Not the path of least resistance

For players obsessed with rewards through safe, unadventurous busywork with long-established best practices, "beastman stronghold invasion" (FFXIclopedia) in "Shadowreign" areas draws almost zero interest, which is also the case for most of WoTG that isn't Campaign so far (deserved or not). Most of the items are lackluster or useless, but it's precisely because of that (nothing lootwhores want) that a group of players shouldn't have too much trouble obtaining access to the culminating battles (18 members maximum) in each of the beastman strongholds. The potential rewards from these battles aren't all that enticing, either--marginal rare/ex items aside from beastman headgear and low-value auctionable stuff--but the battles seem like a fun diversion, and if your group doesn't operate during Japan prime time you shouldn't have any problems with competition for scarce resources. (The NMs are timed spawns and are slow to respawn after being killed.) And even though the items aren't all that, you can still hope "early adopters" blow their gil on these latest toys.

Let's not forget the dealbreaker of having "friends" to do this with, but that wasn't going to stop me from frittering away the exp buffer I unintentionally built up while auto-attacking fortifications last time.

Starting with La Vaule, six of the eight NMs are monks, rangers, or mages (SCH, BLM), automatically ruling them out as feasible for ninja soloing, at least without a dancer sub ("Easy Mode") to try (in vain?) to counter a steady loss of HP. I looked for Coinbiter Cjaknokk (DRK) to see first hand if it would actually spam Shoulder Charge, but it was nowhere to be found.

Dismayed, I shifted my attention to Beadeaux, where five of eight NMs are rangers, monks, or mages, leaving Mu'Nhi Thimbletail (THF), Ga'Lhu Nevermolt (PLD), and Di'Zho Spongeshell (DRK). Spongeshell alternates between melee-absorbing and magic-absorbing states (absorbing meaning healing), and I didn't feel like committing inventory space to elemental ninjutsu (assuming I could even make inroads on it). Thimbletail was nowhere to be found, leaving Nevermolt.

Not unexpectedly, these NMs can be found among the general population, which can make isolating them very time-consuming, but Nevermolt will separate from nearby Quadav, making a pull easy. But, it was painfully obvious after several single-digit katana swings (most for 0) that I had no chance. Even worse, committing to any sort of evasion setup is completely pointless: out of 45 attacks, I evaded exactly 2 and parried 3 and I used my evasion setup exclusively. Paladin spells do consume shadows, yes, but the only reason I even lasted that long was that Nevermolt turns its shell on you periodically, giving you some time to recast shadows. Hojo resisted several times, too.

I hold out hope that I can see Thimbletail in the next 24 hours so I can proceed to get dumped and give up the futility.

pDIF distributions

Anyone with at least a passing interest in how the game calculates physical damage is likely aware of the so-called pDIF factor, which is a function of the ratio of one's attack to the opponent's defense (ATK/DEF). A given value of ATK/DEF corresponds to a specific range, or distribution, of possible damage values constrained by a minimum and a maximum, and one can treat the pDIF graph as a concise summary of the possible distributions from 0 to 2 ATK/DEF.

But what is the underlying probability distribution for all possible ranges? A uniform distribution with the parameters of pDIF minimum and pDIF maximum has strong intuitive appeal because random numbers from a uniform distribution are simple to compute. It would seem impractical for the programmers to mess with normal distributions, and the apparent reliability of pDIF max and min in predicting a range of damage values (at least for one-handed weapons these days) basically precludes the use of standard normal. (It makes no sense to parameterize a normal distribution with pDIF min and max, anyway.)

Moreover, assuming a uniform distribution makes it easier to calculate damage with an expected value of pDIF, which would be just the midpoint between the endpoints of a given distribution if it and all others were really uniform.

But is it really the case? To get a sense of it I considered what would be the easiest, least riskiest, least costly and least time-consuming way to collect data without actually paying attention to the game, which basically meant poking at Campaign fortifications with dual-wield katanas I already had (Mamushito +1).

I acknowledge that my original goal in doing so was not really to gather evidence for a uniform distribution but rather to see to what extent the distribution of damage values might change with an increase in attack from a meat mithkabob (told you I was going on the cheap, and I was thinking maybe the distributions aren't uniform). I also ended up concluding that fortifications are a nice target for testing this, in a way; because of the extremely limited range of actual damage values due to their damage-reduction properties, I had no need to trouble myself with appropriate histogram binning.

