Forum Settings
       
« Previous 1 2 3
Reply To Thread

Maximizing SA/TA Damage (mathy mathy)Follow

#1 Jul 30 2009 at 4:30 AM Rating: Excellent
Scholar
***
1,983 posts
This topic has come up on the boards in the past, about which stats to use for sa/ta. Obviously we've known the wiki basics for awhile: ie, "a single point of dex/agi increases base damage by one. the strike will be a critical which will not miss if landed correctly." Etc, etc. But being the fiends we are, many of us here would like to know "just how much better is a point of dex/agi for SA/TA than a point of attack?" or "is a point of haste ever better for SA/TA than a point of dex/agi?" and "if so, how much better is a point of haste; if not, how much worse is it?"

The veterans here, I'm sure, can easily understand that any answer to this question will not hold for all situations. The game is far too situational. Various factors such as player stats and equip, buffs, debuffs, and mob fought are all interdependent. However, it is still possible and beneficial to speak generally so long as we understand that general =/= always.

I'll post here a very abridged summary version, but for those that would like to know a lot more, the article in my blog can be found here: How to Maximize Sa/Ta Damage. So for detailed reading, I'd go there instead. Posting it all here would be 'tldr'.

I've come up with the following ratios for the relative values of various stats that can affect Sa/Ta damage (/nin).

Sneak Attack
1Dbl.Atk = 1
1TripleAtk = 2.38 (d.a.)
1Dex = 2.58 (d.a.)
1Str = 1.20 (d.a.)
1Attack = 1.19 (d.a.)
1Slow(2xMarch,Haste) = -2.64 (d.a.)
1D.Wield(2xMarch,Haste) = 2.16 (d.a.)
1Haste(2xMarch,Haste) = 2.65 (d.a.)
1Haste(1xMarch,Haste) = 2.22 (d.a.)
1Haste(NoMarch,Haste) = 1.84 (d.a.)

Trick Attack
1Dbl.Atk = 1
1TripleAtk = 2.51
1Agi = 3.35
1Dex = 0.39
1Str = 1.29
1Attack = 1.26
1Slow(2xMarch,Haste) = -3.50
1D.Wield(2xMarch,Haste) = 2.59
1Haste(2xMarch,Haste) = 3.50
1Haste(1xMarch,Haste) = 2.89
1Haste(NoMarch,Haste) = 2.43

These values are all relative to how much a single point of double attack increases damage.

I've also used this to rank SA/TA gear accordingly,

Sneak Attack
(head) Maat's Cap (26.46) > Skadi's Visor (25.41) > Gnadbhod's Helm (13.84) > AF2 (12.9) = Turban (13.25) > Emp.Pin (7.74) > Heca+1 (0.84)
(neck) Love Torque (20.04) > Kubira Bead Necklace (15.08) > Spike Necklace (11.34)
(ears) Suppanomimi (10.8) > Pixie (7.74) => Merman's (7.14) > Hollow/Adroit (5.16) = Coral (5.95) = Brutal (5.0)
(body) Dragon Harness+1 (32.34) > Skadi (26.59) = DH nq (27.38) > Anatares Harness (20.64) > BlueCote+1 (16.32) > WarAketon +1 (12.9) > Rapp (10.6)
(hand) ANNM Dragon Mitts (20.44) > Skadi Bazubands (17.9) > Dusk+1(2xMarch) (17.74)* > Enkidu's (15.12) => Tarasque Mitts +1 (14.28) => Dusk(2xMarch) (13.9)* > Merman's (8.94) > AF (7.74) > Heca +1 (4.02)
(rings) Rajas Ring (18.9) > 5Dex Ring (12.9)
(back) Charger's (23.8) => Cerberus Mantle +1 (22.65) > Forager's (21.45) > Amemet +1 (20.25) > Cuchulain's (15.12) > Commander's (11.34) > AF2 (10.32)
(waist) Cuchulain's Belt (27.38) > Ninurta's (23.04) > Warwolf (18.9) > Speed Belt (15.9) > Swordbelt +1 (14.28)
(leg)Enkidu's (18.9) > Dusk (16.66)* > Hecatomb Subligar +1 (12.44) = Heca nq (12.76) > Dragon+1 (12.9) = Barbarossa's (12.75) > Dragon nq (10.32)
(feet) Enkidu's (17.8) > Dragon +1(13.89) > Dragon (10.12) => Dusk (10.06)* > Leapers (7.74)

