out Carns will be BiS once again for debuff (either way it's not THAT easy to get the perfect augment on Lehbrailg).
Also do not underestimate the additional duration, it's really nice to have.


Regarding the accuracy thing, I asked a similar question a few weeks ago.
Please do not forget that Sing is not the same as Wind/String.
The first is more or less the same as macc (roughly 1sing=1macc).
The other two have a much smaller influence (roughly 1wind/string=0.3macc).

Do not forget that when evaluating gear.
I dunno how much CHR converts to, but as long as your CHR is greatly below the target's CHR I think that 1 CHR should be worth 1 macc too.

In theory CHR should have a dynamic value. The further below your CHR is compared to the target's, the more chances to stick a debuff a single CHR point will add.
The closer you are the less it adds.
Eventually once you're past a certain threshold above the target's CHR, it should stop being relevant.
Not 100% sure about this because I have never read detailed tests about it, but it's what I was taught long ago and honestly it kinda makes sense to me.
 Ragnarok.Sekundes
Offline
サーバ: Ragnarok
Game: FFXI
user: Sekundes
Posts: 4189
By Ragnarok.Sekundes 2014-11-29 08:23:14  
I've heard the .3 thing for one of the other skills a few times. It does make sense but I would like a bit more solidity in my gearing. I've seen info on macc testing before and I don't think I'd be willing to do the tests but just how much testing would we need to do to figure it out, even only vaguely. What kinda of sample sizes and tests would be involved?

Do we even have the meva and stats of any mobs that would work for this test?
[+]
 Lakshmi.Byrth
VIP
Offline
サーバ: Lakshmi
Game: FFXI
user: Byrthnoth
Posts: 6137
By Lakshmi.Byrth 2014-11-29 09:15:34  
We don't, and I'm not even sure what resist states bard songs have. There are certainly mobs where songs regularly last half as long, but I don't think that's a resist. I can't swear that I've ever seen a half duration bard song on a mob where I sometimes get a full duration bard song, which might indicate that they only have full-resist and unresisted as options. However, with Troub on I never see resists at all. Either bard songs cap at higher than 95% magic hit rate, or our resists are not duration based or something?


If there's only two states, the test is simple. You go to Yorcia Weald and cast Requiem over and over on one of those things that doesn't attack you back (if they're susceptible). You want two cases minimum, and I'll list a third that you might want to do for thoroughness:
1) Uncapped magic hit rate. Ideally around a 30% land rate.
2) Add a lot of the stat. If your hypothesis is 0.3, you have a 30% land rate, and you want to be safe, you can add at least 234 skill safely.
3) Add the corresponding amount of magic accuracy. So lets say you went from 30% land rate to 75% land rate, you'd expect it to be 70 MAcc. Put it on and confirm that you have a 75% land rate.

Wash, rinse, repeat for any stat. You want to start with a low hit rate (but above 25%) and stack as much of the stat as you possibly can. The more you add, the fewer samples you need to get the ratio right. For this specific test, I'd recommend one of two approaches:
* Use an equipment set that has a large disparity between string and wind skill. For instance, use a random Harp with no notable stats and don't put on pieces that give only string/wind skill (Hafner Sash and Musical Earring). Something like this:
ItemSet 331150
If you have capped skills and merits in Wind and not in String, this could be a 106 skill differential.
* Other option would be to pick a base set that uses a good MAcc weapon and a high MAcc/stat build and not use an instrument at all. Then put on an instrument, take off the weapon, and see where you're at. You'd also have to figure out the value of "Magic Accuracy Skill" if you were going this route, though.