A rank promotion later, I put together some "composite histograms" in Excel to summarize my peculiar results:

"Lower attack"

"Higher attack"

While the "higher attack" case didn't yield any surprises, the "lower attack" case was definitely not uniform in the least, but why the bias toward 6 damage? What's up with that?

Nonplussed, I attempted to find any snippets of comments regarding pDIF using Google, and I came across an interesting statement about pDIF, which is paraphrased as follows:

"For a given pDIF distribution, if pDIF 1.0 is within the range of possible pDIF values, pDIF 1.0 has a probability of 1/3, with the other possible values being uniformly distributed otherwise."

This statement, if true, would apply to cases of ATK/DEF between 0.5 and 1.5, which pretty much encompasses everyday conditions when fighting. It seems plausible enough in light of the data I collected, but why would anyone go to the trouble of making it so?

At this point, I thought it might help to try some simulation with random uniform numbers to see if I could obtain similar results to what I showed in the graphs above, and by doing so illustrate a possible method for creating a bias toward pDIF 1.0. The biggest problem was making an educated guess about the fortification's attributes, especially the damage reduction property, but I had to run with something.

For the "higher attack" case I managed to get a similar result to my obtained data with a ATK/DEF ratio of about 1.521:


For the "lower attack" case I was unsure how to simulate a result similar to what I obtained from data collection and I looked for further clarification. One idea held that pDIF 1.0 at the endpoint of a distribution is the result of random values below 1.0 (or above) being rounded up (or down) to 1.0. But this doesn't jibe with a large data set I collected while poking at a fortification (when I regrettably neglected to record STR and attack) where 6 damage (the mode) seems to correspond to 1.0, yet 5 damage was recorded also:


But wait! Ignoring the 6, don't the data suggest a long right-hand tail? A uniform distribution doesn't have tails! And why does the range of damage go from 5 to 14? At 395 attack, maximum damage was shown to be 11. I probably wasn't using a meat mithkabob, and I try to maximize attack speed so I don't bother with attack equipment. But, it might be useful for reference later.

So ultimately, I have no conclusion that I'd rely on. I did perform another simulation to demonstrate how the "lower attack" case described a long time ago might come to pass. Let's say about 25% of all pDIF random values on the interval [1,1.65] (ATK/DEF ratio 1.375) end up being converted to pDIF 1.0, ensuring that 1.0 is the mode of any pDIF distribution that contains it. Otherwise, the data are random uniform numbers. Then, this criterion works in my simulation (rather, I ran the simulation a bunch of times until I found a result that looks similar to the one above):


It's too bad getting a feel for the underlying distribution from a random sample is quite annoying in the case of pDIF. Maybe I'll try again with attack lower than 344 next time.

Data collection was made possible with the "offense detail" feature in kparser. Otherwise I wouldn't even bother.

Crystal Stakes results update

After 30 C1 races my chocobo has yet to place worse than 3rd, which is actually kind of nice since I haven't had to write off a disastrous 4th or worse finish. To my delight, some of my PC competitors have actually placed worse than 3rd even with fewer than 3 entrants. (The worst result was 6th for a SS/B/A/C chocobo; I placed 3rd.) Since the Sept 8 update, my chocobo's 1st-2nd-3rd tally is 6-2-2; overall, 14-12-4.

In the last 10 races, I've faced PC chocobos 7 times. I've used the mythril saddle at least twice (the first two races after the update I finished first both times with a mythril saddle) but otherwise I've used the elm saddle. Speed apple used, as always.

I hope in the long run my win rate ends up being closer to 60%, but that's probably wishful thinking because of competition from PC chocobos. It hasn't been uncommon for me to face two other PC chocobos in a race. I have yet to try the C1 slots at 21:45 and 22:45 JST time. I've done either morning or mid-day, and the usual suspects are sticklers for the morning slots (5:45 AM and 6:45 AM)

What are my net gil earnings? In the past week I've hardly been doing free runs, and when I do I just do one cycle of eight races, which costs 3,550 gil to complete. Because 2nd and 3rd place finishes result in a net loss of chocobucks, I factor in the costs of chocobucks, too.

