DA in WS & Gorget Acc Tests

[Mojo]

Double Attack does help two handed weapons more than one handed. From the conventional understanding of Double Attack procs on weaponskills, two handed weaponskills will fully benefit from Double Attack procs on every swing as opposed to every hand. Blade: Jin, for example, is a 4 hit WS(3 + offhand) but only has two chances to double attack. Raging Rush is 3 hits and has different 3 chances at a Double Attack proc. Correct me if I'm wrong, but I don't think Asuran Fists can even benefit from Double Attack because it already reaches the max on attack pers round.

As far as Ares vs Haidate goes. Well, I can understand preferring Haidate over Ares as it offers a lot of accuracy for two handed weapons. It's 15 DEX vs 6 DEX 6 STR 2% Double Attack though, some people might make the swap. Either way, that's where the extra 2% probably came from.


Mod Edit: This thread is a split off of the WAR/SAM in merits debate found here:
viewtopic.php?f=31&t=31532

This thread is to continue a long-winded discussion on double attack in WS, WS acc, gorget WS acc, and blahblahetc. Have a blast.

[Ryda]

Correct me if I'm wrong, but I don't think Asuran Fists can even benefit from Double Attack because it already reaches the max on attack pers round.

This is correct. MNKs in the know macro out brutal for WS because of this.

[Max™]

I know RR can DA twice, I don't believe it can DA 3 times though.

16.7% normal Tp per hit, 18 for full Tp return RR, 20 for one double (the store TP carries over is all I can figure), 21 for two. 19% is rare as hell.

The frequency of 20% returns suggests that it's gotta be a single double attack, working the math on the damage returns shows that too, math for 21% returns shows 5 hits, never any time that I've used RR ever have I seen a 22% return+ 6 hit type damage.

I've never gotten a 5% Tp return after missing the first hit, but I have gotten 3% and 4% before.

[Liquidedust]

I know RR can DA twice, I don't believe it can DA 3 times though.

16.7% normal Tp per hit, 18 for full Tp return RR, 20 for one double (the store TP carries over is all I can figure), 21 for two. 19% is rare as hell.

The frequency of 20% returns suggests that it's gotta be a single double attack, working the math on the damage returns shows that too, math for 21% returns shows 5 hits, never any time that I've used RR ever have I seen a 22% return+ 6 hit type damage.

I've never gotten a 5% Tp return after missing the first hit, but I have gotten 3% and 4% before.

No screenshot of it happening, but a 6-hitter has happened once for me since I hit WAR60 (which is 3 years ago) according to the TP return I had (pretty much certain I didn't get hit it in between but not 100). So it is possible, unless I mistook something, but most likely rare as fuck (since I have only ever experienced it once in my entire WAR career). The damage dealt was absurd as well a lot higher then the other WSs I have done at that level range on WAR (this was 1600-something damage in Bibiki Bay pre-ToAU and pre crit mod on RR as well).

Though don't take it for since I cannot be 100% certain I wasn't hit in between WS and checking TP, just along the lines of 80-90% sure.

[Mojo]

I've heard this before(That RR can't DA more than twice), it's hard to say because the chances of it happening are low. With 24% double attack, the chances of a 6 hit RR would be around 1/200(assuming 90% accuracy). This would be much easier to observe if you had a COR doing Fighters roll, as rolling an 11 would bring those chances up to about 1/10. I'd like to know, either way.

[Max™]

Last night I struck upon a nice way to see some gear change differences for raging rush (just got my heca mitts, yay) over a long period of time.

Fortifications, especially in Garlaige, always the same level, doesn't hit back/buff itself, just sits there and soaks damage.

So yeah, I sat there for about 35-40 minutes just WSing/checking Tp returns/watching damage.

Full DA merits/Pole Strap/Brutal, so 22% DA, 6 hit build, full time Hasso of course, fully merited Warrior's Charge too, and I 2hr/WCed at the start to determine what a 4 hit/full crit would do.

Results were basically that Chiv chain was just barely noticeably better than blank neck slot, and Snow Gorget only produced a couple of missed first hit/17% returns, but definitely seemed to be giving more acc to all the hits than Chiv, still worth more testing, but besides the acc being good (though to be fair I should go almost naked and repeat it, cause I was WSing in like 110 some odd Dex) it also produced the best damage results, cementing it's spot in my WS macro.

Of all the WSes I did, I remember a handful of 21% returns, 3 or 4 maybe.

Think I'll repeat and do better documenting later when I get back on, but yeah... I did hundreds of WSes last night, have done thousands of crit TP RR's alone, never seen a 22% return/damage, honestly don't believe it's possible.