Trick Attack
(head) Maat's Cap (35.21) > Skadi's Visor (33.02) > Denali Bonnet (31.18) > Dragon+1 (16.75) > Turban (17.5) > Dragon (13.4) > Emp.Pin (11.22)
(neck) Hope Torque (16.75) > Crocodile Collar (13.4) > Love Torque (9.51) > Agile Gorget (6.7) => Justice Torque (6.45) => Tiger Stole (6.3) > Spike Necklace (5.04)
(ears) Suppanomimi (19.65) > Genin (13.4) > Drone (10.05) > Merman's (7.56) > Coral (6.3) => Brutal (5.0)
(body) Denali Jacket (44.84) > Dragon Harness +1 (41.3) > Skadi (36.22) > DH nq (35.04) > Antares Harness (26.8) > BlueCote+1(24.76) > Rapp (14.0)
(hand) AF+1 (?, ~54.77) > Skadi (35.8) > Dusk+1(2xMarch) (21.56)* = Dragon+1 (20.53) > Dusk(2xMarch) (16.8)* = Dragon (15.92) => Tarasque Mitts +1 (15.12)
(rings) 5Agi Ring (16.75) > Rajas (8.4)
(back) Charger's (25.2) => Cerberus Mantle +1 (24.06) > Forager's (22.77) > Amemet +1 (21.48) > Commander's (15.09) = AF2 (14.96)
(waist) Ninurta Sash (28.56) > Speed Belt (21.0) > Grenadier Belt (18.87) > Swordbelt +1 (15.12) => Scouter's (13.4) > Ryl. Knight's Belt (10.06) > Vanguard Belt (10.17) > Warwolf Belt (8.4) => Swift Belt (7.7)
(leg) Dusk+1 (18.9)* > Dusk (17.64)* > War Brais +1 (16.75) > Barbarossa's (15.66) > WarBrais/AF (13.4) > Heca nq (-13.68) > Hecatomb Subligar +1 (-17.77)
(feet) Enkidu's (23.26) > Dragon +1 (18.74) > Dragon (13.74) > Dusk (13.3)* > Leapers (11.22)

For the mathematically inclined, I used the following formula to measure the marginal increase of damage from an SA/TA "act":

(SA/TA round Dmg) - (TP round opportunity cost) 
= 
(Initial Critical Hit)+(Extra Swings) - (dDelay Function)*(TP round Dmg) 
 
= 
{(Main. +fStr'a +Dex/Agi)*(pDif'a+1) + [(Main. +fStr'a)*Dbl./Trip.Atks + (Off. + fStr'a)*(1 + Dbl./Trip.Atks)]*(CritRate'a*(pDif'a+1) + NonCritRate'a*pDif'a)} 
- ((delayTP - delaySA)/delayTP)*[(Main. + Off. +2fStr'b)*(CritRate'b*(pDif'b+1) + NonCritRate'b*pDif'b)*(1 + Dbl./Trip.Atks)*Accuracy


There's more detail on understanding the formula in the LJ. Also included is the method that I used to come up with the stat ratios and corresponding gear rankings, excel spreadsheets included. Short story, I used my base sa/ta gear (that I knew for sure were the best options), my tp gear, the standard lvl82 greater colibri, double-march/haste and an accuracy buff, and variable gear options to create a spreadsheet of over 950 gear combinations for SA and over 100 for TA. From there, I found the best sets for each and simply worked out how much a single point of an attribute increased total damage, and then rationalized them.

As I said earlier, these ratios/rankings won't always hold. It's more of a general guide, though I think the gear rankings should generally hold pretty well for most situations though, even if the stat ratios change.

I really hope this can be of decently useful reference to some of the devoted thieves we have on the forums here. This project took far longer to finish than originally anticipated, and I was really glad when it was over. Also, double-checking of the math here is welcome and definitely encouraged, though I feel pretty comfortable with it. The biggest errors would probably be some truncation shortcomings, but I'm pretty confident this hasn't affected the end results in any major ways.