1st: 90,000 × 14 = 1,260,000 gil
2nd: (30,000 - 40/28*3550) × 12 = 299,142 gil (approx.)
3rd: (10,000 - 64/28*3550) × 4 = 7542 gil (approx.)

Total: 1,566,685 gil (approx.)

I really couldn't be happier with this gil fountain that requires minimal effort to tap now (despite the initial outlay of raising a chocobo). I don't covet any big-ticket items (too bad I can't just up and buy Usukane Sune-Ate, Usukane Hizayoroi, and Morrigan's Slops, pieces for which the materials are cheap), but maybe I'll think of something to blow my gil on later on.

I've noticed, or at least it's somehow evident to me, that accidents ("stuck in the mud") may adversely affect other chocobos since most of them appear to be startled when it happens. (The camera "work" is still bad though so sometimes you can't get a good look at how they react.)

I think I'll give a progress report after every 15 races just to give me an excuse to use Blogger once in a while. It's pretty sad I've found occasion to make 8 posts this month already...

Magic damage calculations

As someone who's wasted time caring about magic damage on Ebony Puddings in a vain attempt to approach 9,000 exp/hr any day of the Vana'diel week, I've found the so-called magic damage formula to be quite useful in getting my hopes up. I've also wondered why no one ever bothered listing the "base damage" values ("V values" in the FFXIclopedia article) of offensive spells other than black magic for the sake of reference.

After further investigation, I managed to dig up a reference of "magic D values" not only for black magic but also for white magic and ninjutsu, along with some other sundry details. Earlier, I endeavored to calculate the D-value for the ni line of elemental ninjutsu, and I was happy to find out that I managed to get the right value (78). Interestingly, the san line has a D-value of 105 and a multiplier of 1.5, so you can pretty much dismiss claims of 400-500 damage from san spells as the result of targeting tiny mandies. (You certainly can't do that well on Ebony Puddings.)

As an example, suppose you had +40 MAB (24 from /BLM, 5 from Moldavite, 8 from Uggalepih Pendant, 3 from Denali Kecks), 40 INT from equipment and an elemental staff to boost damage further. If you're a taru with no INT merits (74 INT), your dINT will be 25 going up against an Ebony Pudding (INT 89). Based on this, I computed the damage to a pudding from a san spell to be 285. 400+ damage, where?

Going back to my previous rigmarole about great axe break weapon skills, I confirmed for myself that the unresisted duration of their debuffs should be 180/270/360 seconds corresponding to 100/200/300 TP; in a partially resisted case, the duration is lowered to 90/180/270 seconds, or by exactly 90 seconds. (Finally, I got a Full Break where the accuracy down was partially resisted and the other effects weren't. The time difference between accuracy down wearing off and the rest wearing off was exactly 90 seconds.)

It's kind of funny that these durations are easy to confirm compared to the degree of the debuff effects, yet the reported debuff effects are treated as gospel while the reported durations are obviously wrong now (allowing for the possibility that they used to be 3/4/5 minutes unresisted).

Based on what we know about the effects of the break weapon skills, I have to wonder whether the accuracy down effect of Blade: Kamu can be resisted (lowered duration). That would depend I suppose on whether Blade: Kamu has an associated element (probably earth, but who knows yet...). Break WSes are never used in practice, anyway, and I wouldn't count on Blade: Kamu either, especially if its effect is really accuracy -10 as listed on Studio Gobli currently. There are probably better things to do with your TP, especially subbing dancer in a soloing context.

Ninjutsu damage and break WS duration, two things that I've always wondered about and that I finally know more about. Hooray for utterly useless knowledge, yet arguably better than talking about income from chocobo racing or heaven forbid, gear I want to primp and preen in.

Break weapon skills

(I should let it go, but there are a few things I've always wanted a decent explanation for...)

Studio Gobli has summarized the properties of the new job-specific weapon skills, and the traits of some of these WS are intriguing if only in a tangential "oh, that could be useful in some situation that will never come to pass."

1) King's Justice (WAR) has a primary skillchain attribute of Fragmentation (thunder/wind), giving the warrior the ability to participate in a Light skillchain using a two-handed weapon without having to obtain Ground Strike (great sword). (Its secondary SC attribute is Scission or earth.)