In terms of the pure sample size, if everything worked out well you would expect to need about 2100 casts in each condition to get +/- 2% confidence intervals, which would give you +/- 3% on your difference estimation.
[+]
 Ragnarok.Sekundes
Offline
サーバ: Ragnarok
Game: FFXI
user: Sekundes
Posts: 4189
By Ragnarok.Sekundes 2014-11-29 11:35:17  
K. Thanks as always for the detailed info. I may try to do a few tests but only if I can get it automated and have a parser gather the data. Then I can let it go all night and get a decent sample size.
 Ragnarok.Martel
Offline
サーバ: Ragnarok
Game: FFXI
Posts: 2894
By Ragnarok.Martel 2014-11-29 12:18:09  
Ragnarok.Sekundes said: »
K. Thanks as always for the detailed info. I may try to do a few tests but only if I can get it automated and have a parser gather the data. Then I can let it go all night and get a decent sample size.
Kparser is working, and does record enfeebles. For automation, I can set you up an enfeeble spam cycle pretty easily. Then if those field objects that aoe enffeble you(spiny protrusions, and those poison spamming mushrooms, etc.) have normal resistance properties then you can set it spamming on them all night.

Just be sure to document your test conditions well.
[+]
 Asura.Sechs
Offline
サーバ: Asura
Game: FFXI
user: Akumasama
Posts: 9876
By Asura.Sechs 2014-12-02 06:23:24  
Assuming that they balanced songs with the same "chance to stick on target" as other debuffs (Enfeebling Magic) there has to be a difference in the sing and wind/string conversion, compared to other skills conversion.

Enfeebling skill +1 roughly converts to macc+1, but jobs with Enfeebling Skill normally have a higher cap than Singing Skill is.
This is explained by the fact that the "chance to stick" formula for songs combines two different skills (sing + either string or wind).


Following this logic and considering the initial premise to be true (maybe it's not?) then it's obvious that song-related skills have got to have a different conversion and that it cannot be 1:1 like for enfeebling skill, because otherwise songs would end up with much more "macc" than enfeebling spells, and since this doesn't seem the case to me, it means the formula likely gives different conversion values to the 2 main skills that constitute the combined one for songs.

The two hypothesys we can think of are:
1) Both Sing and Wind/String have reduce macc conversion (lower than 1:1)
2) Sing is 1:1, but wind/string has a lower conversion (the hypothized 1:0,3)

The first sounds unlikely because I think there have been some tests for singing/macc, I remember reading something.
Which leads us to say that there's gotta be a different ratio for wind/string. (wether it's 0.3 or not, I don't know)


Given three items one with macc+20, one with sing+20 and one with wind+20, the first two should produce similar results (if not even identical) whereas the third should produce worse results.

Talking just in theory with no test or data to support, I can say I'm convinced by the logic of this theory.
 Shiva.Kingmancat
Offline
サーバ: Shiva
Game: FFXI
Posts: 332
By Shiva.Kingmancat 2014-12-02 07:55:41  
Asura.Sechs said: »
Saldy BRD atm has no good shield option, once a nice shield comes out Carns will be BiS once again for debuff (either way it's not THAT easy to get the perfect augment on Lehbrailg).

This has a been a pet peeve of mine for years!! The sad truth is, SE thinks BRD is supposed to sing, afk, sing, end fight. We are a total of 19 shields in the whole game. 16 of them are the "All Jobs" lvl 30 latent shields. The others are a really crappy lvl 40 one that BLM(?), SMN(??), and PUP(???) are also on, Legion Scutum (which is also All Jobs and isn't really all that helpful for BRD), and Genbu, the ONLY useful one since it can have Cure Pot, Cure Cast, and MAcc.

BRD gets left off of ALOT of useful support gear. Chanter's Shield that was just released, Sors Shield, and the support class clubs from Yumcax and Wokpet come to mind as glaring examples.
[+]
 Bismarck.Ihina
Offline
サーバ: Bismarck
Game: FFXI
user: Ihina
Posts: 3187
By Bismarck.Ihina 2014-12-02 08:13:19  
Augment +5macc onto your genbu shield. It's as good as any strap there is.
 Asura.Sechs
Offline
サーバ: Asura
Game: FFXI
user: Akumasama
Posts: 9876
By Asura.Sechs 2014-12-03 01:46:33  
Yeah Sadly even after 3,5 stacks of Genbu's things I could only manage to get Macc+4 :'(
Genbu's Shield is ok, but a new option with macc and whatnot wouldn't hurt.

Altough I guess unless we get another option with PDT, Genbu's Shield would still have a place at least in idle sets.
I don't remember but I think atm I might be using it even for my Curepot set on BRD.
Log in to post.