Edited, Jul 30th 2009 12:33pm by Shamaya
Edited, Sep 24th 2009 5:12am by Shamaya
Edited, Sep 25th 2009 12:51pm by Shamaya

Edited, Sep 25th 2009 3:16pm by Shamaya
____________________________
Shamaya, Asura
#2 Jul 30 2009 at 6:00 AM Rating: Decent
*
91 posts
I've always had a suspicion that attack did more for SA/TA than people gave credit for.
Glad to see I wasn't doing it wrong when I used my Coral earring over my Minuet.

Also using your 'point system'

Mirke (10attack/3dw lol i know) = 37.19
DH+1 = 31.53
Mirke (10atk/10acc) = 30.2
DH = 27.1
Mirke (10acc/3dw) = 25.79
Antares = 20.4

Haha Mirke 10atk 3dw wins by a significant margin, although I doubt its worth specing that way over the 10acc 3dw, or haub style.

Quick question, are Skadi and Striders ranked high because of the increase in movement speed? If so shouldn't that lower the Dusk rankings?

Great post, will help tremendously with the advent of FOV augments.

Edit - NVM the last the strider/dusk question, saw answer on your LJ.

Edited, Jul 30th 2009 10:02am by Zelphan
____________________________
//Zelphan/75THF/Ragnarok//
TP//DE
#3 Jul 30 2009 at 7:25 AM Rating: Decent
Scholar
**
485 posts
Interesting stuff Shamaya.

Especially the TA list:
(leg) Dusk+1 > Dusk > War Brais +1 > Barbarossa's = WarBrais/AF > Heca nq > Hecatomb Subligar +1 >

I thought I was doing okay with macroing NQ heca for TA. It has 20 attack and all. But it is nearly at the bottom of the list because of the 12% slow.

I am going to make a few changes to my TA setup now.

#4 Jul 30 2009 at 8:34 AM Rating: Good
Sage
**
523 posts
Not to nitpick but it looks like you left out Cuchulain's Mantle from your Sneak Attack options, and provided I'm not a complete idiot it looks like your rankings would have it in between the Amemet Mantle +1 and the Commander's Cape.

Out of curiousity, did you figure in a food with your numbers? I saw the songs mentioned but no other buffs and I'm just wondering, really.
____________________________
Server - Bahamut
Name - dawdr
Mithra 75thf/40rng/37nin/37war/37drk/37sam
#5 Jul 30 2009 at 8:43 AM Rating: Decent
***
2,976 posts
I think im reading it wrong, but how does haste affect damage? SA and TA are a one-off attack and while Haste will let you land it quicker, I don't see how it will actually make it land harder. Don't misunderstand, I realise that Haste is great for the DoT and raising TP and everything, but for the actual SA/TA, I fail to see where it's benefitting you.
____________________________
R.I.P. Crystan FFXI
75WHM 75SCH 75RDM 75SMN
#6 Jul 30 2009 at 9:07 AM Rating: Good
*
91 posts
The haste benefits you by letting your next TP swing land faster.
With 4/5 heca it isn't any stretch to state that you will be missing out on 1 TP round, So depending on how good your TP set is and especially how potent haste buffs you're receiving, you may actually have less damage over time. The extra damage from the slow may not outweigh the damage of an extra TP round, not to mention extra TP gain.
I would wager in zerg capped haste situations it would be best not to swap out of TP gear for SA/TA

same goes for dual wield.


Edited, Jul 30th 2009 1:13pm by Zelphan
____________________________
//Zelphan/75THF/Ragnarok//
TP//DE
#7 Jul 30 2009 at 9:15 AM Rating: Good
Avatar
****
4,327 posts
Here I was thinking that my lack of a decent SA/TA macro set was ruining my THF, but according to this I've been doinitrite :D
____________________________
drk = 80 sam = 76
pld = 79 thf= 80
nin = 80 drg = 75
mnk = 76 war = 52

Retired for now ^ ***** you Abyssea. FFXIV woo eh..
milich wrote:
buttfucking
#8 Jul 30 2009 at 9:26 AM Rating: Decent
Scholar
**
485 posts
Please lets make one thing clear for new thf's. You _must_ make proper gear swaps in order to gain maximum dmg output over time. Please do not be lazy i.e. don't swap for SA/TA.