2) Vidohunir (BLM) lowers magic defense (as though you would ever melee) and the duration of this effect is presumably 60/120/180 seconds with 100/200/300 TP.

3) Blade: Kamu (NIN) lowers accuracy and the duration of this effect is 60/90/120 seconds with 100/200/300 TP. Ninja can now participate in a Light skillchain (Fragmentation attribute), as though you'd want to. I assume Blade: Kamu is earth element given the accuracy down effect.

There are a few other WS with "duration of effect varies with TP" (RDM, SMN), but I'd really like to know whether these effects are explicitly stated in the chat logs when they actually process (even better, whether they're resisted or not). If they actually are, then why not do the same for the great axe "break" weapon skills?

I always disliked the fact that you have no idea whether the "[attribute] down" effect actually kicked in immediately after using one of the break WS. In practice, you have to look at the chat logs to see if the effect eventually wears off ("The [monster's] [Attribute] Down effect wears off"), implying that the effect was actually applied by the WS, and if you don't see any message, you usually end up concluding one of the following: (1) the effect was still active when the mob died (if the effect is really there, shouldn't it be obvious, you say?); (2) the mob used an ability/spell to override the effect, or; (3) the effect was never there to begin with.

Why would you care? Well, for those of you that proselytize low-level WARs into using Shield Break exclusively, you might actually want to know:

(1) how long the evasion down effect should actually last
(2) whether it actually works reliably on exp mobs of interest

Regarding (1), Studio Gobli lists durations of 3/4/5 minutes with 100/200/300 TP for all the break weapon skills, and this seems to have been generally accepted as true, if hardly widely known. Yet I just had to convince myself that this was true, and after messing around in Lufaise Meadows for a bit, I was surprised to find that, on the neighborhood mobs at least, the effect duration is more along the line of 90/180/270 seconds with 100/200/300 TP, as summarized by the following graph:



I had no evidence of partial resists (just an all or nothing effect); this was all the data I got from break weapon skills where the effect was applied.

Most of the Shield Break results were from bees, which are supposedly weak to ice. The one Weapon Break was used on an orc, which is supposedly weak to water, and the Armor Break was used on a bugard, which isn't known to be weak to wind.

As for (2), one way to convince yourself that Shield Break should work, without waiting for a "wearing off" message that you may or may not see, is to target mobs that are known to be weak to ice. This is merely a rule of thumb, as Shield Break might work on mobs that aren't weak to ice (such as Goblins) because resists should come into play, and it's possible Shield Break won't work on mobs whose crystal drop actually contradicts a "known" ice weakness. Bugards, which drop fire crystals yet are putatively weak to ice, are one example (though admittedly not an exp target).

I myself was more interested in whether I could ever see a full complement of debuffs from Full Break on mobs that aren't weak to earth, to no avail:

Makara
113% TP
101 seconds
accuracy and evasion ONLY

Makara
114% TP
103 seconds
evasion ONLY

Gigantobugard
138% TP
nothing in 8 minutes!

Death Jacket
167% TP
150 seconds
accuracy, attack, and evasion (no defense)

Death Jacket
123% TP
111 seconds
accuracy and evasion ONLY

Gigas Warwolf
185% TP
nothing in 3 minutes!

Gigas Martialist
112% TP
100 seconds
evasion ONLY

Orcish Bowshooter
171% TP
154 seconds
attack and evasion ONLY

Note that earlier I said the above results applied only to the mobs in Lufaise Meadows... to my chagrin, I later obtained the following preliminary results from Bull Dhalmels in Buburimu Peninsula:

Bull Dhalmel
128% Full Break
230 seconds
evasion and defense ONLY

Bull Dhalmel
300% Full Break
360 seconds
evasion and defense ONLY

Bull Dhalmel
150% Full Break
135 seconds (Windsday)
attack and defense ONLY

Bull Dhalmel
107% Full Break
96 seconds (Windsday)
full debuff!

From this data alone, I could conclude that the duration of the debuffs from Full Break is up to 6 minutes at 300% TP. The other data, hmm...

So, what gives?

Did Studio Gobli used to be right? The evidence doesn't support 3/4/5 minutes at the moment.

