PUG/MNK Optimization

[Cynric]

EasyMode is 100% correct on all of what was just said. FF 14 ARRs mechanics for certain things are a lot more forgiving than popular opinion would have you believe.

[Mishaela]

Maybe you should show us instead of calling people idiots?
Assuming higher than 2.0GCD and using True Strikes you will have a total of 6GCD between Twin Snake applications (Twin - Coeurl - Opo - True Strike - Coeurl - Opo - Twin). Your twin snakes buff will drop off before your second Twin - meaning you lose 10% on Twin Snakes and any possible autoattacks/off gcd attacks you might apply. If somehow this is not the case can you explain why?

What EasymodeX is describing is something like this:


You don't need 2.0 but you can't quite make it work with 2.2 either. 2.1 (or just under to pad for lag, button delay, etc.) is all you need.

[enil]

I'm aware of the animation delay and the fact that it checks at the start of the animation for buff refreshing (obvious to any who has seen GLx3 fall off and then come back with x3) - but has it been conclusively tested that it calculates for damage/buffs at the start of the animation as well?

I have not read anything - the easiest way to test this I think would be to pop B4B and use an ability and then try again as the buff is falling off.

edit: Also can you link your math to why True Strike is still a DPS gain?

Twin Snakes is 140 Potency. then multipled by 1.1 to simulate 10% damage
True Strike is equivalent of 150 Potency. then multiplied by 1.025 to simulate the extra 5% crit.

Given all other stats equal - I'm not sure which is the most accurate or accepted damage formula as of now but from my understanding the scaling works somewhat like this in regards to potency. (AP + Potency / 100) x WDMG and other modifiers. Since only the potency matters we only need to calculate the modifier for that specific part.

[(AP + 140)/100] x 1.1 (6.644)
[(AP + 150)/100] x 1.025 (6.2935)


My current Attack Power is 464 unbuffed.
I get a multiplier of 6.644 for Twin Snakes and 6.2935 for True Strike. This does not include the chance for an auto-attack which would greatly imbalance it towards the side for Twin Snakes.

[EasymodeX]

A Twin spam scenario with a 20% baseline crit results in 154 * 1.1 * 2 = 338.8p

A Twin+True scenario even if you flub the buff on Twin results in 165*1.125 + 154 = 339.6p

If you flub the True position as well, 165*1.1+154 = 335.5p
If you successfully get the Twin buff on Twin: 165*1.1 + 154*1.1 = 350.9
If you gain the True position and Twin buff: 165*1.125 + 154*1.1 = 355.0

The only case where the raw Twin and True numbers are lower is when you lose the buff and miss the True position. (-3.3p). However, even in that case you gain 10 TP.

10 TP to spend on Fracture pays for 10/[Fracture cost over normal average cost] or 10/[80-59] = 48% of a Fracture. 1 GCD of Fracture is worth about (110 + 120) * 1.1 [crt] over an average MNK ability rotation of 185p*1.1 -- e.g. 49p. 48% of 49p is 23.5p. 23.5 - 3.3 = 20.2p. Well now, that was a bigger gain from the TP savings than I thought.

So even if you flub the True position, and you drop the Twin buff for Twin, you still gain 20p every True Strike as long as you re-invest that TP into more Fractures (assuming you are able to invest TP into more Fractures; the result for adding Impulse Drives is not as good). E.g. this is the worst-case scenario and the TP efficiency you gain is still Worth™.

This also means that if you succeed in any of the other conditions, you're further ahead of damage output using True.


All of this analysis assumes that you're failing to maintain the Twin buff for the second Twin, which shouldn't happen if you use short cycles with moderate (+50) SS or better without lagging or disengages. So in general, you should not be landing the "worst-case" scenario.

All that said, you don't want to be throwing out True Strike all the time -- it's probably best to match it with Bootshine when you're "safe" for both, as a practical matter.

Edit:

I originally did an inaccurate value assessment of the 10TP savings. After revising it, the benefit is even stronger and also correlates with a simple "TP value" analysis. In my model, a standard Monk sequence has an approximate potency per TP of 5.0. Saving 10 TP on True Strike for "roughly the same damage" means you can spend 10 more TP elsewhere at 5p per -- so you gain 50 potency elsewhere. That's pretty big ... even if Fracture is terribly inefficient and (based on the earlier numbers) you only actually get ~20 potency, not 50.

[EasymodeX]

but has it been conclusively tested that it calculates for damage/buffs at the start of the animation as well?

