What's Better?

[Fwahm]

no, level correction is applied first.

crato caps at 2.0 on even matches.

on level 82 it caps at 2.0-7*0.05=1.65

1.65*350=577 <-- actual attack cap.

Where's the actual tests on this?

It seems quite silly, as it means you need less attack to cap on level 83 Greater Colibri than you do for level 82 Greater Colibri, even if your damage is lower on the 83s.

[Hitoseijuro]

actually it's the converse. recent pDIF research proved that attack caps much lower than expected, because c ratio is supposed to cap before level correction. In practice it means than using more than 500 attack during asuran fists is useless ( even less if the mob is dia3'd). So the only remaining way to improve it is play on str and vit. So you just try to reach a "tier" of str+vit, which are 150, 160, 180, 190, 200, etc ( 170 isnt a tier due to roundings).

Right, but its still dependant on what buffs/debuffs(some rdms dont dia, inorite) he has. Outside of /war and cor he wont be getting 500att or above with just str gear(485-ish with kaboob, relatively close, though theres boost too which is probably what im missing on that ;. And as he meantioned he does /war I did say it might not apply to him as much.

Theres no reason not to have multiple macros for different camps and different buffs. Saying something as an absolute isnt always the right way to go about things. I normally have a couple ]att/acc macros depending on buffs/camp and str/att/acc macros depending on buff/camps.

Also 170 is a tier, 180 is not a tier(maybe you were confusing that....)
160*.10 =16 16*.83 = 13
170*.10 = 17 17*.83= 14
180*.10= 18 18*.83 = 14

[Howard Roark]

since this is pretty much getting ignored in the magian thread, lay out the h2h weapon hierarchy for me

is it hades+1 > destroyers > Mantis > the dex4 acc8 cat claws?

where does the occasionally attacks twice (assuming joyeuse proc rates of attacking twice, 45%) fit in that hierarchy

feel free to toss in PW weapons and pre and post upgrades of spharai/glanzfaust for completion's sake so people can stop asking
repost:

I might have completely missed it, but has anyone done any number crunching to see if the Dmg +21 delay +55 Mantis weapons would beat destroyers/hades +1? might think of working towards them if they do.

this wasn't answered, bump

currently working on this path, wondering what to upgrade besides the occasionally attacks twice and possibly those dex4 acc8 ones (path for this btw? was unclear in francisco's post at the time)

[mdkuser]

since this is pretty much getting ignored in the magian thread, lay out the h2h weapon hierarchy for me

is it hades+1 > destroyers > Mantis > the dex4 acc8 cat claws?

If you ignore the +9 accuracy on hades+1, destroyers are ~1.2% better. For hades+1 to beat it, you need to be able to swap those 9 acc in another equipement slot by something that improves your damage more than 1.2%; it's easy, just swap your torreador for a str ring if you are not str caped, and if you are str caped do hollow earring > mermans or black Tath. > white tath. Basically 1 base dmg doesnt beat 6% crits, but 2 base damage do.

Mantis ( I assume the D21 delay 55 one?), is equialent do destroyers. Well there are many ways to do the maths, but my maths find destroyers is +0.2% better.. which is neglectible.


cat claw I assume DMG+18 delay+61 dex+4 acc+8?

With this delay you are getting 0.1 more tp/hit than destroyers or HS+1 btw, which changes WS frequency.

If you ignore those +10 accuracy, detroyers are 6% better, and NQ hades sainti are 5.2% better. Are you gonna be able to swap 10 accuracy slots for something better than +6% damage? nope: they are wors than destroyers. Are you gonna be able to swap 3 accuracy slots for 5.1% damage? nope: they are worse than hades sainti NQ.

remark: the maths were done for mnk/nin not holding their TP, with caped fSTR during melee or WS, str+vit=200 during WS, pdif=1.3, pdfif_crit=2.3 and asuran_pDIF=1.6 and double march (basically no minuets or chaos's). I can provide details later if needed... all is on my spread sheet.

