**Dragoon Damage Formulae**

Damage/Weapon Skill Formulae

((Potency/100)*(WD/25+1)*(STR/9)*(DET/7290+1)*BUFFS)-1

Auto-Attack Formulae

((WD/3*aa_delay/34+1)*(STR/6.8)*(DET/6795+1)*BUFFS)-1

**Monk Damage Formulae**

((Potency/100)*(1+WD*0.0432544)*(STR*0.1027246)*(1+DET/7290)*BUFFS)-2

**Summoner Damage Formulae**

((Potency/100)*(1+MD*0.0394452)*(STR*0.1274670)*(1+DET/8079)*BUFFS)-3

**Other Job Damage Formulae**

N/A

**Critical Hit Rating Formulae**

Critical Hit Chance

((CRT-354)/(858*5))+0.05

Critical Hit Damage

((CRT-354)/(858*5))+1.45

**Skillspeed Formula**

Skillspeed on GCD:

=(ROUNDDOWN((ROUNDDOWN((1-ROUNDDOWN(((SS-354)*0.13/858),3))*(gcd_casttime),2))*(1-ssbuff),2))

Skillspeed on DoTs:

((SS-354)/6800+1) – Note, this fits into “BUFFS” for damage calculation]

**Parry Formula**

(((Parry-354)/(565*5))+0.1) * (1-block_rate) * (1-enemy_crit)

Base enemy crit appears to be 5%, therefore it would be *(1-0.5), or *0.95

### Like this:

Like Loading...

Is SMN actually calculated with STR or is that supposed to be INT in (STR*0.1274670)

LikeLike

Hey, is that SMN formula still correct for HW? Its off by a few 100 when i test ruin on a target dummy, or am I doing something wrong?

LikeLike

Could you post your data? I’ll have a look

LikeLike

Wow, you’re quick 😮

https://docs.google.com/spreadsheets/d/10IhLJH1I597PGbZaMJV0TALO6m1KknQI-aDNkUbWFZw/edit?usp=sharing

Thats how im calculating it, I just put the formula in a sheet, its probably me putting in the wrong values?

LikeLike

http://ffxiv.consolegameswiki.com/wiki/Maim_and_Mend_II

🙂

LikeLike

http://ffxiv.consolegameswiki.com/wiki/Maim_and_Mend_II

🙂 That’s your problem!

Though, it does look like the formula is over-valuing your non-crit values by a decent amount… Interesting.

If I used 1.2 as the modifier, it hits right on your numbers. Must’ve been an error on my side most probably.

EDIT: Yes, looking at it, I solved the formula with *1.2 and not *1.3… If you use 1.2 it’ll work. I’ll fix it for the actual *1.3 modifier later on.

LikeLike

Aaaah ok! I assumed that was already baked into the formula >.< my bad 😛

LikeLike

No problem mate 😀

LikeLike

[…] accurate since it was data-mined…) critical hit rate and damage formulae are posted here at https://dervyxiv.wordpress.com/damage-formulae-other-mechanics/, but I have noticed that he does not take into account at what point the game floors the value. I […]

LikeLike

Is the auto-attack formulae universal across all melee, or is it dragoon only?

LikeLike

Greeting!

I recently became interested in finding out how the game calculated everything (hopefully gonna try to create a dps simulator if I keep up on it :P) and decided to start with the japanese formulas. I found them to be closer to what I feel a game dev would do since they use one formula for everything and use tables to plug in various modifiers for stance, buffs, job, etc. After playing with the Auto Attack formula and testing it with a couple of classes thanks to the help of friends, I believe the formula to be correct (hoping to get a better test set when I finally level a mch to 60). However, I found it to be lacking when I tested it on my SCH. I found that after simplification the formula looks like:

(2*(((AA_POT/AA_DELAY*3)+(8.5))*0.00458)*STR*((DET-218)*0.000137+1.00)*STNC*(AA_DELAY/3.00*1.1))/2

When plugging this in and getting the min/max values I should expect, I found it to be off by 1. I also found that Cleric’s Stance did not change the values.

The interesting things this formula brings up is that MIND/INT mean nothing to SCH/AST/WHM/SMN/BLM when using auto attacks. I know that this is probably a useless thing to explore since magic people never use auto attacks, but I wanted to be sure that it was correct for all classes. For reference, the original AA formula was:

(2*((WD-(WD*IS_MAG)+(AA_POT/AA_D*3*IS_MAG)+JOB-(JOB*IS_MAG)+(IS_MAG*8.5))*0.00458)*((ATK-(ATK*IS_MAG)+(STR*IS_MAG))*ATK_BUFF)*((DET-218)*0.000137+1.00)*STNC*DMG_UP*DMG_UP2/RES_DWN*(AA_D/3.00*AA_EXISTS))/2

Thus I found that, for magical jobs, the JOB modifier is irrelevant, the ATK is irrelevant, the STNC should be irrelevant but is being said to make a difference, and I think the AA_EXISTS modifier is supposed to be 1.11, but I’m not sure. I was wondering if you could garner some insight into why the formula seems to break if I try to use a healer? Is it just some math error I made? Or is there an actual problem with the formula?

LikeLike

Is that from http://ku-so.hatenadiary.jp/?

If so, it looks like he’s updated his formula. My Dragoon formula is off by 1 damage values also. I’ve not checked for high potency values (in full i245), but it’s very accurate for the data-samples I tested with.