[RKenshin]

I also remember seeing what I thought were 3x DA RRs, one of which was 2-3 years ago in Dynamis with 2h, for a ~2.5k RR in Sandy. I can't really think of a way to explain a 2.5k RR without SV or any defense downs on a dynamis mob, without DA'ing three times, so until then I'm reluctant to consider otherwise! Not to mention my TP return suggested it as well, and I doubt they changed anything since then.

So yeah, I'd like to think that short of getting cut off by the 8 swings per WS/attack round cap, you can DA as many times as there are swings in your WS. Of course though, the odds of it happening and landing so many hits is still rather low.

With RR in particular, having 85% acc and a 15% DA-rate you're looking at about a 1/1000 chance of it happening. If you include the fact that at least 50% (in reality, probably much higher) of the time a 5-hit RR will kill a mob before the 6th hit could land, it's more like a 1/2000 chance. Then there are other factors like Flashga/Windwall/Stoneskin/Evasion boost that all decrease the chances further. All in all, it is not something you'd typically see.

In addition, it has been proved that every hit has a chance to DA through DA percentage testing as a DRK with Guillotine. IIRC the test subjects averaged like a 35% chance to DA once in Guillotine, which would not be possible otherwise if each hit didn't have a chance to DA.

Ares Flanchard. Also, something I think that is interesting to note about WAR/SAM and DRK/SAM is how much they benefit from a COR in party over other jobs. Fighters Roll can bring your Double Attack up by 24%. Having 48% DA on Raging Rush is amazing. DA helps other jobs too of course, but to a much lesser extent.

No, it helps them to a further extent, DA has diminishing returns. Using ares legs over haidate also makes no sense.

For multi-hit WS no, the higher amount of DA you have in a multi-hit WS the higher your chances are of double attacking at least once.

For TP gain, yes. I'm pretty sure you meant for TP gain though.

[Max™]

Ok, assuming that a Fortification is lvl 50-55, with 22% DA, 112 Dex, full merits, and about +30 Acc on top of that besides Hasso 95% Acc shouldn't bee too hard to reach, how many WSes would it take statistically to see a 3x double?

As for 2.5 RR, it used to have a higher ftp modifier before changed to Crit Tp didn't it, 2 doubles+forced crits, assuming 500ish for 5 crits with the higher mod doesn't sound unbelieveable, especially since Orcs are squishy as is.

[Mojo]

You'd have a 0.78% chance of a landing all 6 hits on a Raging Rush with those numbers, or around 1/129. However, the farther a chance is from 50%, the more prone it is to variance. If you did 512 Raging Rushes with that set up, you would only have had a 98.2% chance of seeing a six hit Raging Rush. This is high, although it would still be possible to not see one. 256 was around 10% of not seeing one. I definitely think the easiest way to approach it would be to get a COR.

[RKenshin]

I might actually do some tests of my own if time permits.

However, the most accurate and easiest way to test it is in a set of gear that as a much +DA as possible, followed by as much accuracy as possible. If you think you're at 95% WS acc when you're really at 90% WS acc, it will make a huge difference and greatly distort results. Because of this I'd also recommend using sushi and maybe even attempting to get a COR. Although with a COR you lose some accuracy in the part of test, as you may have different values of DA at different times depending on the roll.

Also, subbing SAM would likely not be the best idea as well, since it appears it can result in some inconsistent returns? Unless you're certain you've had a few 19% TP returns with no other factors to explain it. In which case I'd advice against any STP at all aside from Brutal's. I'd probably /DRG just for the haste to help TP gain, and a little extra accuracy too.

But, with 95% accuracy on a 3-hit WS with a 22% chance to DA, it's about a 1/130 chance. A reasonably accurate test would be 1000 WS without seeing a single 6-hit RR. However while doing this you might as well keep track of all 3 hit, 4 hit, and 5 hit WS just for statistical purposes, and to help back up the accuracy of your results.

EDIT: Mojo beat me to part of it but yeah lol.. Same numbers.

[Starr]

I've already discussed this with Mojo but until I see a SS of a WAR doing a 6 hit Raging Rush (or an 8 hit Penta or a 7 Hit Guillo) I don't believe it's possible. I believe WS are capped at 2 DA for a 2 handed weapon skill and once per hand for a dual wielded WS.

[Max™]

I'm going to go and put on my own personal LS later, filter everything and make a log, just because I've always wondered about this.

I'll go full acc, but as for the /Sam thing.

