(Don't) Fear The Reaper: A Dark Knight Guide

There was a typo, I meant to say 3 hits per 2 rounds. Which is what I covered with everything else.

Another note: the "law of averages" isn't a mathematical theorem, it's a layman's logical fallacy (the notion that small sample sizes will naturally conform to the underlying statistical distribution). You were looking for the law of large numbers, but the way you expressed it is perhaps more in line with the aforementioned fallacy. I can see the intent but again, could be stated more clearly.

Useful information, I will keep that phrase in mind for next time, probably won't change the way I state things as long as people understand what I meant.

if that's the case then you're looking at 49.5% chance at 1 DA, 30.25% at 2 DAs, and 20.25% at 0 DAs with 55% DA.

You should probably do it properly then. A QA adds three hits, doesn't multiply your current hit by 4. So 4 QA would be 12/100 not 16/100 since you'll always have the initial hit.

I am honestly trying to figure out if you are trolling me or we don't see things the same way. I am leaning towards we don't see things the same way.
Just because you proc a multi-attack doesn't mean you lose your initial hit. You are checking the difference between no multi-attack and multi-attack I can see it your way, if you want to want know your total attacks per round then you would go 16/100

in 100 rounds you will get 16 hits from 4% QA. initial hit giving you 4 of those hits and QA giving you 12 more.


49.5% DA isn't 1 DA, 49.5% is 0.495 hits per round. 55% is 0.55 attacks per round. Therefore if you have 49.5% DA, then in 2 rounds you will get 2.9 attacks per 2 rounds. 55% DA in 2 rounds will give you 3.1 attacks per 2 rounds and so on.

We are literally saying the same thing, just differently.

No, you're saying it completely wrong. 55% DA literally gives you 1.55 attacks per round, not 3.1, and 4% QA gives you 12 hits in 100 rounds.

no, I keep making typos, 3.1 in 2 rounds.

QA doesn't magically lose the 4th hit, you will still get 4 hits when you proc QA

I'm showing you why. Yes 55% DA will average 1.55 per hit and 3.1 over two. I'm telling you 55% DA and two attacks will net one DA 49.5% of the time, two DA 30.25% of the time, and 0 DA 20.25% of the time.

In 100 rounds with no QA, that's 100 hits. If you have 4 QA, that's 112 hits, not 116. Which is my point. You don't multiply your QA by four, you multiply by three since you're only gaining three additional hits from a QA.

The original hit is counted in the 100 rounds.

If you put on 4% QA, you gain 12 extra attacks in 100 rounds from it, meaning that 4% QA gives you 12/100. It's basic English.

Ok, then you guys aren't getting what I am saying

100 is the value you need to reach a multi-attack based of the gear you are using.

0/100 means no multi attack

2/100 means 1% double attack

if you have 100/100 (50%) double attack, since it is double it would take 2 rounds to achieve 1 extra hit, thus 100.

If you are using 100/100 (33.3%) triple attack, it will take 3 rounds to hit a multi-attack round. thus 100.

If you are using 100/100 (25%) quad attack, it will take 4 rounds to hit a multi-attack round. thus 100.

You are aiming for 100 to get your multi-attack round in.

If you have 400/100 (100%) quad attack, you will get 1 round of quad attack every round. Thus 100 being the goal.

16/100 DA = 8%
16/100 TA = 5.3_%
16/100 QA = 4% (not 12/100)

Going by those values, you can understand how long it takes to get a multi-attack round.



Maybe I should word it differently, this isn't a round, or a set number of hits. This is a value to understand how many rounds until you get a multi-hit.

You proc 4% QA, that will be 16 of your 100. Initial + 3
If you proc 4% DA, that will be 8 of your 100. Initial + 1

The higher the x/100 value is, the less times you will need to swing until you hit multi-attack.

This is where the confusion is!!!

Yeah, that's not how multi-attacks work. They aren't "counters" that build up until you qualify for an extra hit. Your terminology is misleading and useless.

I am not talking extra, I am talking total. You can't double the DA value without tripling the TA value or quadrupling the QA value.

Considering these are the numbers I have been using for years and consistently put up some of the highest dps numbers people have seen... Doesn't seem so useless. Not to mention this is the only time I have ever dealt with people that don't understand it, anyone I have ever explained it to with less detail has understood it and gained drastic improvements on their own dps.

Yeah, you don't do any of that. DA isn't multiplied, TA is doubled, and QA is tripled.

If you use the terminology "x hits from QA", you don't count the initial hit, because it's not a part of QA.

Doesn't looking at it that way assume that your hits are guaranteed? vs being rolled on every swing?

The rolls for QA TA DA are on each swing, - you could DA 100 times in a row, with 1% DA, it is just unlikely; you could also never DA with 99% DA, just unlikely.

The premises people are viewing this from are different.

Most are assuming that if you counted 100 avg swings (no counting QA TA DA) you would get an extra 12 hits with 4QA or an extra 12 hits with 6TA or an extra 12 hits with 12DA.