My function is exactly like his; we have a base coefficient which scales damage. Where has Arisu_Neetsha adds the base Job Coefficient to WD, mine is a derivative of their AP/MP values. The WD across all Jobs are roughly similar and the only drastic changes are to the AP/MP coefficient values.

I’ll be collecting a lot more data in 3.35 and rework on my formulas, but thus far, mine are pretty accurate for the data-samples I’ve received.

LikeLike

Are these the same formulas used in http://www.eonet.ne.jp/~versatile/ku-so/ff14dps.html?

LikeLike

Yes and no.

Arisue Neetsha is using my Determination Coefficient and my Critical Hit/Damage formulas. The rest of the formulas are independant.

LikeLike

Ok. Seems I have a lot to learn, but I’m looking forward to it 😀

Any advice for moving forward?

LikeLike

Yeah, best advice is to have a lot of time on your hands, haha.

LikeLike

Not sure if you’ve updated your comments on ss since this post https://www.reddit.com/r/ffxiv/comments/4jbscn/blm_stat_weights_simulation_methodology/d35hkp3 but it’s inaccurate to say you just take the base cast time and add .01 to the start of the ss formula. You have to actually apply the multiplier to both sides of the equation. I would propose the more accurate formula to be BaseCast * (1.004-((SS-334)/6602.5) where base cast is in seconds.

LikeLike

Also to note, haste buffs are calculated off the rounded value of the pre-buff cast time. ie 702 ss on raise ~ 7.727873 if the base cast was set to 6.4 seconds for PoM it would come out to ~6.182298 (displayed as 6.18 cast time), but it actually calculates as 7.72 * 0.8 which is ~6.176 (displayed as 6.17) This is of particular note to monks where moderate amounts of ss are a good thing for their rotations because it means fewer ss points between cut offs.

LikeLike

in the smn formula, where do the 0.0394452, 0.1274670, and 8079 come from?

LikeLike

Data points in a non-linear regression solver in Excel.

LikeLike

Hello! I’ve been following the trail of formulas and numbers you solved and am thoroughly impressed! I am relatively new the FFXIV math/theorycrafting scene and was looking into why EVERYONE is so inclined to hop on the “stat weight end-all be-all train.” I’ve come from games where you solved for stat breakpoints to be able to achieve optimal rotations and such, not “well this has the greatest weight so throw all my points here.” I’d like to hear your opinion or any math support you have as to why, let’s say for BLM, why someone would choose the ilvl 270 Alex pants (169Int,123Det,114SS) over the ilvl 260 Diabolic Pants (157Int,110Crit,123SS). Does the +6.6 stats outweigh the potential damage from +110 Crit (RNG forgiving)? Is Det not so bad as people are making it out to be?

Any input would be awesome and i hope to see more especially going into stormblood 🙂

LikeLike

You can use the formulas to check the increase in DPS. Let’s say we have a base potency of 100. Using the formulas on my page, (Using Dragoons, i’ll explain later), we know that every point of INT/STR/DEX/AP/whatever increases our damage by 1 + 0.1111 (1/9). The Critical Hit formula is =1+((((CRT-354)/(858*5))+0.05)*(((CRT-354)/(858*5))+0.45)), where X is your current CRT As an example, we’ll use a base of 700 CRT, and +110 crt for the other legs. Determination is simply expressed as 1+det/7290.

So for the i270 Alex Leg, we have 100 * (169/9) * (1+((((700-354)/(858*5))+0.05)*(((700-354)/(858*5))+0.45))) * (1+123/7290) = 2041.8456232

For the i260 leg, we’re doing 100 * (157/9) * (1+((((810-354)/(858*5))+0.05)*(((810-354)/(858*5))+0.45))) = 1896.11552453

There’s only a 9 SS difference between the two. This 9 SS will be impacted based on what your current SS is, as SS is tiered. If that 9 SS will make you go up a tier in your casts, you’ll see a DPS increase. If it doesn’t, no DPS increase. Nonetheless, 9 SS will not make up the difference of 145.73009867 raw damage per 100 potency.

I used the Dragoon AP/STR scalar (AP/9) as people have reported that the Dragoon formula works much better for the BLM, than the actual BLM formula I have, which makes 0 sense.

LikeLike

Interesting. Maybe you could help me calculate something? As of now, I can cast five consecutive Fire IV while under the influence of Ley Lines and four without. How much spell speed would I need to push six consecutive Fire IV with Ley Lines (and/or five without)? My spell speed (on one gear set anyway) is currently 1305, and what I’ve come up with so far is subtracting the cast time (2.18 with LL) for the first Fire IV from the Astral Fire buff (12 – 2.18 = 9.82) and then dividing 9.82 by the recast time (1.81), resulting in 5.42 consecutive Fire IV (which makes no real difference in that regard compared to my other set with 1070 spell speed and 5.15 consecutive Fire IV, and even though I know there’s a 235 point difference and you should subtract the base 354 when calculating what difference another point does, I still can’t seem to figure this out.) What bugs me is that I can’t seem to find out why the spell speed shaves off a different number from the cast time compared to the recast time, so I have no idea how to solve this.

LikeLike

Use the SS formula provided on the formulas page. The cast and recast times are different and scale differently.

LikeLike

I’m sorry but I can’t understand the new SS Formula.

Could you please explain it?

Why there’s 0.85 and where the 2.51 should be placed?

LikeLike

I copy pasted the wrong formula, sorry. I updated it.

And you don’t use 2.51, you use the base cast/recast/gcd. So it’ll be 2.5.

LikeLike