?

Log in

Or connect using:

FFXI damage equations and their implications explained

« previous entry | next entry »
Jul. 30th, 2007 | 07:30 am

FFXI damage equations and their implications explained in simple (!) but accurate terms

This post is aimed at explaining some more advanced concepts on how damage works in FFXI, so that everyone (and monks especially) can more easily understand how to make the right choices to maximize their potential.
I spent a lot of effort in trying to make this post as easy to understand as possible, by working out examples for different situations, since people have a tendency to have shorter attention span when just formulas and terms like pDIF are thrown around.

To start, for any setup choice you really want to hang on to and stay behind, you really must ask yourself and answer two very important questions:
-1) Why is this better and, just as importantly,
-2) by how much exactly is it better.
If your answer to [1] is just that you heard from a bunch of people that the alternative is "gimp", or are easily convinced by persuasive reasons such as "because it's LOL" or
any other similar appeals to people's fear of being ridiculed and desire to conform with the group, you might want to consider growing a backbone and realize you actually don't know any real reason for holding that particular belief or conviction, and should really look into finding actual reasons like "it gives a damage increase of around X% over this alternative in this situation and increasing damage is one of my goals", etc. Too much crap happens because people just follow in with a sheep mentality and don't bother questioning their reasons or looking for facts or details. Anyways... back on topic..

If your answer to [2] is "it depends", you are not trying hard enough. You should be able to at least think about the best case and worst case scenarios so for example you don't spend months of over-opinionated verbal arguments trying to s*#!-talk your friends over a less than 1% improvement...

This post is about how you answer those questions now that we have all the game damage equations reverse-engineered to an art from the huge data collecting effort over the past 5 years of StudioGobli etc, but explaining in simple clear terms all the consequences so everyone can have a clear understanding of what those equations actually mean.

There are also a few FFXI damage calculators around, and I definitely recommend people use those and not necessarily go through the tedious process of working out every situation by hand, but I am spending time here explaining the implications of these damage equations which might not be so obvious and which are often misunderstood. There are too many wrong rules of thumb phrases misused all the time, like "ATK makes the curve more consistent and STR raises it up (wrong)", or "ATK > STR > DEX for WS (wrong)" which have nothing to do with the actual reality of how the game works. I had to explain why STR+4 will do more than ATK+5 on a ws for example, which is something that anybody could realize by themselves if they ever did the math instead of regurgitating one-liners they heard somewhere and took as truth without ever checking.
If you disagreed with any point mentioned in this topic, by all means try to find a situation that acts as a counterexample, we would all learn a lot more and be more easily persuaded if people actually had some numbers and real arguments to back them up.

[1]. Damage equations basics: fixed and variable range components
I have heard some people say things like "damage in this game is random, so I can believe what I want and from my experience math doesn't matter"... That's a big mistake. The damage in FFXI is accurately computed based on your stats, and then the game draws a very specific range around this and distributes random values within it, so you can tell what your max and min damage are and even estimate your average, all of which depend on your stats.
Believing something like "ATK makes numbers less random"(?!) for example, as I heard people say before, shows some people just don't understanding how damage works in this game.
Your normal melee damage is simply:
meleeDamage = baseDamage x pDIF
The baseDamage number is a simple fixed number mostly dependent on your STR and weapon. The pDIF number is a randomized number that falls within a max and min which are determined mostly by your ATK, plus a correction for level difference to the monster you are attacking.
It's important to realize that the randomness added is very specific, so while looking at a few attacks might seem like the damage is arbitrary and unpredictable, the effect of specific stats is always the same and is in fact very predictable.

To give you an idea of the relative magnitudes, the highest value that pDIF can take is 2.4, and usually the range between min and max is exactly 0.8, 0.5 or less. Critical hits damage is computed just by adding 1 to the normal pDIF value (and change the cap to 3), which proportionally is a very large increase and one of the reasons why critical hits are so good.
Let's look at an example:

Using some mock values to illustrate this before getting into complex cases and how these values are calculated, at 75 with destroyers and merits and a reasonable amount of STR and ATK, say 80 and 300 to make things round, your base damage against a lvl75 monster with 80 VIT and 300 DEF would be 52 (50 from your weapon choice and 2 from the fact that your STR matches the monster's VIT) and your pDIF would be between 0.7 and 1.2.

So your attacks would fall between 52*0.7 = 36 and 52*1.2 = 62. In this same situation your critical hits would fall between 88 and 114 (52*1.7 and 52*2.2).
Try this:
To get a general feel for the increase you can expect from adding a certain stats, since people have endless flame-wars over atk food or certain gear, try guessing the answer to the next question:

"How much will adding 5 ATK, 10 ATK, 5 STR, 10 STR add to these damage ranges?"

