Talk:Thief Weapon Skill Damage Guide by Malizia
From FFXIclopedia, the free Final Fantasy XI encyclopedia
Looks like I found information on multi-hit weaponskills at the VZX FFXI Doc. I'll update the math to include this new information when I get the chance. It seems the key difference is later hits have an fTP of 1, are still affected by Trick Attack, but are not affected by Sneak Attack. They also have a lower pDIF since they are not guaranteed crits. --Malizia 15:19, 23 April 2007 (EDT)
Contents |
trick attack mods
I'm not following how you figured agi modifies DE more than dex? Which formula are you referring to?
AGI only modifies DE because of the Trick Attack damage. On the first hit, DEX is significantly more important. The thing is, Dancing Edge is a 5-hit weaponskill (6 if you're THF/NIN), and Trick Attack applies to every hit. Sneak Attack does not. So for those 4-5 extra hits, AGI (in the Trick Attack bonus) becomes about three times as important as DEX (in the weapon skill modifier bonus). This is why AGI is more important in the end. --Malizia 02:44, 25 April 2007 (EDT)
Oh, one more note... there's nothing in the Dancing Edge weaponskill that's affected by AGI. The bonus from AGI is entirely due to Trick Attack. Without Trick Attack, AGI won't affect the weaponskill at all. --Malizia 02:53, 25 April 2007 (EDT)
More discussion on Trick attk/Agi mods on DE
I understand completely what you're referring to. It's just that it is so completely against intuition. Why would agi affect Dancing Edge more than Dex? I've searched that page and don't see a formula for the extra hits that would make me thing agi affects where Dex doesnt. Which formula are you referring to? Only difference I see is that the Agi bonus from Trick attk doesnt give it the pDif modifier which explains why TA's are weaker than SA's. BUt like I said this whole agi > dex is confusing me and I'm trying to understand where you got that formula.
From VXZ's page here is what I'm seeing (perhaps I'm blind) and it's not matching in my mind to your theory.
For THF Main, during the activation of Sneak Attack and Trick Attack, the total DEX and the total AGI of your character is calculated as base damage respectedly.
For Sneak Attack: Damage = (Base Damage + Total DEX) x critical hit pDIF
For Trick Attack: Damage = (Base Damage + Total AGI) x pDIF
For Sneak Attack and Trick Attack: Damage = (Base Damage + Total DEX + Total AGI) x critical hit pDIF
For Sneak Attack and Trick Attack enabled WS: Damage = ( Base Damage + Total DEX + Total AGI) x critical hit pDIF
fTP fTP is the Base Damage multiplier when you perform weapon skills For multi hit weapon skill, fTP of 2nd hit and the following hits will have value 1.00
fTP of Dual Wield and Double attack Dual Wield and Double attack will add another extra hit into your WS For example : Dancing Edge and Rampage are 5-hit WS. when the user sub Ninja or when double attack effect kicked in, the damage and TP gain will be counted as 6-hit WS. However, the added attack will have fTP= 1.0 as well as the consecutive attack.
So, the WSC calculation is
WSC = floor ( floor((stat_percentage_1 x stat_1) + (stat_percentage_2 x stat_2)) x α )
For Weapon Skills: Base Damage = floor( (D + (aD) + fSTR(2) + WSC) x fTP)
That's all standard stuff but nowhere do I see anything that would suggest that agi keeps modifiing where dex doesn't for the ftp=1 hits. From the formula above it would suggest no change except that fTP=1 after the first hit.
SATA and multihit weaponskills
You pulled the right information from the VXZ page. The extra information I'm using is found on the Trick Attack page:
- "Unlike Sneak Attack, when used in conjunction with a weapon skill, this ability will enhance all hits of a multi-hit weapon skill."
and the Sneak Attack page:
- "When used in conjunction with a non-elemental weapon skill, this ability will enhance the attack. Only the first hit of a multi-hit weapon skill is enhanced unless the user is holding two weapons or using Hand-to-Hand, in which case two hits will receive the effect."
So, the Sneak Attack modifier wears off after the first (two) hits, but the Trick Attack modifier lasts through all the hits. I actually made a mistake in the guide - I thought the Sneak Attack modifier only lasted for the first hit, regardless of Dual Wield, so I need to fix that.
The Trick Attack page only says "enhances," which means possibly only the hate-transfer effect is applied to all the hits. I took it to mean the AGI damage bonus is also applied to all the hits. --Malizia 12:35, 25 April 2007 (EDT)
Thats where I think you are probably wrong. The hate transfer is indeed all hits, but the agi modifier is most likely not. ALso we dont know where they got their information from. I have never seen any evidence showing that agi would keep modifing where dex wouldn't. Thats a bold statement and I've seen nothing to substantiate it thusfar.