The post Shamaya made is not an excuse to be lazy and don't swap for SA/TA. He even says so in closing statement (see blog):
*Likewise, laziness and not swapping for sa/ta does decrease your damage output.
*Mid-range equips, with a combination of good stats, seem to be the best.
#9 Jul 30 2009 at 9:44 AM Rating: Good
Scholar
***
2,881 posts
its not for the actual SA/TA round, its for the melee round after it.
____________________________
Lv75 Elvaan Thief, Dragoon, Warrior, Black Mage, Ninja, Bard
Number of times I have reached 75 because of death: 100
Total Merits: 432 Levels until Maat Cap: 37
Why fight like a man when I can fight like a chick with a bear?
#10 Jul 30 2009 at 10:16 AM Rating: Good
Scholar
*
57 posts
Same question here as I asked over at KI, why the skadi / trotters.
____________________________
http://ffxi.allakhazam.com/profile.xml?193950
NEW PROFILE -YaY-
#11 Jul 30 2009 at 10:26 AM Rating: Good
**
269 posts
All the movement + / - options are italicized, which I'm guessing means it's hard to quantify them. If you don't have to run anywhere for your SA/TA, you can effectively take trotters/striders off the list. If you do, though, they could be better then anything else (and by default, dusk would then be worse then everything else)

Thanks for the list Shamaya. For now I'll just edit my suppa in place of the brutal and call it a day. Also might have to work on a chargers.
____________________________
Wolfheart 80 THF, DRK, MNK but not bard. No, no I didn't level bard.
Phoenix server.
( ')> Shadow Duck approves of this message.
#12 Jul 30 2009 at 12:40 PM Rating: Decent
******
22,699 posts
Just to clarify, these ratios do not take into account places where you're getting attack boosts like x2 minuet or chaos roll and etc, right? (Or is there an option for me in the excel sheets to change the attack and get new ratios?)

Edited, Jul 30th 2009 4:42pm by Deadgye
____________________________
Dear people I don't like: 凸(●´―`●)凸
#13 Jul 30 2009 at 2:20 PM Rating: Decent
Avatar
***
1,713 posts
I think attack is a bit over rated in this system. 2.2 attack is not 1 dex in SA.
____________________________
-LordTrey
99 THF, COR, DRK, BLM, WHM, DNC
Twashtar (90)
Leviathan
#14 Jul 30 2009 at 3:42 PM Rating: Good
***
1,387 posts
These ratio probably works only work merit mobs etc etc.. I would still gear towards "Biggest-bang-of-the-slot" if you are fighting things with higher defense.
____________________________
THF75 BLM75 WAR75 NIN75 RDM51 BST38

lolgaxe+1 wrote:
Shock value isn't always a bad thing. Square Enix does it all the time. I mean, put a Ninja in a Walahra Turban and have him two hour. Is that politically correct? NO.

#15 Jul 30 2009 at 4:36 PM Rating: Decent
*
69 posts
These are all for SA/TA with out weapon skill. How would they compare with weapon skills?
#16 Jul 30 2009 at 4:58 PM Rating: Good
***
2,270 posts
WS is all out. WS is a damage ability, not a melee hit. Like Jumps on Drg. They have ZERO effect on the melee rounds, so the whole idea of slow/haste on WS is irrelivant. I WS in 4 pieces of Heca (3+1s) and the slow doesnt mean anything. The slow/haste only applies for SA/TA because it is a melee hit so the following hit will be significantly slowed down. It doesnt happen with "instant" JAs like Jumps/WS etc.

Also, <3 Shamaya. Is there anyway you could upload the spreadsheet? I would LOVE to test these a little bit on non-mandau setups. A little bug is bothering me saying that the ratios might change for others because Mandau/Blau attack rounds are significantly stronger than Blau/sirocco while SA/TA strength is almost unchanged. Putting on slow gear or removing haste on SA probably hurts mandau more as your melee is much stronger. The relative strengths of SATA and melee are probably skewed by mandau into melee's favor. REALLY want to check it on my own gear if you could upload.
____________________________
Aanalaty, Mithra of San d'Oria
99Thf/99Drg/99Rng/95Pld/Secret Job 92
Bonecraft-100+3 7/29/08
Lu Shangs-5/28/08
Twashtar-90
Ryunohige-75: 1/7/2012
Mandau-75: 3/9/2012