Answer:
Base[36, 62]
+5ATK[37, 63]
+5STR [37, 64]
+10ATK [38, 64]
+10STR [38, 65]

There is nothing hard about this, but what makes things complicated and tedious is that these two numbers are tiered (divided into regions or levels), capped in certain ranges, modified in others, and corrected for level differences, which also results in a lot of unintuitive effects where adding stats might have no effect until a certain threshold, and then create a sudden jump in damage at that point. Understanding this and how to always reach these thresholds optimally can increase your performance significantly and save you from making very wrong assumptions on the relative importance of stats and armor.
I will start with a basic explanation and slowly add complexities, exceptions and their implications.
[2]. The fixed part of your damage: Base damage and fSTR

For the most part, you only need to look at a couple of stats. Your accuracy will determine how often you will actually hit, and I'll discuss that in the next post. Your DEX and critical hit rate will determine how often you will hit as a critical hit and also comes later in the next post.
Aside for those the other main stats that determine your damage are your weapon damage rating, your STR, your ATK, and your LVL vs the monster's VIT, DEF, and LVL.
As mentioned above your melee damage is a product of two numbers, the first being the fixed baseDamage.
BaseDamage = WeaponRating + fSTR
The first parameter is your weapon's damage (see insert below) and it's usually the larger contributor (at 75 it's 50 with destroyers and merits and 32 with bare hands).
The other parameter is basically a fraction of the difference between your STR and the monster's VIT (to see the relative contribution, at 75 this number can never be higher than 13 while we saw the weaponRanking was 50).
Note that ATK and DEF here have no effect at all on this base damage (they will come into play later on with ranges and pDIF).
So your base damage is ultimately a function of only 3 things: weapon damage, STR, and the monster's VIT.
The game makes things a bit more complex, by tiering things, so instead of taking the straight number of (STR-VIT), they make a parameter people called fSTR which is slightly altered at different ranges and is divided by 4 and truncated, so that it can take around 4+ STR to raise fSTR by 1. This value is TRUNCATED, which means that if your fSTR just raised by 1, and then you add 1,2,3 STR to it, you will see no additional effect! (aside from the side increase in ATK which will come into play later and which might also give no immediate effect). Adding that extra 1 STR after that however will make fSTR suddenly jump up. That's why sometimes adding a little STR seemed useless and other times it might have made a bigger difference. It is a bit difficult to control when the jump happens for this because you don't have control over the monster's stats however. This is one area where WS modifiers have an advantage as you can always predict it exactly for those.
Insert 1: Calculating Weapon Damage Rating:
I will show how to compute your weapon's basic damage in more detail here, since for monk it depends on both your hand-to-hand skill and weapon rating.

In more details, to compute your weapon rating take your H2H skill, multiply by 0.1, add 3, then add your weapon's +damage, so for destroyers that is +18, and you have your weapon damage rating.

For quick reference, the base weapon damage of a fully merited monk with destroyers is 50.
(276+16)*0.1 + 3 + 18 = 50
NOTE: It turns out that SE made things a little more complicated specifically for H2H... It basically computes weapon rank, weapon damage ranking and the max cap on fSTR differently than with all other weapons. SO if you are using a damage calculator and can't specify you are using H2H, use the formula above and 50 for destroyers as a good approximation. However this will have slight small differeces in a few combinations of H2H skill and weapon +damage because the actual formula and values are H2H*0.11+D, so for destroyers this would be 53. If you use these numbers however remember that you can't use current damage calculators because you also have to adjust fSTR max accordingly, where weapon rank is only based on the +D part without H2H skill contributions (so for example it's 18/9=2 instead of (292/10+18+3)/9=5 for destroyers, and the max fSTR will be 10 instead of 13).
Note however that for things like computing the max critical damage with destroyers and max skill these two formulas result in the same value of 189. One place where you can see the difference is with the predicted value of a D+5 weapon vs D+6 weapon, which is off by 3 if you use the studioGobli formula.
Here are two screenshots as a sample test demonstrating where this slight adjustment is more correct than the normal damage equations with D+5 and D+6 (if H2H and fSTR max were computed like other weapons you would expect a max crit of 147 for D+5 and 150 for D+6, instead you get 144 and 150, which are predicted correctly by the above adjustments):



Insert 2: fSTR tiers table

fSTR is also tiered, just like DEX/criticals and so many other things in the game so if you fall within certain ranges you have to basically use a look up table to find its value:

(STR-mVIT)fSTR2 approximation
12 ~ ...((STR-mVIT) + 4) / 4
6 ~ 11((STR-mVIT) + 6) / 4
1 ~ 5((STR-mVIT) + 7) / 4
-2 ~ 0((STR-mVIT) + 8) / 4
-7 ~ -3((STR-mVIT) + 9) / 4
-15 ~ -8((STR-mVIT) + 10) / 4
-21 ~ -16((STR-mVIT) + 12) / 4
-... ~ -22((STR-mVIT) + 13) / 4
On top of that, the game also caps several things, you probably noticed that your total damage potential is capped if you ever tried to punch a level 5 monster vs a level 40 monster at 75 and did exactly the same damage. fSTR itself is also capped according to your weapon rating (take your weapon's damage of 50 and divide by 9, then truncate and you get 5, your higher cap on fSTR is that number+8, so 13, and the lower bound cap is -5).
The proportional increase is a bit less at higher levels because fSTR is relatively small compared to the higher weapon damage rating, but still generally more impactful than ATK as we will see later.
Summary of 2: impact of STR:
So, the summary points here are:
1) the fixed part of your damage is determined only by STR and weapon (with weapon being the largest contribution), not by ATK
2) it can take sometimes just 1 STR and sometimes 4+ STR to make your base damage jump up, because of truncation.
3) your base damage is further tiered and capped at different ranges.
4) The bonus from either stat is not really all that large for small increases
Intermission 1: Soloing Jelly NMs

If by now you are bored to death with all these numbers, here's a side story.
I went out and solo a NM I didn't get around killing yet. As we know jelly are difficult monsters for monks, and this jelly NM had even more melee resistance, so I wanted to try soloing it, plus I didn't get around getting my Jelly Ring yet (phys.dmg-5%).
It was a very nice experience as nobody else was camping it (the drop is rare ex) and the pop conditions for this one are not time-based, you just have to go there and kill enough slimes until it pops. The slimes and ghosts in the area are only EP-DC to 75 so I had no problem getting several stacks of slime oils and I finally popped the Sewer Syrup NM. This guy is described as very hard for melee because it's very resistant to melee damage and has high HP, over 10k, and it's instead very weak to magic. Still I thought it was worth trying, and I decided to fight it as MNK/RDM instead of the usual nin sub to make it more interesting. As you can see it went very well, it took a while (about 8min), but I didn't even use my 2hr or any items for it. So for the books, this NM is soloable by monk.