(1) This was done re: DOT clipping testing.
(2) In order for Snap Punch to re-apply the GL3 (and not GL1) in the scenario you observed, it must read your buffs at the beginning of the animation (or rather, the end of the instantaneous spellcast), because you lost GL3 mid-animation. Sidenote: this is observed for AF3/UI3 stacks for BLMs as well, similar to the GL3 case.

[Size]

I have a question regarding impulse drive. I was using it in rotation along with bootshine, but as I've been off loading SS in place of Crit and Det, and when encountering bosses where it is difficult to reposition (like a spinning ADS) I've been resorting to flank only rotations using Frac. I also feel like I'm losing some time in repositioning, and wondering if this is made up by the booshine crit and impulsedrive spam.

I don't have access to a parser so I'm wondering if some of you more experienced monks (looking at you Easy) can say definitively if all that repositioning and impulse drive is actually worth the dps gain (if there is one), or if it is more efficient (especially TP wise) to just stay on flank and focus on Fracture and maximizing your DoTs and offensive CDs in between attacks.

[Mishaela]

I think it depends on the fight, too. Maybe it's because I'm at +101 S, but I have a hard enough time with TP as it is tossing out fracture a few times on turn 1 and 4 when I don't get a Paeon; even with Invigorate I usually have to slow down skill use and/or drop Fracture to not run out of TP. Turn 2 I never have TP problems--I have threat problems until rot, then after that since I pass rot across the room I can't even rely on having GL3 up full time so I typically don't have time for non-form skills even with TP excess.

[EasymodeX]

I also feel like I'm losing some time in repositioning, and wondering if this is made up by the booshine crit and impulsedrive spam.

1. If you ever delay your attack to reposition you are virtually always losing DPS. There are certain thresholds where this is not the case ("I lose only 0.00001s repositioning"), but for the vast majority of the time, it holds true. For example, by the time you notice your positioning is costing you GCD time, you're way past that point and should have pressed a button.

One general rule for a Monk is similar to a caster's ABC's (Always Be Casting): always be pressing an attack button. The positionals for Monks are moderately important, but they are not critically important. Always look to position your next attack, not your current one, because you should already be executing your current attack.

2a. If your positioning is completely random, Bootshine > DK (as long as the DK buff is up) because DK without flank is 100 and BS without rear is still 130.

2b. The mental attention required to analyze this condition will probably cost you more damage in the long run. Typically you just want to press DK again if you are uncertain on positioning. Each player manages this differently however*.

3. Impulse Drive has very limited use cases that are isolated to immediate direct burst damage or high DPS burn where the last 1% DPS matters to you, even at the cost of all your. For anything else, ID is not a good idea.

4. In a practical case, just wait until the fight is a little stable before you Bootshine.

if it is more efficient (especially TP wise)

ID is horribly inefficient from a TP perspective. In terms of TP, your basic priorities are:

1. Use a Monk purist set of attacks.

2. If you have excess TP to trade for damage, add Fractures with Demolish during cooldowns like BFB.

3. If you have more excess TP or need more DPS, add Fractures every Demolish.

4. If you still have more excess TP or you still need more DPS, add Impulse Drive.


* The way I run my Monk is pretty simple:

1. I attack with a priority of: SP TW DK.
2. Check for Demolish, if no DM / DM is about to expire, plan to use DM. If I am using Fracture, I tie the FR with DM.
3. Same for TOD.
4. Use oGCDs.
5. Compensate for mechanics.
6. If the fight is stable and DK debuff is >8s, plan to Bootshine-True.

If I have any uncertainty at any point in the fight, do #1.

Personally, the only time I'd ever use ID is for Twintania short conflags and I don't run my MNK on T5 so I don't have practice with ID. Anything else seems silly.

[Rochetm]

I'm losing some time in repositioning, and wondering if this is made up by the booshine crit and impulsedrive spam.

If you are using ID and not running out of TP you need to look at what you are doing. Your gear is either very low or you are not attacking as fast as you should be.

For positioning you should be moving as soon as you use your abilities. The flank/rear check is at the start of the animation. For big bosses don't use the lock-on strafe method that I see everyone using. Just run through the boss right to where you need to be (location) and hit the ability. The game will spin you around and hit the spot if you have the option on for it to do it.

For Mnk/Drg every fight is very different... you can't do the same things on every boss. Personally I think that is what makes mnk/drg so fun. In all content up through turn 4 there is only one target dummy like boss. Tonberry king stands perfectly still and ignores you but every other boss does something that is going to mess with you.

[Size]

Thank you both! Good advice.