Just trying to keep things as accurate as possible. I've been a thf for a long time and have never done real tests on agi modifier on trick but from what I've always seen dex/str > agi even for TA modified ws's(to a reasonable extent of course).--Tdizzle00 13:54, 25 April 2007 (EDT)
Personally I agree, I was surprised with the end results too. I was really doing all of this to see how much I should focus on STR, not to see whether DEX > AGI. How about this: until I have evidence that the AGI damage modifier from Trick Attack is used in all of the hits, I'll calculate both possibilities. Meanwhile, I'll try to think up some way to do some testing on this. It's going to be tough, considering I'll need to find a monster that can withstand all 5 hits of a weaponskill without dying, but is weak enough so the pDIF is at max. Thanks for bringing this up! --Malizia 15:44, 25 April 2007 (EDT)
Questions
So after the above discussion, here are things that need verification for multihit weaponskills:
- For which hits does the Sneak Attack DEX damage bonus apply?
- For which hits does the Sneak Attack guaranteed critical hit bonus apply?
- For which hits does the Trick Attack AGI damage bonus apply?
Other questions I have:
- How does DEX affect the chance to critical hit?
- How does DEX affect accuracy? More importantly, what's the formula to convert accuracy to a chance to hit?
--Malizia 18:09, 25 April 2007 (EDT)
My response to your questions: -Dex dmg applys to first hit only on SA modified ws's -Crit hit applys to first hit only -Trick attk agi bonus applys to first hit only in the same manner as SA and Dex.
-Dex affects the chance to critical hit in a non linear fashion depending on mobs agility (the exact stat may have to do with level difference also). But it is not a 1/1 or 1/10 or anything liek that. There seems to be tiers that you need to reach before dex starts to really increase crit hits. There has been recent testing on BG and Ala regarding this. IT seems that when you are close to a mobs 'tier' a few dex will dramatically increase critical hits (something like 20 dex will give an approx 10% increase). But when you get farther below or beyond that tier whatever it may be for dex vs mob adding dex does little (see adding 40 dex giving 1-2% increase).
-Dex is said to affect accuracy at a 2:1 ratio and at some levels it may be true. I believe it was tested on very low level characters on respectedly low level mobs but am not sure if it still applys at higher levels.--Tdizzle00 18:40, 25 April 2007 (EDT)
I don't think I made my questions clear enough:
- About the sneak/trick dex/agi/crit damage questions: what about the second hit if you're dual wielding? I quoted above the Sneak Attack page which says the bonus applies to the first two hits of a dual-wield WS. Is this not true? Does it also apply to crits and the AGI bonus from TA? I'm going to try to test this anyway when I can.
- I just read the BG forum topic on crit hits and tiers. I don't think I have access to enough DEX equipment to test this myself, and their results don't give me a mathematical formula to work with. I'm looking for that formula.
- I saw the DEX -> Accuracy in a 2:1 ratio bit on the Accuracy page. Once again, I'm looking for a formula: if I know my accuracy, what is my % chance to hit a monster?
--Malizia 19:35, 25 April 2007 (EDT)
Accuracy hasnt been testing extensively but from testing that has been done, there seems to be a level correction much like the dmg formula. So if you are much higher or much lower in level you have a bonus or hinderence to your accuracy. This is fairly easy to see when you go naked with a weapon with a skill of 60 and can hit level 50 mobs easy with you level 75 character but go on a level 45 character with the same skill and you will miss almost every time. So it seems that when you are close to the mobs level there is more of a straight correlation to the mobs agi/evasion vs your skill/accuracy. Probaly much like how the attk ratio is figured in dmg.
I dont think the dual wield statment is true also. Just use SA alone and see how it reacts. I think perhaps a long time ago it would guarantee crits for both dual wield hands. First one was modified by the SA mods but the 2nd hit (offhand) was merely a crit. Now it just makes the offhand hit 100% to hit with no crit involved. I think it's similar with Trick attack. I'm fairly sure of this statment. The offhand hits dont recieve any of the dex/agi mods. For TA enabled ws it will receive the hate transfer though but nothing has ever made me think it will receive any of the dex/agi mods from trick or sneak (nor do any of the hits of a multi hit ws other than the very first.--Tdizzle00 09:33, 26 April 2007 (EDT)
Sneak Attack test results with THF/WAR
Test setup: THF75/WAR37, DEX 90, D = 33/26 (Blau Dolch), no offhand. I fought against White Lizards. I always opened the battle with Shark Bite at 300% TP, combined with Sneak Attack but not Trick Attack. WSC = 37. By itself, Sneak Attack would max out at 399 before 100% tp and 378 after. Using the non-weaponskill damage formula for Sneak Attack, this implies fSTR = 10, which is the maximum for Blau Dolch. pDIF is always variable, but at least I know I can hit the maximum (2.4) against the lizards because that would be the only way to hit those maximum Sneak Attack numbers. This implies my cRatio is greater than or equal to 2, so my minimum pDIF is 1.6. My weaponskills would vary between 5 and 7 TP returned, indicating Double Attack or Triple Attack kicked in. Only two of the 7-TP Shark Bites ever managed to kill a lizard (with 1454 and 1472 damage).