Formless strikes was not quite as useful on this because of his very high HP and the long 10 min timer on it, but it did about 2-3x more damage while it lasted:

Also evasion and defensive stats from gear like osode are always so nice:

[3]. Damage ranges and pDIF

Here is where the randomness and your ATK vs monster's DEF come into play.
After you determined your base damage, you just have to multiply it by a number called pDIF to get your final damage for each punch. This number is a function of only 3 things: your ATK, the monster's DEF, and your level difference.
To get this you just compute ATK/DEF, and get what people call cRatio. Of course the game didn't want your damage to go unchecked and get ridiculously high or not scale well with monsters and different situations, so this is again tiered and capped. The interesting point however is that it's not quite truncated (well it might be truncated to 3 decimals but still allows you to see very small differences).
cRatio is capped at 2, so if your ATK is ever double the monster's DEF, adding more ATK will have absolutely ZERO effect. You can refer back to some of my parses on farming monsters with and without ATK food if you need extra convincing on this.
cRatio is further adjusted by the level difference between your LVL and that of the monster. a 0.05 is subtracted to the ratio for each level difference (or added if your level is above).
Now, pDIF is just a function of cRatio , and is a random number between two bounds (min and max pDIF), and the two bounds are computed specifically from cRatio multiplied by some number plus some constant.

Insert 3: pDIF max and min table:
As I mentioned, pDIF is a function of cRatio. However, the game likes to make things a bit more complicated and as I mentioned pDIF is tiered, so there are ranges of cRatio where either the upper bound or the lower bound are not related to your ATK at all but are fixed to a specific value, and the number that is added and multiplied to cRatio is different for different ranges of cRatio.For pDIF max and min, you have 3 tiers or ranges:
cRatioMax pDIFcRatioMin pDIF
0 = cRatio < 0.5pDIF(Max)= 0.4+ 1.2xcRatio0 = cRatio < 1.25 pDIF(Min)-0.5 + 1.2 x cRatio
0.5 = cRatio = 5/6 (8.3)pDIF(Max)= 1
5/6 < cRatio = 2 pDIF(Max)= 1.2 x cRatio1.25 = cRatio = 1.5pDIF(Min)= 1
1.5 < cRatio = 2pDIF(Min)= -0.8 + 1.2 x cRatio
So for example, if you ATK is 300 and the monster's DEF is 400, if the monster is at the same level as you in this example, your cRatio is .75, so you are in the middle range for pDIF(max), and in the first range for pDIF(min).
That means that your max pDIF and therefore your max damage will not be affected at all by adding more ATK, until you add 34 ATK, at which point each ATK you add will increase your pDIF according to 1.2xcRatio, so going from even 200 ATK to 333 ATK (an increase of 130 ATK!) will not change your max damage at all , while adding even 10 atk from 334 ATK to 344 ATK for example will increase your pDIF max again, from 1 to 1.03.
Of course in that range pDIF min will still increase, raising your minimum damage, so adding attack in that range will still raise your average damage, but you can think of it happening twice as slower as in a range where both max and min are uncapped.

The fact that increases in ATK only show up in your damage as a ratio to the monster's DEF also makes it a lot less linear or intuitive. The same increase in ATK, say ATK+20, is a lot more significant at low levels than at higher levels (for example, take ATK=100 DEF=100 vs ATK=300 DEF=300, adding 20 to each case you have a cRatio change of 20% vs 6%). That's one reason why people who have recently leveled up monk through low levels tend to get an over-inflated expectation of ATK compared to what it actually does at 75.
One other important consideration about pDIF is that critical hits raise pDIF by a whole point, which is a LOT, and one of the reasons why critical hits form merits and Destroyers, and stats like DEX are so important. But I will talk more about critical hits in the next post.
Summary of 3: impact of ATK:
So, the summary points here are:
1) the effect of ATK is smaller at higher levels
2) in many ranges any increase in ATK will have no effect at all or half the optimal effect, because both of capping and tiering effects.
3) In general just adding a couple of points of attack will not be enough to raise your melee damage by even a single point (it might take 4-6 atk to raise your damage by 1).
4) In contrast, critical hits raise your pDIF by a whole point, which can sometimes be 2 to 8 times your normal damage (especially on high level NMs).
Intermission 2: The new expansion
By now I am sure most people watched it, but here are the two trailers for the new expansion:

[4]. WeaponSkills damage and WS modifiers:
If you ever hear anybody say something stupid like "modifiers are LOL", that's another person that doesn't understand how damage works in FFXI, and prefers to mindlessly repeat words he heard somewhere without any reason instead of trying to understand how thingsd work. That's especially true for monks, because we have very low modifiers for our best WS, the bulk of our damage comes from non-ws damage (usually 30% is from ws and the rest is from melee DoT). and we have some WS modifiers like VIT that are not otherwise offensive stats, so there is some reason to ask if and when modifiers are useful.
That however doesn't mean that it's ok to be ignorant of how modifiers work or that they are less important than other stats on WS, especiially considering that you can use macros to swap gear on WS and that modifiers damage increase never caps while other stats can reach a point where adding more has no effect at all on damage.
The damage from WS is easy to compute, it's basically the same as that for normal hits, plus a few extra terms:
WSDamage = ((BaseDamage + WSC) x fTP) x pDIF
For multiple-hit WS you just multiply this by the number of hits you will do. WSC is the famous WS modifier.
fTP is always 1 for asuran fists, but it increases with TP for Dragon Kick, Howling fist, Raging Fists and Combo, although in most situations you are better off doing 2 WS instead of waiting for 200 or 300TP because for example fTP usually goes from 1 to only 1.5 from 100 to 200 TP. Notably Dragon Kick starts at fTP=2 for 100TP.
Anyways, since everything else is the same, and I already covered fTP, the only term left is WSC , the WS modifier. As you know that's farily high for dragon kick, 50% STR and VIT, and only 10% STR and VIT for asuran which is our favorite WS because of its high consistent damage.

Side Discussion 4: Effect of stats on AsuranFists WS:
Let's consider the case that you are near a jump in damage, and you are considering if it's worth wearing an extra 1VIT and 1STR.

The figure on the right shows that situation in two scenarios, with ATK in its ideal growth range and with attack maxed with respect to the monster's def (mDEF=200).

Your max asuran fists damage with all 8 hits connecting will be 1189. Now let's assume we wear an extra 1 STR and 1 VIT. Your asuran fists damage jumps up to 1248, an increase of almost 60 points (~5%)on your max WS damage. So again, when you are on the edge of a transition point you can gain a lot from adding very little thanks to modifiers.

These are also very nice because they don't depend on enemy's stats, only on yours, so you always know where the transition points are, and they are not capped so they can continue to increase your WS damage whereas ATK or STR can't.
A few Examples of WS Damage Increase:

Different asuran fists values (using ATK= 400,STR=79,VIT=79, mDEF=250, mVIT= 80):

base 555~952
+1 STR, VIT 581~997
+15 ATK 590~987
+11STR, +1VIT 621~1057

usingmDEF= 200
(atk caps):
base 793~1189
+15 ATK793~1189
+300 ATK793~1189
+1 STR, VIT 830~1248
+11 STR, 1VIT 869~1305
In contrast, in the mDEF=200 situation the effect of ATK was already capped (since the enemy's DEF was half our ATK), so adding even 600 ATK will not change damage at all (you can actually test this with boosting to 999 ATK btw).

One of the most frequent misunderstandings that people have is that if you have a 10% modifier, each point of that stat you add will add 0.1 to the extra modifier damage. That's not how it works however because of truncation. Instead, adding a point of that stat will either add nothing at all or add a full point. That's the same as a 90% modifier by the way. The difference is that for a 90% modifier you are pretty much always justified in adding more, especially a lot more, but for a 10% modifier it only makes sense to add more if you are right at the next transition point, where adding one will make the truncated value jump up, while in other situations adding a lot more would still get truncated down to the same value so it wouldn't make sense to add it.
In addition, since STR is used for the fSTR term too, something like STR+4 for example will always be better than ATK+5 (that was a question that was raised at some point on macro-swaps for WS) even if it didn't help break one of those modifier thresholds (unless it comes with added penalties like -DEX which would hurt your accuracy and criticals).

Another important misconception is that since asuran fists is 8 hits, you need MORE accuracy than normal.
I will discuss accuracy fully in the next post, but I really have to mention this because it's a very wrong assumption. Accuracy is the same and has the same caps regardless of how you group your hits. If your accuracy was only 50%, you should be equally worried whether you are using a 1hit WS or a 8hit WS. In fact if you care about skillchains you would be more worried about it in the single-hit WS since you might actually miss the whole WS instead of just some hits.
For the same reason you also can't gain any advantage from adding more accuracy when your melee accuracy is already at its cap of 95%, there is no reason to swap in more accuracy for WS at that point, because it will have no effect. Each hit will still be 95% chance of landing, and you will likely get a perfect full-TP return asuran about two thirds of the time.
Summary of 4: impact of ATK, STR, VIT on WS like AsuranFists:
So the summary points here are:
1) WS damage is almost the same as normal melee damage, with the addition of modifiers to the baseDamage.
2) Even on AsuranFists, adding just 1 point to a modifier stat can have a large impact on damage when near a threshold point
3) If you are far from a threshold point it might not be worth trying to add too much of a stat (aside maybe for STR which will also raise your fSTR in sufficient amounts)
4) Similarly, if you are in a region where damage contribution from your ATK is capped at either or both ends (e.g. bards/farming), adding more ATK will have no effect.
5) AsuranFists does not require any more accuracy than any other WS. If your melee accuracy is capped (95%), you don't need any more accuracy for AsuranFists and will gain no benefit at all from it.
In the next post:
I already made this post too long, so I will leave the following topics for the next post(s):

Additional Topic 1 (next post): Accuracy and evasion
Additional Topic 2 (next post): Weapon Delay and damage ratio: faster weapons vs higher damage weapons
Additional Topic 3 (next post): Critical damage and tiered DEX effects
Additional Topic 4 (next post): Computing Kick attacks damage
Additional Topic 5 (next post): Computing Chi Blast damage
Additional Topic 6 (next post): Soloing Avatars (videos?)
Additional Topic 7 (next post): Extra damage bonuses on monster weaknesses and weapons types.
Additional Topic 8 (next post): Adding kick attacks parsing to your parser.