Maat-1/1 Thf Meleed (pansies steal)
Some of mah gearz: http://www.ffxiah.com/player/Lakshmi/Aanalaty#item-sets
Join 'People Against The Heart Snatcher' and help put Thfs on the right P.A.T.H.S. today!
#17 Jul 30 2009 at 5:02 PM Rating: Good
*
91 posts
He has the spreadsheets up for download on his LJ, the link is in the OP.
I was gonna tweak them with Dolch/Siro myself, but I'll jus let ya do it. ;p
____________________________
//Zelphan/75THF/Ragnarok//
TP//DE
#18 Jul 30 2009 at 5:29 PM Rating: Decent
Edited by bsphil
******
21,739 posts
LordTrey wrote:
I think attack is a bit over rated in this system. 2.2 attack is not 1 dex in SA.
It might be... if anything in here accounted for the defense of your target.

Great for maximizing SA/TA based on your gear/buffs on level 82 greater colibri... but doesn't hold much water beyond that.
____________________________
His Excellency Aethien wrote:
Almalieque wrote:
If no one debated with me, then I wouldn't post here anymore.
Take the hint guys, please take the hint.
gbaji wrote:
I'm not getting my news from anywhere Joph.
#19 Jul 31 2009 at 12:04 AM Rating: Good
Sage
Avatar
*****
10,815 posts
well, first of all, the equivalencies plainly don't generalize. compare X ATT vs gcolibri against X ATT vs debuffed kirin against X ATT vs non-debuffed kirin. also compare the value of 1 DEX in a heavy DEX setup vs in a not very heavy DEX setup. also note that 1% DA on top of 0% is a different DoT increase than on top of 5%.

however, to the people saying that ATT is being overvalued here, don't be so sure. notice:

a) DEX is 1 base damage. "hoho, must be a big increase!" well, if you have a lot of DEX, 1 DEX will be well below 1% damage increase.

b) "hey, i've been at this for a long time. 1STR generally = 2ATT for 1handers with a decent amount of ATT." this is true, but see (a). base damage gets bumped the fuck up by all the DEX (or AGI). this lowers the value of STR. ("but what about the ATT that STR adds?" not sure if OP factored that in; no offense, i can't really be bothered to read the whole thing;;;).

c) "yeah but surely DA doesn't care about all the DEX..." but yes it does. your first hit is gonna crit and have a high base damage (the DA isn't a guaranteed crit nor does it have the DEX enhanced BD does it?). therefore, as on hagun WSs (for example), the DA's value takes a big hit.

(a)-(c) is what's inflating the goodness of ATT. it's hard to swallow still; we're talking about a guaranteed crit, so you wouldn't expect ATT to do much (as crits simply add 1 to pDIF). nonetheless, the biggest source of damage in a THF's SA/TA is the base damage increase. therefore, once your base damage reaches a certain level (relatively ill defined and situational what that level will be), it makes sense that you'd want to raise the pDIF side of the equation.

edit: note,
Greater Colibri  Level:81-82 Def:322-327 VIT:67     
Kirin            Level:92    Def:413     VIT:120


Edited, Jul 31st 2009 4:07am by milich
____________________________
pahn
retired monk

i wish to be the red comet.
#20 Jul 31 2009 at 1:04 AM Rating: Decent
******
22,699 posts
Quote:
(a)-(c) is what's inflating the goodness of ATT.


I think it was more the not including any forms of attack boost, even food, that inflated the goodness of attack.
____________________________
Dear people I don't like: 凸(●´―`●)凸
#21 Jul 31 2009 at 1:11 AM Rating: Good
Sage
Avatar
*****
10,815 posts
The One and Only Deadgye wrote:
Quote:
(a)-(c) is what's inflating the goodness of ATT.


I think it was more the not including any forms of attack boost, even food, that inflated the goodness of attack.


no, run the numbers. i thought the value of ATT was high, but when i thought about it and postulated some base damages, i saw that 1 DEX was only giving like .6% DoT.