[pchan]

since this is pretty much getting ignored in the magian thread, lay out the h2h weapon hierarchy for me

is it hades+1 > destroyers > Mantis > the dex4 acc8 cat claws?

where does the occasionally attacks twice (assuming joyeuse proc rates of attacking twice, 45%) fit in that hierarchy

feel free to toss in PW weapons and pre and post upgrades of spharai/glanzfaust for completion's sake so people can stop asking
repost:



this wasn't answered, bump

currently working on this path, wondering what to upgrade besides the occasionally attacks twice and possibly those dex4 acc8 ones (path for this btw? was unclear in francisco's post at the time)

My own math :

*Mantis (D21 delay 55) is ~1% better than Hades Sainti +1 without the +9acc, so HS+1 will almost always beat it.
*destroyers are 1.2% better than HS+1 without the +9 acc so HS+1 will almost always win
* destroyers are 0.2% better than mantis...
* destroyers are 4% better than cat's claws *IF* you don't need the acc (10 acctotal). So If you use those acc fully they will beat destroyers.
* HS+1 are 2.8% better than cat's claws *if* acc on both weapon is not accounted for, so in all cases it doesn't beat it.


So general comparison :

acc needed
HS+1 > cat's claw > destro > mantis
Part of the acc is traded for other stats on other slots (merits typically)
HS+1>destro>mantis>cat's claws
acc not needed at all
destroyers>mantis>HS+1>cat's claw

If you ever manage to do a decent crit build cat's claw is probably winning, good luck with that, it's probably done with byakko/usu body and sushi, so inferior to full usu.

Also stat wise, shenlong's are superior to all. And mythic and relic are gaining free damage (5 and 6 I think) and a fstr so nothing can beat them now (mythic used to suck, not now, good news..)

[Howard Roark]

thanks for replies, mildly depressing to still need a pair of HS+1s lol

at what point does MNK cap accuracy (situational I know) so let's say bird camp versus north mamool camp for 2 sample situations

assume full usukane 8/8 h2h, what else is needed? I'm guessing you'll assume pizza for mamools based on your previous posts but I'm wondering which rings/neck/back/ammo to aim for

[xerodok]

Odin+7 ATK/Mermans Earring or Odin+Brutal for MNK TP.

[Raen Ryong]

5% DA >>> 6 Attack. Or am I missing something? Seems too easy...

[Cream Soda]

5% DA >>> 6 Attack. Or am I missing something? Seems too easy...

[Hitoseijuro]

thanks for replies, mildly depressing to still need a pair of HS+1s lol

at what point does MNK cap accuracy (situational I know) so let's say bird camp versus north mamool camp for 2 sample situations

assume full usukane 8/8 h2h, what else is needed? I'm guessing you'll assume pizza for mamools based on your previous posts but I'm wondering which rings/neck/back/ammo to aim for

407acc is cap for birds at lv 82. Obviously less for 81.

Full Usukane mnk with 8/8 H2H merits with 71 dex(Hume) will net you 360 acc roughly. Which means you have about 47 acc to make up in the rest of the slots when Focus is down and about 17 acc when its up.


For mamools the range is a bit different, *generally* they will range around 381-430acc to cap out on(varies between type/level), with nin(450s) and(thf495+) being higher.

With the same setup Hume, full Usukane 8/8 H2H merits with pizza+1 , it will put you at 400acc. So for the other slots you only need about ~30acc for things like drg/blu. Can save focus for thf/nin types if you like or w/e you feel is best.(if no focus is up for nin types, it would be around 50acc needed,30-40 is still within decent hitrate I guess...).

Hope that helps.

[Norellicus]

Was this changed more recently than the 2007 fix? Because I'm pretty sure that made your att:def cap apply before anything else was done, then level correct is applied, then pdif.

I mean, it makes logical sense to level correct first, because it creates situations where you cannot feasibly attack the monster for any damage, no matter how many buffs you have applied to yourself, but seems contradictory to historical information

Where's the actual tests on this?