Link | Click HERE to leave a COMMENT (no LJ needed, FFXI anon posters please write name and server) | Share

Comments {30}

talifey

(no subject)

from: talifey
date: Jul. 29th, 2007 11:29 pm (UTC)
Link

Very interesting post. Are there any sites with all of this documented besides that JP one you posted?

I'm really looking forward to your next post on "Weapon Delay and damage ratio: faster weapons vs higher damage weapons" since I played thief I always felt there was something unfair about low dmg low delay weapons (though every drk also complains about high delay high damage which also makes sense because of lost swings at death) but against HNM seemed like low delay was hugely disadvantageous.

Also I hope when you do your critical analysis you comment on sneak/trick, while I know you're a monk I'm sure you went through the mnk/thf stage with dragon kick and I would find anything on that incredible interesting!

Good work :D!

Reply | Thread

genomeffxi

(no subject)

from: genomeffxi
date: Jul. 29th, 2007 11:41 pm (UTC)
Link

Thanks. All the other sites with the same equations documented are basically translated and transcribed from Studio Gobli.
A good translation is at http://www.freewebs.com/vzx-01

About SA/TA maybe I can add a little about it but usually as monk those are not very common concerns aside from very specific situations.

I know people complain about high delay weapons like gaxe or gkt, but then again it goes a little both ways because it's annoying to do a WS like asuran at the same time as someone else and only get 1 hit before the monster dies, while someone who does a big TachiGekko gets the full 1k damage parsed =P
SE says they are still planning on doing something to make single-wielders happier, like allowing you to wear something with stats on the other hand anyways, I hope they include h2h in that too.

Reply | Parent | Thread

Funny

from: anonymous
date: Aug. 14th, 2007 04:00 pm (UTC)
Link

Skitso again,

It's kind of funny that you predicted this. But it seems they're only including 2 handed weapons in this adjustment. Too bad eh. Do you have any thoughts as to whether they'll do something similar with h2h?

Thanks for the parser link. I forgot you'd probably not want to disclose your email either, duh.

Reply | Parent | Thread

(no subject)

from: zerozephr
date: Jul. 30th, 2007 04:58 am (UTC)
Link

Good to see you posting again!

Completely unrelated to the post, but how's life on Diabolos? I'm thinking of migrating since Ragnarok's end game scene really stinks.

Reply | Thread

genomeffxi

(no subject)

from: genomeffxi
date: Jul. 30th, 2007 05:20 am (UTC)
Link

Thanks.
Eh I wouldn't recommend it here =P
I might be wrong of course because I've been pretty detached from LSs for such a long time and just playing on my own schedule but from the little I can tell, there is not much to pick from.
There are a lot of young LSs and inexperienced LSs which might be good for entry-level people, but I don't even know of many people doing new activities like salvage or einherjar consistently and successfully. Even older LSs have been plagued by a lot of drama and a lot of revolving new players or weird or biased distribution schemes.
Plus it costs real $$$ to move. If you really want to move you might want to make a mule a few times in a couple of worlds first or see if you can't make the changes you want in your current world.

Reply | Parent | Thread

awsome

from: anonymous
date: Jul. 30th, 2007 09:53 am (UTC)
Link

love to read all of your posts, can't wait for the next one. ^^

Reply | Thread

genomeffxi

Re: awsome

from: genomeffxi
date: Jul. 31st, 2007 08:32 am (UTC)
Link

Thanks :)
it might take a couple week for the next post though =P

Reply | Parent | Thread

DeAnna

(no subject)

from: lurkakitty
date: Jul. 30th, 2007 07:11 pm (UTC)
Link

Good to see ya posting again, stranger =P

Reply | Thread

genomeffxi

(no subject)

from: genomeffxi
date: Jul. 30th, 2007 08:55 pm (UTC)
Link

Thanks :)
Although you probably won't find my posts very interesting now that you play a different game :p

Reply | Parent | Thread

DeAnna

(no subject)

from: lurkakitty
date: Jul. 31st, 2007 12:49 pm (UTC)
Link

Hey now, I still read and even post on the ffxiljcommunity :P

Reply | Parent | Thread

(no subject)

from: anonymous
date: Aug. 3rd, 2007 02:58 am (UTC)
Link

great read as always, keep up the excellent job.
cant wait for next one.

Reply | Thread

Accuracy

from: anonymous
date: Aug. 3rd, 2007 04:15 pm (UTC)
Link

I've read quite a bit of the things you've posted here, including gear suggestions and this bit. I'm curious as to how much accuracy gear I should actually have.

As it stands right now, the only acc+ I have on is a Chiv. Chain, and the additional accuracy from having capped hand-to-hand merits. Most things, aside from mamool, either show up as low evasion, or no message. The problem I'm having comes from the exceptions, mostly merit party things. Before I spend a lot of time and money on gear that I may end up not actually needing, I was wondering what you would suggest.

I'm aware that getting Byakko's Haidate would give me a decent accuracy boost (from the DEX+15), but I'm not certain if that's enough. I was considering changing my Chiv. Chain out for a PCC, and getting 2x Snipers. Most of my gear, aside from Limbus gear, falls in line with your gear topic. Osode, Destroyers, Okote, etc.