I've had 19% returns with normal 3 hit damage, so what I think is happening is this: after the usual first (16.7%) and second hits (+1, +1) Doubles get the Store TP carried over (20%).

If you miss a second hit, and land a double, the STP doesn't seem to add up and you get 19% with normal 18% return type damage.

21% looked like it was a store TP issue for a while, but now with it crit based, and a better understanding of damage formulas, I can tell it is 5 hits (2 DA's).

[RKenshin]

If there is a 2xDA cap, I'm actually more concerned over whether it would be a cap implemented during or after the total swings are calculated for a WS. If it was during, it would reduce the effects of a higher DA rate by a few %. If it was after, it would remain untouched. But, considering how it does the 8 swing/round limit it's likely to be the first.

Which is why I think it'd be worth keeping track of the number of no DA, DAx1, DAx2, etc.

[Raen Ryong]

I'm almost certain I have done a 3x DA Guillotine, just once. 23% TP return on /sam, 20% being 4hits and I did not get hit, nor had extra StoreTP on than usual (no COR in party). Sadly, I don't think I SS'd it before I took another swing which kinda ruins it. Regardless, I am positive you can DA more than twice but the chances of doing so are obviously very low.

[RKenshin]

Funny timing for this topic to come around.

In the Salvage run I just finished another WAR (Blackice), others there, and myself witnessesed Ice pull out an 11 TP return Vorpal Blade that hit as hard as a 2xDA Vorpal would suggest. Normal TP return is 9, and DA TP return is 10. He was NIN sub and single wielding for Flat Blade to prevent Frog Chorus (which worked extremely well).

I didn't think to screen it at the time, but I'm sure he and others there could come and vouch for it. We discussed it for a bit after it happened.

I've also claimed to receive a 2x DA Vorpal Blade while single wielding Ridill, about a year and a half ago. I was playing with WAR/DRG and Ridill in an exp PT, and boom... 11 TP Vorpal for around 1.2k. I don't single wield it often at all so seeing it even that one time was probably rather unlikely.

[Max™]

2 birds with one stone.

185 total WSes.

125 Using Snow Gorget, 44 Dex, 13 Gear acc, Hasso.

3 ~21% Returns
30 ~20% Returns (a flukey streak in the last 14 kind of skewed this, it WAS right at 22/100, then I got 7 in 14 >.<)
80 ~18% Returns
12 ~2 hit Returns (17% mostly, except for 2- 2% wses)

All in all 403 WS hits I believe it was, and only 12 missed WS hits... for 97% WS acc?

Snow= gewd for RR? I know the damage was consistently better.

(although, when I went to Chiv/almost nothing else for next test, I did 19% return, 441 Dmg to a fortification twice in a row, followed by a 446)

did 40 more with Chiv and almost nothing else, and Snow and almost nothing else.

That sample wasn't large enough to show the Acc rate better, but over that almost 200 I saw no 3x Doubles yet.

[Ferien]

2 birds with one stone.

185 total WSes.

125 Using Snow Gorget, 44 Dex, 13 Gear acc, Hasso.

3 ~21% Returns
30 ~20% Returns (a flukey streak in the last 14 kind of skewed this, it WAS right at 22/100, then I got 7 in 14 >.<)
80 ~18% Returns
12 ~2 hit Returns (17% mostly, except for 2- 2% wses)

All in all 403 WS hits I believe it was, and only 12 missed WS hits... for 97% WS acc?
...

Any of those 18% or 20% (and theoretically, the 21s) could have had a missed double in them. How much DA were these run with?

[Capo]

Funny timing for this topic to come around.

In the Salvage run I just finished another WAR (Blackice), others there, and myself witnessesed Ice pull out an 11 TP return Vorpal Blade that hit as hard as a 2xDA Vorpal would suggest. Normal TP return is 9, and DA TP return is 10. He was NIN sub and single wielding for Flat Blade to prevent Frog Chorus (which worked extremely well).

I didn't think to screen it at the time, but I'm sure he and others there could come and vouch for it. We discussed it for a bit after it happened.

I've also claimed to receive a 2x DA Vorpal Blade while single wielding Ridill, about a year and a half ago. I was playing with WAR/DRG and Ridill in an exp PT, and boom... 11 TP Vorpal for around 1.2k. I don't single wield it often at all so seeing it even that one time was probably rather unlikely.

ive had an 11tp vorpal for sure, i had to check because of the crazy damage even though they were too weak mobs. was war/thf with joyeuse so tp calculation should be the same as ridill, so im pretty certain vorpal can DA more than once

[Suiram]

I did a similar test on Fortifications a while back. Never once in 101 vorpal blades (BLU/WAR + Brutal) did I see a TP return that reflected a 2x DA, even though I calculated a 99.9941% chance of seeing at least one, were it possible. The details are here if someone wants to quibble with the mathematics or the methodology.