It seems quite silly, as it means you need less attack to cap on level 83 Greater Colibri than you do for level 82 Greater Colibri, even if your damage is lower on the 83s.

Quoting these again because I'd still like to see more info; there's a significant difference in the result depending on how you determine the order of these functions. Consider:

Level correct maximum ratio for 82: 2-0.05*7=1.65
Ratio: 577/350=1.65 (no change)
pdif: 1.18-1.98
Output w/ destroyers and capped fstr: 74~130, asurans 739~1303

vs

Ratio: 577/350=1.65
Level correct final ratio for 82: 1.65-0.05*7=1.3
pdif: 1-1.56
Output w/ destroyers and capped fstr: 63~103, asurans 626~1026

That's an astounding 23% difference in maximum output just by swapping the order of the corrective functions, and the result for level correcting first seems really high for birds, but I've merited on them all of twice ever and it was more than a year ago so I may just be in shock at how easy they are to grind into a pulp.

So if I'm misunderstanding you or there's some element here I'm missing, I'm all ears; I love doing the math here but want to make sure I'm doing it right

Edit: Plus, cRatio seems to logically stand for corrected ratio, no? Wouldn't that imply adjusting for level after finding the actual att:def value?

[mdkuser]

Quoting these again because I'd still like to see more info; there's a significant difference in the result depending on how you determine the order of these functions. Consider:

Level correct maximum ratio for 82: 2-0.05*7=1.65
Ratio: 577/350=1.65 (no change)
pdif: 1.18-1.98
Output w/ destroyers and capped fstr: 74~130, asurans 739~1303

vs

Ratio: 577/350=1.65
Level correct final ratio for 82: 1.65-0.05*7=1.3
pdif: 1-1.56
Output w/ destroyers and capped fstr: 63~103, asurans 626~1026

That's an astounding 23% difference in maximum output just by swapping the order of the corrective functions, and the result for level correcting first seems really high for birds, but I've merited on them all of twice ever and it was more than a year ago so I may just be in shock at how easy they are to grind into a pulp.

So if I'm misunderstanding you or there's some element here I'm missing, I'm all ears; I love doing the math here but want to make sure I'm doing it right

Edit: Plus, cRatio seems to logically stand for corrected ratio, no? Wouldn't that imply adjusting for level after finding the actual att:def value?

actually, in your 2nd scenario, nothing prevents you from having more than 600 attack. look, with just pizza, standard attack gear ( kotes, togi, usuleg, feet) and boost, I'm already at 600+ attack ( ~ 500 pre boost). so in this case, once you have minuets and/or dia2 you will have like 650, up to 700 with chaos's, which is caped attack since more than double colibris's or MMj's DEF ( they have 330 ish def actually..). So again, in your second case scenario,

Ratio=attack/def=caped at 2.0 with minuets, or without minuet but with dia2 ( dont even need dia 3...)

using the data provided by masamune...

http://www.bluegartr.com/forum/showthread.php?t=87774

associated pDIF varies between 1.55 and 2.5

so a 8 hit asuran fists ( ftp 8.1, destroyers, caped str and 15 WSC)
damage varies between 979 and 1579

I have never seen 1600ish asuran fists, and I guarrantee you I have done under 979 8-hit asurans ( the average asuran is lower lol).


In the first case scenario, cRatio is caped at 1.65, the corresponding pDIF varies
in the range 1.15~2.1 which gives a range for asuran fists of 726~1326 which is more accurate with what i remember lol ( average around 900 dmg).

I think these considerations are enough to claim that level correction is applied first. Unfortunately I dont have colibri parses anymore, but I remember breaking 1k damage occasionnally, getting as high as 1100-1200 maybe but never more. and that's with minuets because back then brds sing march/minuet. Also, I heard about this cap beeing applied first here on BG from hearsay, on various threads, and the wiki on ffxiclopedia has been updated around christmas with this info.