Are there actually partial resists? Perhaps a partial resist reduces effect duration by 90 seconds flat regardless of TP level. The dhalmel data suggest an unresisted duration of 180/270/360 seconds with 100/200/300 TP. I propose a "partially resisted" duration of 90/180/270 seconds with 100/200/300 TP, which would match up with the results from Lufaise Meadows. But if those are really "resisted" durations, why didn't I ever see unresisted durations ever in Lufaise Meadows?

What other factors am I not considering (day of the week)? No, I didn't use a Martial Bhuj.

Dormancy

There are not a few interesting, new, even completely unexpected things from the latest version update, such as the new "arena" battles in beastmen strongholds of the past, not that I should keep deluding myself about being able to participate in these successfully within the next year. Level Sync seems nice enough. Heck, even reducing the HP of certain beastmen's pets is a completely unexpected change because you'd never think SE would've ever tacitly acknowledged leveling off those as a legitimate form of experience point acquisition. (Yes, don't give SE too much credit yet; it's possible that there aren't actually any good camps in ToAU or WoTG areas...) These new things are genuinely praiseworthy by my standards.

On the other hand, there are changes to existing "content" that fall under the "it's about fucking time" category. For example, Nyzul Isle progress is now recorded for all players (under certain conditions) instead of just for the annoying asshole leech who demands that he use his runic disc. Chocobo rentals in the past after nine months is infuriating when they could've introduced them in the March or June updates. Still, sometimes these things are welcome even if unreasonably overdue.

Then you have "what the fuck?" changes whose ramifications are not immediately apparent...

"The strength of Zeni NMs will now gradually be adjusted after a predetermined amount of time has elapsed following their being claimed." What does this mean? Can you actually elaborate without having us figure it out through trial and error? Can you actually stop having zeni costs be maximized after every maintenance?

I think it might be okay to disincentivize solo or low-number attempts on ZNMs on a temporary basis if there is actually high demand for these at "peak" hours currently, but is this really the case? Doubt it. Wanna bet SE just keeps this crap in permanently?

Another example is that "the effects of certain items used during chocobo races have been adjusted," whatever that means. Why did this have to be changed to begin with? It apparently was fine for over six months!

Overall, though, this update seems solid enough, although it isn't enough to keep me wasting excessive amounts of time for minimal "enjoyment" return. (I've gone well past the point of diminishing returns about a year ago.) Even the simplest things take forever to do, and it's not like I'm actually going to rely on someone else to honor his or her "word" about returning a favor at a later date. That shit has never, ever happened, ever, and I don't need to waste anymore time so that I can someday be wrong.

Nonsense

Since the June update, I've been eagerly awaiting more than one "casual," cutscene-heavy quest to do for the past three months since the fare offered in June was almost non-existent. (Mythic Weapon quests and complete garbage like "The Miraculous Dale" don't even count, but Babban Ny Mheillea was amusing.) Yet the "developers" apparently forgot that key quest-related NPCs disappear during Campaign battle in Bastok Markets (S) and Windurst Waters (S) (and probably San d'Oria, too, depending on your server population's collective ineptitude at Campaign), a ridiculous oversight on their part given that they now allow campaign arbiters (providing teleportation services) and Sigil NPCs to remain during battle (not to mention some Campaign Ops NPCs whom no one cares about). Don't you bootlickers forget that back when "Wings of the Goddess" was released that SE promptly issued an update/fix allowing some quest-related NPCs (for maps) to remain even during Campaign battles. Fuck, even Diordinne sticks around during Campaign.

Obviously, disappearing NPCs in the "hometowns" wasn't a problem before Campaign was fucked with (fixing something that wasn't broken) in favor of the beastmen confederates, and maybe they think this will be moot once loser nations finally get their acts together or the initial surge of interest in the new quests dies down, but it's still puzzling they won't fix something that seems so basic given the fact that they've done it before.

Less time butchering once-orphaned "content" like adventuring fellows, where the "devs" saw fit to reinvent the wheel--same grinding shit just to make entirely cosmetic changes to our fellow's appearance--yet no ToAU availability (who cares if losers "solo" on greater colibri with their fellow?) and no free fourth armor lock. More time on quality-of-life issues, thanks.