Anyway, sorry for the long-winded question, I would just like to know if I NEED to replace my gear, or if I'm doing fine.

Thanatosmazus, Asura Server

Reply | Thread

genomeffxi

Re: Accuracy

from: genomeffxi
date: Aug. 3rd, 2007 06:23 pm (UTC)
Link

well as I said I will do a post on calculating accuracy etc later, but you can expect that switching to p.charm from c.chain will increase your hit% by 2.5% (unless you are near the cap of 95% hit rate already).
The same is true for a sniper's ring (unless you are switching out a rajas which I wouldn't recommend and would only give you 1% more).
If you want to know also how that will feel you can try to get similar effects from focus (focus with af1 and then switching back your ohat can give you 10% more hit rate).
Check what your meleeHit% is in general and on the high-eva monsters in merit PTs specifically (keep in mind some THF monsters etc just need more acc than others even in the same camp). After that you can better decide what you feel comfortable with in terms of hit% and decide if you want to change gear, use more accuracy food, etc.

Reply | Parent | Thread

genomeffxi

(no subject)

from: genomeffxi
date: Aug. 6th, 2007 09:43 pm (UTC)
Link

I am replying by quoting your text because I am not sure you wanted others to know your email, so I am omitting it (you can let me know and I will just turn on your post to visible).


Your question was:

I was wondering you said 5 str base i better then 5 attack base
but dont 2 str= 1 attack
so 5 str would be 2.5 attack correct?
and str would maximized your full cap on ur attack base??
when attack won't correct???
let me know what you think about that.. not sure i'am just asking....
just that when i had like 10-16 stats to all... i felt like i was weaker
at soloing then my over all stats but then again i might be wrong cuz i 'am a RDM where i should be focusing on int and mind and enfeebling to be able to gravity debuff and melle... oO XD then again i'am a crazy RDM/RNG XD
Gukswar IFRIT SERVER Oo tell me what you think of it o_O.... i won't say your wrong because i'am not positive about DAMMAGE calculations o_O...
I'am a good range dd tho XD as rdm buffer debuffer

The answer is yes 2 STR= 1 atk, which is why when comparing small amounts of the two usually the base stats like STR win over.
For example with 5STR vs 5ATK you are really comparing 5STR vs 2.5ATK

About ATK capping, that definitely can happen, but mostly when you are fighting easy monsters (farming materials on too weak or low EP) or have a lot of buffs from bards (most likely not in your case as RDM). Increasing STR however will NOT change this capped situation, that cap on ATK is only based on your ATK vs the monster's DEF. I think as a RDM you probably won't be in a capped ATK situation unless you are fighting too-weak for farming.

Aside from that, that's fine to go around and have fun soloing with strange combos like your RDM/RNG. I think you can experiment with interesting things like holy bolts which I think do higher damage when your MND is high (so should be good for RDM/RNG maybe?)
But as far as meleeing RDM has low offensive melee stats, low attack and accuracy also from low skill rating on weapons so you might need ATK the most and you are not likely to risk capping it much, although if you are really just deciding between 5STR and 5ATK you would want to go with 5STR.

For monk we often have to choose between more unbalanced choices, like 6STR vs 22 ATK, where it makes more sense to pick the larger ATK boost often and it is more important to know where your ATK might be capped to switch gear appropriately.

Reply | Thread

A good word

from: anonymous
date: Aug. 7th, 2007 03:25 am (UTC)
Link

Cobhc (Monk)
Hades

I've been reading your work for a few months now and i just wanted to say, great work man, truly amazing stuff

Reply | Thread

genomeffxi

(no subject)

from: genomeffxi
date: Aug. 13th, 2007 07:13 pm (UTC)
Link

For Skitso, there's a good parser I know of that now is even open source (so people can program new features in it), called dvsparse, it's at http://sourceforge.net/project/showfiles.php?group_id=197684

Reply | Thread

[2]. The fixed part of your damage: Base damage and fSTR

from: anonymous
date: Aug. 13th, 2007 08:42 pm (UTC)
Link

You mention two things:

"monk with destroyers is 50. (276+16)*0.1 + 3 + 18 = 50"

This makes sense to me, the equation works out to 50.2, and rounded down it would be 50.

Then you mention that the equation is different for H2H, H2H*0.11+D.

"H2H*0.11+D, so for destroyers this would be 53."


This does not make sense to me. I assume that D stands for the Damage+ on the weapon, so 18, but 292*0.11+18 = (aprox.) 50 not 53. 292*0.11+18+3 = 53, the value you give. Is this a typo in your writing, or am I misinterpreting "D"?

This is my first time trying to tackle any conception of Damage Equations and I'm still having trouble trying to figure out the base equation without truncation etc. so I'm just hoping for some clarification.