Damage calculations using the above statistics (D1 is the first hit, D2 is for the other hits):
D = (floor((D + fSTR + WSC) * fTP) + DEX)pDIF
D1 = ((33 + 10 + 37) * 3 + 90)pDIFcrit = ((330)pDIFcrit
D2 = ((33 + 10 + 37) * 1 + DEX)pDIF(crit?) = (80 + DEX)pDIF(crit?)
|
Maximum possible damage: pDIF = 2.4, no DEX bonus on extra attacks: pDIF = 2.4, no DEX bonus, yes CRIT bonus on extra attacks: |
Minimum possible damage: pDIF = 1.6, no DEX bonus on extra attacks: pDIF = 1.6, no DEX bonus, yes CRIT bonus on extra attacks: |
Actual damage results:
5 TP returned:
1057, 1067, 1090, 1109, 1145, 1150, 1155, 1166, 1168, 1177
6 TP returned:
1280, 1314, 1363
7 TP returned:
1392, 1403, 1454, 1472, 1499
Analysis:
I wanted to determine if Sneak Attack affected more than the first hit of a weaponskill, in terms of the DEX damage bonus and the 100% chance to crit. From looking at my actual test results, my numbers only match the formulae if Sneak Attack does not affect the later weaponskill hits at all. This means neither the DEX bonus nor the 100% crit bonus applies to more than the first hit of a weaponskill.
However, I have not had the chance to test using THF/NIN (for Dual Wield) nor using a trick buddy for Trick Attack analysis. The Dual Wield tests should not be too hard to do in the near future, but getting a dedicated trick buddy might take a while.
--Malizia 14:52, 7 May 2007 (CDT)
I think the weapon rank for Blau is rank 2 because you have to use the non latent damage.--Tdizzle00 20:27, 8 May 2007 (CDT)
That's correct. Thus the maximum fSTR = floor(D/9) + 8 = floor(26/9) + 8 = 2 + 8 = 10, where floor(D/9) is the Weapon Rank. --Malizia 02:21, 9 May 2007 (CDT)
Trick Attack test results with THF/WAR
Test setup: THF75/WAR37, DEX 90, AGI 79, D = 33/26 (Blau Dolch), no offhand (same setup as last time, except with a trick partner). WSC = 37. In this case I would always do a 300% TP Shark Bite with Trick Attack and no Sneak Attack. Only one of the 7-TP Shark Bites ever managed to kill a lizard (with 1409 damage).
Damage calculations using the above statistics (D1 is the first hit, D2 is for the other hits):
D = (floor((D + fSTR + WSC) * fTP) + AGI)pDIF
D1 = ((33 + 10 + 37) * 3 + 79)pDIFcrit = ((319)pDIFcrit
D2 = ((33 + 10 + 37) * 1 + AGI)pDIF(crit?) = (80 + AGI)pDIF(crit?)
|
Maximum possible damage: pDIF = 2.4, no AGI bonus on extra attacks: pDIF = 2.4, no AGI bonus, yes CRIT bonus on extra attacks: |
Minimum possible damage: pDIF = 1.6, no AGI bonus on extra attacks: pDIF = 1.6, no AGI bonus, yes CRIT bonus on extra attacks: |
Actual damage results:
5 TP returned:
991, 1042, 1044, 1047, 1048, 1051, 1070, 1091, 1118, 1140
6 TP returned:
1292
7 TP returned:
1370, 1409
Analysis:
It's the same results as with Sneak Attack!! Trick Attack adds the AGI bonus to the first hit of a multihit weaponskill, and only the first hit.
All that's left is testing with THF/NIN. THF/WAR has the advantage of Berserk and Warcry, so I can be sure to maximize Attack before firing a weaponskill. The NIN sub will not have that, but it should be interesting to see how SATA affects the second hit when considering Dual Wield. More to come soon! Meanwhile, the guide is updated with this new information.
Finally, Tdizzle00, this seems to verify your intuition. As THF/WAR, DEX is certainly more valuable than AGI with most weaponskills. It would be highly unusual for THF/NIN to have a different result.
--Malizia 01:41, 20 May 2007 (CDT)