[Norellicus]

I think we're still missing eachother here, because I can't fathom how you're getting a 2.5 pdif...if Ratio caps at 2.0, and then you apply a delta-7 level correction, you get 1.65 cRatio, which results in the same 1.18-1.98 pdif distribution. That's the only reason I didn't simulate 700+ attack, because if case 2 is true and as we've already established we can easily hit 700 attack against birds regularly then we should see the results posted in case 1, which doesn't seem to be reality (but again I may be severely underestimating the expected returns since I don't have personal experience to back me up).

To reiterate, I'm not saying there isn't a cap, I'm saying that Ratio, i.e. your attack/target defense, is found to be an integer value not greater than 2.0 with any excess being ignored, thus making 2.0 the effective "cap"; then level correction is applied on top of that, which attains your cRatio.

It seems to me that if you level correct first then you're trivializing the effect of fighting monsters far stronger than you, creating a forced statistical barrier that simply states "must be at least this level to attack this mob in any fashion" and neuters any possible effect of gear+buffs beyond this statistical point. One could argue that the other variables involved (accuracy, fstr) can create similar scenarios, but you can overcome those simply by buffing those individual stats further (though this is hardly economic given the way our equipment is budgeted and the requirement to have a healthy amount of all three). I've seen cases of people at least being able to passably attack something that they'd usually have no business attacking thanks to just having sheer quantities of buffs to get their stats up which is why I have to question level correct applying first (but admittedly all cases I can recall are no more than ~20 levels of disparity, so that simply may not be enough to invoke the proverbial point of no return).

[mdkuser]

Level correct maximum ratio for 82: 2-0.05*7=1.65
Ratio: 577/350=1.65 (no change)
pdif: 1.18-1.98
Output w/ destroyers and capped fstr: 74~130, asurans 739~1303

vs

Ratio: 577/350=1.65
Level correct final ratio for 82: 1.65-0.05*7=1.3
pdif: 1-1.56
Output w/ destroyers and capped fstr: 63~103, asurans 626~1026

the latest pDIF formulas give a range pDIF=1.15~2.1 for cRatio=1.65,
and pDIF=1~1.68 for cRatio=1.3, by the way. Now I understand what you wanted to say. Well it IS possible to get more than 1100 or 1200 asurans with about 570 attack, its just rare since all your 8 hits need to reach the highest value. So you'll always be stuck in the ~900 range.

[Norellicus]

I was using the wiki formulas for lack of a better option

Are these new ones documented somewhere? I don't see anything definitive in that Masamune thread you linked earlier, just the speculation on the relative caps. I can't really extrapolate much from the graph as I don't understand all the values represented (I know how to use the formulas, but I'm not very good with reverse engineering results to get back to the formulaic source).

Edit: Nevermind, I guess I was looking at them backwards or something, I see the relationships now and can at least use the graph to find values, even if I can't write out the actual formula.

[Staleyx]

Hagun: DMG:75 Delay:450 TP Bonus
or
Kurodachi: DMG:40+28 Delay:464 Occasionally attacks twice.
or
Radennotachi: DMG:79+10 Delay: 450

[Nightfyre]

If it procs like Joyeuse, Kurodachi destroys the other two. Hagun > Radennotachi

[Shinryuu]

What's better? Well, not really 'what's better', but what would the distribution of occasionally attacks twice on Luchtain have to be in order for it to beat Bravura with maximum augments? Assume 6-hit for both, capped accuracy for both.

[Bikpik]

Is aiming for a 5 hit feasible for DRK/SAM without SAM roll, or does it sacrifice too much haste/accuracy for the store TP?

With at 528 delay scythe, and standard weaponskill gear, it takes at least 25 Store TP on gear to get a reliable 5 hit going for DRK/SAM.

It actually only take 23 stp in gear to hit a true 5hit drk/sam, but you need to use stp food for this, and I found the ACC a bit lacking on anything worth while. (unless I'm missing a piece of gear somewhere)

[Shinryuu]

Can't use pizza and 5-hit DRK so it's going to depend on your hitrate. Shame Apocalypse isn't 528, would be so good.