Reply | Thread

genomeffxi

Re: [2]. The fixed part of your damage: Base damage and fSTR

from: genomeffxi
date: Aug. 13th, 2007 09:43 pm (UTC)
Link

Yes so it is actually confusing, more so for H2H.
The right equation is the one that adds up to 53, with the +3, because even without skill (think of when you first start the game and have 0 H2H and bare fists and still do damage), you have a natural base +3 damage even without weapons.
The confusing part is that the damage of destroyers is in fact 53, but if you used that in a damage calculator or tried to compare to other weapons you would only get the right number for 50 and not 53, that's why for a long time people thought it was 50 and not 53. Basically destroyers is 53 but it caps at the same damage as a weapon with 50 damage (think swords etc). That's because the rank of destroyers is just 2, not 5, because the rank is computed off of just the 18+3, as a static value (18+3)/9=2 rank, instead of using the weapon damage rating of 53 to compute rank as other normal swords etc do (53/9=5 rank). The numbers for max damage of 189 per punch happen to be the same because that 3 difference between 53 and 50 happens to balance out to the 3 difference between rank2 and rank5. In general you can use 50 in damage calculators as a decent approximation, but the correct value is really 53 with a lower cap (equivalent to a 50 weapon) due to the rank from the static value on the weapon that doesn't consider H2H in the rank (maybe it was easier to code).
It was sort of an afterthought in the post, and didn't really change all the examples much so I left it as a side discussion to not complicate things too much. For most comparisons or understanding the effect of adding X STR or Y atk it doesn't really matter much anyways.

Reply | Parent | Thread

Re: [2]. The fixed part of your damage: Base damage and fSTR

from: anonymous
date: Aug. 14th, 2007 05:09 pm (UTC)
Link

ive been reading a lot of your posts, and i just have to say that its amazing that someone is actually testing all this stuff.. im really glad that your not just some one mindless zombie who just agrees with "common knowledge" but doesn't really know anything.. for that i respect you. i just got mnk to 75, not fully merited, and this post helped me out a lot, so thanks for testing, and posting results =] i cant wait till your next one.

Reply | Parent | Thread

Awesome

from: ajinin
date: Aug. 15th, 2007 06:09 am (UTC)
Link

As always Genome, your posts both amaze me and astound me. Your work and dedication to Monk and its capabilities is something to be admired. Keep up the good work, and I'm glad you're still around providing useful info. for the rest of us aspiring monks. All the numbers are somewhat confusing to me, but the gist of it is placing mods into both STR and VIT are good, especially when you are near a 'threshold,' if I am not mistaken? Also, even if you aren't near a threshold for one or the other stats placing a bit into each category will more than likely produce better results?

I am a big fan of + attack gear but now I see that isn't always the best route to go. I always wondered why the hell I see so many monks with + STR gear for AF, which is not a bad thing but I always thought to myself, well if the mods are 10% for both STR and VIT then there's only so much STR you can put into the WS for mods before it reaches the highest point. I believed that and your posts showed me my assumption was correct :D! I don't have the best gear, but my mods for AF are I believe + 15 STR + 17 VIT.

I was in a pt with a friend and he was decked out in STR gear while I invested in both STR and VIT gear (sticking to my belief / assumption having an equal amount of both will help more) and sure enough when we did our AF I out dmg'ed him. Needless to say he was like, "Wtf!" It just goes to show ya that not everything is about the best gear, but it's more about how much knowledge you have of your job and its abilities.

Reply | Thread

genomeffxi

(no subject)

from: genomeffxi
date: Aug. 18th, 2007 11:23 pm (UTC)
Link

In reply to Cruven (I am not sure if you wanted your email made public)
Cruven- Ragnarok

Im a level 60 Elvaan monk...my gear isnt outstanding....Full af...no SH, no Chiv chain...one STR +3 ring and a DEX +3 ring.

Alot of peopel say i should use my Okotes over my temple gloves. I know temple gloves only add +5 STR; however i prefer to use them over the Kotes 20 attk, as my argument i say that attk doesnt factor into WS. I know that since i lack ACC rings (which even if had the money to buy i wouldnt...i like my STR and Dex) could yes help my DoT would hurt WS DMG. Im also aware that people say i should use Raging fists over Howling. But on EXP mobs since ive recieved Howling ive done more DMG per Howling then i have with Raging. (Howling fist macro has Temple gloves for boost). I generally keep my temples on at all times but i will oc cationally switch to Okotes.

Im not all that sure what to do gear wise etc..here i know ive been trying to save up for a SH not sure if ill use it or not.

(email: (removed))

P.S.: monk is my highest job so i have no merit points

ATK does factor in WS except in very specific conditions where it is already capped and in your case you are probably not in that scenario.
STR is very good, but if you are comparing 5 of one vs 20 of another, you have to consider the quantity too, I would still recommend wearing okotes over temple for normal meleeing and even WS after boosting with temple except in again very specific situations.
Your experience might have been bad with raging and better with howling for a couple of reasons, one might have been low accuracy giving you low dmg on raging, the other might have been you were fighting something like crabs/steelshells a lot where fewer-hit wss can have some advantage over raging. If those conditions remain true, then I think you can keep using howling fist. However you should switch to raging fists as you improve your gear, and use something like a parser to better compare the two, sometimes the impressions you get can be a bit biased, like you will tend to remember only the highest damage and forget the times howling completely missed (something that is less likely with raging) etc.
Scorpion harness is ok but it's only 5acc over your AF1 body, so you might want to see if you can get even more accuracy somewhere else (rings, neck etc) for the same cost.

Reply | Thread

Cruven

from: anonymous
date: Aug. 19th, 2007 04:39 am (UTC)
Link

well ive been so hurt on cash im tempted to sell my Okotes (got em before the R/ex update). It was ont he bird there in Wajom. I dont go out of my way usually to fight soemthin else and i also havent XPd in 2/15 months. I got G3 doen a while ago just cant get up the energy to get out and PT (when i broke G2 i went up 5 levels in less then 24 hours...so it took it out of me.)