500/322-.3 = 1.252
502/322-.3 = 1.259
-> 5%~ increase.

it was definitely more (a)-(c) than any underestimation of how much ATT you have.

edit: p.s. by the way, i find relativizing everything to DA to be pretty bizarre. and by bizarre i mean... a bad enough idea to justify redoing the whole thing. the value of DA on SA/TA is based on

a) how much DA you start with,
b) how much DEX/AGI you stack,
c) how much ACC you have,
d) how much ATT you have,
e) your crit rate on the additional hit

shamaya, if you read this, i urge you to redo the whole thing, b/c making it all relative to DA is a bad choice. pick some amount of ATT and a target mob and relativize to that, or something else more stable than DA or base damage.

Edited, Jul 31st 2009 5:15am by milich
____________________________
pahn
retired monk

i wish to be the red comet.
#22 Jul 31 2009 at 1:47 AM Rating: Decent
******
22,699 posts
500/327 - .05*7 + 1 = 2.179
502/327 - .05*7 + 1 = 2.185

That's only a.. .28% increase isn't it?

Let's assume I have one brd for minuet+march; and I'm using kabobs against 82 colibri.

33 + 8 + 133 = 174, increasing dex by one would be 1/174 = ~.57%
401 + 66 + 60 = 527, 527/327 - .35 + 1 = 2.261;
531/327 - .35 + 1 = 2.273; ~.53% increase.

1 dex still increases my damage more than 4 attack. This does however ignore my offhand weapon.. which attack would have a bigger effect on than my main hand. I don't really want to factor than in though. Smiley: lol

Edited, Jul 31st 2009 6:03am by Deadgye
____________________________
Dear people I don't like: 凸(●´―`●)凸
#23 Jul 31 2009 at 1:50 AM Rating: Decent
Silent But Deadly
*****
19,999 posts
milich wrote:
500/322-.3 = 1.252
502/322-.3 = 1.259
-> 5%~ increase.
0.55% actually.
____________________________
SUPER BANNED FOR FAILING TO POST 20K IN A TIMELY MANNER
#24 Jul 31 2009 at 1:58 AM Rating: Good
Sage
Avatar
*****
10,815 posts
The One and Only Deadgye wrote:
500/327 - .05*7 + 1 = 2.179
502/327 - .05*7 + 1 = 2.185

That's only a.. .28% increase isn't it?


yeah. it's only a .32%~ increase if you start at 400ATT too. i wonder if s/he even considered the crit when valuing ATT.

how much DEX do you SA in anyway? like i said in my edit, this makes a big difference to everything. if BD is 200~, 1 DEX is .5% DoT. if BD is 100~ it's 1% DoT.

also, s/he seems to have 2.2 ATT = 1 DEX. note that this makes no difference at all to the cRatio you'd end up with in your example, which sort of touches on how trifling some of these differences/comparisons are. still, BD is going to be the difference maker, even if ATT was just mistakenly overvalued.

edit: in case anyone's confused, by 'BD' i mean "base damage" not "blau".

further edit: fixed decimal point typo;;;.

Edited, Jul 31st 2009 6:18am by milich
____________________________
pahn
retired monk

i wish to be the red comet.
#25 Jul 31 2009 at 2:10 AM Rating: Decent
Silent But Deadly
*****
19,999 posts
GRAAAAAAAAAAAAAAAAAGH RAGE MODE

Decimal points are going in the wrong places in milich's posts due to beer (either a shortage or a surplus, I'm not sure which yet).

0.32% increase, not 3.2%.
____________________________
SUPER BANNED FOR FAILING TO POST 20K IN A TIMELY MANNER
#26 Jul 31 2009 at 2:12 AM Rating: Good
******
22,699 posts
MDenham wrote:
GRAAAAAAAAAAAAAAAAAGH RAGE MODE

Decimal points are going in the wrong places in milich's posts due to beer (either a shortage or a surplus, I'm not sure which yet).

0.32% increase, not 3.2%.


Well, it being 6am could also effect his decimal placement.
____________________________
Dear people I don't like: 凸(●´―`●)凸
« Previous 1 2 3
Reply To Thread

Colors Smileys Quote OriginalQuote Checked Help

 

Recent Visitors: 1 All times are in CST
Anonymous Guests (1)