I am not sure why you would count the initial hit (as it isn't QA TA OR DA, just a regular hit; and that initial hit is assumed in your 100)

Agree to disagree, the way I have done it has worked for many years and done wonders on increasing my dps output by using this math. I have also had people I told to do it this way come back and tell me how much more dmg they are doing now. The fact that only a few people in the forums after nearly 10 years of explaining it this way can't understand it isn't my concern. I tried to explain it, you as well as a few others didn't get it.

edit: always count the initial hit, that is part of your build.

The way you're talking about it is literally mechanically incorrect, and it doesn't matter how many people claim to say it helped. Your explanation literally says that if you have 50% double attack, you will alternate single attack rounds and double attack rounds.

Wait, I see how you guys are explaining it and it makes sense. I have just been going off a different x/100, where as you guys would go 55% = 55/100, I just use a different value in my head. I see 110/100, as a different number in my head than you guys and it creates confusion. I will try to use your value instead of mine from now on.

55 DA would be 155, you're just adding the 55 from the DA. You have a 100% chance to swing once already.

I remember people saying adding more DA has diminishing returns? I'm not really big on the math, but the way I understood it was that the difference between 1% DA and 10% DA is huge, but the difference between 50% DA and 60% DA isn't that big of a deal.

Is it really so cut and dry as 10% DA = 110 swings in 100 rounds, 60% DA = 160 swings in 100 rounds?

yes there is diminishing marginal utility and no, it isn't always that cut and dry

You don't get reduced chance at a DA for adding more DA, it just doesn't have as large an impact on your DPS

I uploaded my accuracy gear spreadsheets to Google Drive. I used these to figure out what pieces had the max accuracy and where I could upgrade some pieces in my high accuracy set.

These only contain 119 gear. Lustratio gear is in the lists, but obviously we would not want those for a TP set.

I based the augments for reisenjima gear on what BGWiki has listed for Oseem, so there are some values missing.

The accuracy formula I used for overall accuracy is a snippet from the Accuracy formula on FFXIclopedia (DEX Contribution + Equipment, where DEX Contribution = DEX * .75).

Adding DA will only diminish if you are lowering your sTP to the point of loss of balance between DA and sTP.

Assuming you are 5 hit, (4 hit + ws) and you can increase your DA%, do it, won't hurt at that point.

However if you are raising your DA% and losing your xhit to a lower tier, you have to start asking if that extra DA is worth it. In the case of war/sam you can use rag with 100% DA and by 6hit, but with 100% DA, that is only 3 attack rounds. (don't factor in a missed hit, there is a chance you will miss a hit on a 5hit build anyway.)

The goal is to get tp as fast as possible to the point of using tp at an appropriate level. In the event of most ws, you will want the 200% tp, or 175% tp with moonshade, 125% with aeonic/moonshade (although I hear it might actually be better to spam at 100% on aeonic)

200% (minus moonsahde) seems to be the sweet spot for dps with ws such as reso/CR, yes you can hold to 300% and spike higher ws, but 200 apparently is the sweet spot. Not something I have personally tested, just a long running rule I have been following for a few years now.

I definitely agree with the STP statement, but I don't think you're necessarily right about DA not diminishing. I'm almost certain it does, as Austar said.

QA procs before TA which procs before DA, making statements like 10 DA = 110 attacks per 100 kind of misleading as QA and TA are almost always present these days.

Also what is this about WS @200%? I wasn't aware that was better than spamming @100%

Edit: Spamming Reso @200%?

It can be really good on WS that are based on amount of TP.

DA gives diminishing returns in a percentile way, but not in a raw way.

For example, lets say you have 100 DPS with 0 DA. Going from 0 DA to 1 DA raises your DPS by ~1, a ~1% increase. Going from 50 DA to 51 DA still raises your DPS by ~1, but it's now only a ~0.67% increase.


As for WSing at different TPs, it depends on several factors, most significantly on the fTP of the WS and how fast you gain TP.

For example, in my default sets at capped haste and accuracy (including Moonshade Earring), Resolution is best at ~1500 w/ Gavialis and ~2250 w/o Gavialis (though it doesn't matter much for Resolution as long as you WS before 2750), Torcleaver is best at 1000, and Cross Reaper is best at ~2000 with Liberator and 1000 with Anguta.

Conversely, when I only have 70% haste, Resolution and Cross Reaper are both also best used at 1000 TP.

Faster TPing favors higher optimal TP, and higher multi-attack on WS favors lower optimal TP (for non-fTP transferring WSs only).

So what's the optimal amount of STP for getting to 2250 tp?

Edit: Ok this is how I'm finding the answer: 2250-TP after fisrt hit of Reso / number of hits. Lemme know if that's giving me the wrong answer.

For me, looks like I'd need 89 total STP for this with a base of 174 TP returned from WS.

It's not significant enough to actually gear for it. Just WS at TP close to that when you can.