Stat wise....

with my melee gear on...

Str+15 (+21 with sole sushi)
Dex 57+9
Vit: 67+7
Attk: 281
Def: 240

Okote stats i lose 4STR, Def lowers to 234 and attk raise to 299.

H2h skill is capped at 202.

Most Damage ive done so far in ym monk career is 1174 on a TW Floating eye in Ranguemont Pass. With no food. a Boost before Raging fists and temple gloves on (AF is the only monk gear i have at the moment im using. everythin else is a bit pricy for me.)

of course i am still "noobish" (self term)on the monk job since i was told my LS has had a history of good monks. I jsut however cant get them to give me any tips >_

Reply | Parent | Thread

(no subject)

from: valyana
date: Aug. 21st, 2007 05:01 am (UTC)
Link

I used to regard the tiers in the pDIF graph the same way: when you hit one of those tiers, adding more attack is suddenly only half as effective. However, more recently it occurred to me that the pDIF 1.0 mode effect indicates that this isn't the case.

At high and low cRatio, pDIF has a range of 0.8. For cRatio near 1.0, it only has a range of 0.5. I believe that the game still rolls a random number from 0-0.8, but that 0.3 of that range (37.5%) gives a pDIF of 1.0.

In the tiers just above and below the middle range, I believe something similar happens. At cRatio = 0.5, pDIF 1.0 is no more likely than any other value. But as cRatio increases, it becomes more and more likely until it hits 37.5% at cRatio = 5/6.

If my theory is valid, this makes attack more effective in those tiers (and less effective in the middle area) than you would expect by just averaging the min and max values. You still see a bump up in effectiveness going from high def to neutral def, but it's much smaller.

I derived formulas for average pDIF from this theory; you can find them in

http://wiki.ffxiclopedia.org/Maximizing_Physical_Damage_Guide#Attack

Reply | Thread

genomeffxi

(no subject)

from: genomeffxi
date: Aug. 21st, 2007 05:35 am (UTC)
Link

Both are definitely a possibility, but it's a bit harder to test than just looking at endpoints of a range.
I guess one way would be to look at the distribution of hits over a long series of battles using the same stats. You can basically plot all those values as a histogram and see what is the most likely value and how it tapers off on both sides.
One possibility is that the peak of the distribution curve is at the average between min and max, and the other possibility that you are suggesting is that the peak is further, basically the distribution gets one of the slopes chopped off.

In ascii art that would be something like:
|...*...|''''''|...*..|
|.*****.|''->''|.***.|
|*******|''''''|*****|
VS
|...*...|''''''|...*.|
|.*****.|''->''|.****|
|*******|''''''|*****|

If you run some tests to get this type of data I'll be interested in seeing it. Either way attack is less effective in these areas but it would not be as bad as half the normal effect if this was the case.

In areas where atk is capped however I have not seen any increase at all in the average either, you can see back to my post on critical hits and kick attacks when I did meat vs no food on farming mobs. If your suggestion was true I would expect to see an increase in average even in a capped situation as the distribution would still shift up and just get truncated differently, so I suspect that the distribution might still be set based on the endpoints in that situation and probably also in the half-capped ares.

Reply | Parent | Thread

(no subject)

from: valyana
date: Aug. 21st, 2007 05:42 am (UTC)
Link

I think the attack cap is a completely different situation; the graph shows it as a cap on cRatio, so adding more attack has no effect at all.

Reply | Parent | Thread

(no subject)

from: anonymous
date: Jun. 9th, 2008 02:01 pm (UTC)
Link

hi. How would you know the value for monster's VIT or monster's Def to put in the equation?

Reply | Thread

(no subject)

from: anonymous
date: Jun. 15th, 2008 09:53 pm (UTC)
Link

[quote]so going from even 200 ATK to 333 ATK (an increase of 130 ATK!) will not change your max damage at all [/quote]

Hmm...that seems unbelievable. Will it give you more consistence max pDIF damage instead of min pDIF though? or is it totally useless?

Reply | Thread

genomeffxi

(no subject)

from: genomeffxi
date: Jun. 15th, 2008 11:40 pm (UTC)
Link

If it helps convince yourself, just try punching a level 5 monster as level 75, notice that you don't punch them for 4629342394 damage... Then eat atk food and punch that lvl 5 monster again, your damage isn't changed. Your consistency isn't changed, you will still do damage in the same range of values.

There are specific caps on your damage range and they are used before adding the randomness in the range of damage. The game doesn't care about how much more ATK you have outside of those caps.

Reply | Parent | Thread

(no subject)

from: anonymous
date: Aug. 30th, 2008 12:04 am (UTC)
Link

Ok mabye im missing something, but how can you calculate mob VIT?

Reply | Parent | Thread

genomeffxi

(no subject)

from: genomeffxi
date: Aug. 30th, 2008 02:08 am (UTC)
Link

The short answer is, you can but don't really need to most times. Usually you just consider a few cases and say work the numbers out for VIT=50, 80 and 100 or whatever, most of the general comparisons and results won't vary and you can get an idea of which range you are at.

The longer answer is those values are fully known for a few common xp monsters, and for all the others there are well-known formulas take use family type, job, level to determine defense and even evasion, and you can use that to determine VIT too. See for example the more complete http://www.geocities.co.jp/Playtown-Part/6371/memo.html
(if you can read japanese)

Reply | Parent | Thread