[RKenshin]

Nice job on the test, I'll try to get some statistical representations out of it. I'm assuming you had 22% DA for every WS.

2 birds with one stone.

185 total WSes.

125 Using Snow Gorget, 44 Dex, 13 Gear acc, Hasso.

3 ~21% Returns
30 ~20% Returns (a flukey streak in the last 14 kind of skewed this, it WAS right at 22/100, then I got 7 in 14 >.<)
80 ~18% Returns
12 ~2 hit Returns (17% mostly, except for 2- 2% wses)

Ok, so I have to count in the fact that DA/Zanshin can activate on missed hits and make it appear you land a full-hit/DA once/etc when you really had another DA. Over these 375 swings, 356 of them should have hit. Resulting in 19 missed swings. Not going to factor in Zanshin for this just because it's a little more complicated than I want to get. But, on 58% of those missed swings, you DA'd at least once in place of a miss. I'm just going to assume that you only missed once a WS.

So in reality you likely had closer to:
03 2x DA
36 1x DA
68 All-Hit, no DA
19 Missed-Hit

However, the possibility does exist for you to have missed 2 times in a WS, and just have DA'd twice.

With that in mind, this is your result:

1 Miss: 15%
no DA: 54%
DAx1: 29%
DAx2: 3%
DAx3: 0

And, the statistics that you should have seen based on figuring out your odds mathematically:

Chance of 1 Miss: 15%
Chance of no DA: 41%
Chance of DAx1: 32%
Chance of DAx2: 9%
Chance of DAx3: .5%

Comparing your results with what's expected, with the difference between them with green being neutral, red being your result was larger than expected, and blue being smaller than expected:
1 Miss: 0%
no DA: 13%
DAx1: 3%
DAx2: 6%
DAx3: 0

Pretty close to what was expected mathematically. With obvious small deviations from the small sample size.

All in all 403 WS hits I believe it was, and only 12 missed WS hits... for 97% WS acc?

Not quite.

Every time you miss one swing in RR, there is a ~34% chance that one of the other swings will double attack and make it appear as if you never missed a swing. If Zanshin does in fact proc on WS (which I honestly don't see why it shouldn't, but thus far there has been no conclusive evidence to support either way), and it's a base 10% rate, then this is bumped up to a ~44% chance.

So, in reality your WS accuracy was ((100 - X) * 44%) + X = 97, where X is your real accuracy. Solving for X we have about 94.8% accuracy as your real accuracy. The combination of DA and Zanshin procs made it appear to be 97% when it was really just about 95%.

This is also just as mathematically expected.

That sample wasn't large enough to show the Acc rate better, but over that almost 200 I saw no 3x Doubles yet.

It was a good sample and nice job with it, hopefully we can get some more tests to get some real conclusive evidence either way.

And, actually, the most recent idea that comes to mind is that if anything all WS in the game are capped to 2 DA. It's not a 2 DA for 2h, and 1 DA on each hand for dual wielding... As there's no other real way to explain getting 2 DA with a 1h weapon. Which appears to be a possibility.

Geno: shut up. remember why my drk stomped all over everything in merit parties? it's because I didn't waste time spamming pointless dark magic.

There was a topic awhile back during some of the initial 2hander tweaking where Ryko and I were doing some testing/parsing, and some idiot drk tried to call me out for not casting drains enough XD I was saying back then that they're generally not worth casting, and that abs-tp isn't worth casting in the slightest unless you're pretty damn sure you can get more than 35 or so TP (realistically you should be shooting for significantly more than 35).

I remember this lol, the guy was an idiot. Once again I completely agree that in most situations casting dark magic will only hinder your parse if you're going for max damage potential, not help it.

I did a similar test on Fortifications a while back. Never once in 101 vorpal blades (BLU/WAR + Brutal) did I see a TP return that reflected a 2x DA, even though I calculated a 99.9941% chance of seeing at least one, were it possible. The details are here if someone wants to quibble with the mathematics or the methodology.

It's not a 99.9941% chance. There's only about a 4% chance for you to see a 2x DA, meaning that over 101 WS you should've seen about 4. The odds of not seeing any is more like a 98.4% chance against it happening.

And, I wouldn't completely discount the possibility either way. I went 1/34 on HQ stave synths recently, and 99.94% of the time that's impossible.