Friday, June 20, 2008

Parser-mony

As someone who participates in merit parties about twice a year (learn to allocate merit points efficiently according to your everyday needs instead of being a "fully merited" retard), I rely on log parsers to keep me interested when the action in itself is boring as hell. I find KParser a very useful diagnostic tool in alleviating the boredom of meriting.

Why would I use KParser instead of other parsers? Parsers that I have previously used tend to cater solely to people jacking off to how much damage they do in a parsing session; their output doesn't tell you anything about the proportion of the monster's attention that each damage-dealer commands, which might be worth considering before one makes ridiculous generalizations regarding the relative effectiveness of a job combination or equipment set. In contrast, the sophisticated KParser actually crunches detailed numbers for defensive statistics (among other things, too many to discuss briefly)! KParser is not only interesting enough to keep me from being bored to tears, but it's also a pretty neat diagnostic tool to help me delve deeper into a parsed situation.

I put KParser to the test during a pickup merit party at Tadjana Isle (Greater Colibri with the occasional Mamool Ja mixed in) with the following party configuration:

Myself as WAR/NIN #1: used Perdu Voulge exclusively; 12% Haste (Walahra Turban, Swift Belt, Dusk Gloves), mediocre WS equipment.

WAR/NIN #2: Used both Byakko's Axe and some dual-wield combo with main-hand axe, at least 17% haste probably.

WAR/NIN #3: Maneater/Ridill, Adaman Hauberk, at least 17% haste probably. (Kupo member.)

WAR/NIN #4: Maneater/Juggernaut, Ares's Cuirass, at least 17% haste probably.

RDM and BRD/NIN

Looking at the offensive numbers alone (102 monsters, 19,385 experience points over 1:09), I can't really conclude that any of the warriors is loafing; our damage proportions are within 5% of one another (this is just a rule of thumb of mine, not based on any statistical criterion):


Damage Summary
Player Total Dmg Damage % Melee Dmg WSkill Dmg
WAR/NIN #1 124488 26.17 % 74760 49728
WAR/NIN #2 108562 22.82 % 73771 34736
WAR/NIN #3 112252 23.60 % 77009 35242
RDM 1080 0.23 % 926 153
WAR/NIN #4 128528 27.02 % 85934 42582
SC: Reverberation 776 0.16 % 0 0
Total 475686 100.00 % 312400 162441


However, looking at the numbers for "attacks against," we see here that WAR/NIN #3 (Maneater/Ridill) took the fewest attacks of all the damage-dealers:


Attacks Against:
Player Total Attack# % Avoided Avoid %
WAR/NIN #1 174 24.58 % 145 83.33 %
WAR/NIN #2 118 16.67 % 85 72.03 %
WAR/NIN #3 83 11.72 % 54 65.06 %
RDM 8 1.13 % 1 12.50 %
WAR/NIN #4 179 25.28 % 142 79.33 %
BRD/NIN 146 20.62 % 113 77.40 %


KParser even has a distinct category for Utsusemi casting. Notice that WAR/NIN #4 (Maneater/Juggernaut, Ares's Cuirass) cast Utsusemi: Ichi 25 times during the 1 hour, 9 minute session, which may have had a deleterious effect on his damage output:


Utsusemi

Player Shadows Used Ichi Cast Ichi Fin Ni Cast Ni Fin
WAR/NIN #1 133 15 11 48 43
WAR/NIN #2 72 4 4 34 30
WAR/NIN #3 49 4 4 28 24
RDM 1 0 0 0 0
WAR/NIN #4 130 25 19 44 42
BRD/NIN 112 4 4 43 40


The best feature of all is KParser's ability to filter types of chat, including Arena and NPC chat! Well, it can help when you want to review NPC dialogue by filtering out annoying /party and /shout crap. It could also be useful to revisit what someone said in /tell that you happened to gloss over hundreds of chat lines ago.

From a meriting standpoint, I don't use parsers to make sweeping, erroneous generalizations but as a diagnostic tool to help me fine-tune my own performance. There isn't much of a take-home message from a typical parse of a merit party; all I really care about is that the group as a whole is getting the job done and that I am pulling my own weight.

As far as soloing concerns go, of my outlandish, totally unrealistic "ninja solo" goals is to acquire the items chest on the second floor of SE Apollyon. KParser can help me determine the (lack of) effectiveness of my evasion setup and (poor) dagger accuracy when fighting the Adamantshells, and also the average time spent fighting each Adamantshell.

Note: parser output was truncated for clarity.

Are chocotrain tokens worth it?

Searching for a new challenge, I have been revisiting "content" that has basically been orphaned by the general playerbase. Chocobo racing, yeah, that's a hot ticket.

Having recently entertained thoughts of putting my long-retired and seldom-used chocobo through the gauntlet of Chocobo Circuit races in the hopes of obtaining eventually any Crystal Stakes trophy, I considered the possibility of using Chocobo training tokens to raise its receptivity in order to mitigate adverse status effects inflicted by the other jockeys during the races. For example, one may use a Gysahl Bomb to reduce the speed of the other chocobos.

The following asides will help to explain my motivation for raising receptivity:

Aside 1: Chocobo racing in general is a microcosm of the FFXI experience where cheap-ass AI tactics or onerous restrictions provide most of the obstacles to gratification. The chocobo attribute "receptivity" is a measure of how well your chocobo copes with such tactics in the domain of chocobo racing. A higher receptivity level for your chocobo supposedly increases the probability that your chocobo "evades" the adverse effect altogether. And if your chocobo does suffer the adverse effect, receptivity supposedly lessens the effect in some way (reduced duration, attenuation of the effect itself, something).

Aside 2: Chocobo raisers generally accept the idea that the game measures chocobo attributes on a hidden point scale of 0-255, which corresponds to the eight discrete levels that the game actually reports:

0-31 - "Poor"
32-63 - "Substandard"
64-95 - "A bit deficient"
96-127 - "Average"
128-159 - "Better than average"
160-191 - "Impressive"
192-223 - "Outstanding"
224-255 - "First-class"

Aside 3: 80-chocobuck training programs are available to "adult chocobos still residing in the stables" (not retired) and are understood to raise attributes by 5 points. But these are not available to my retired chocobo.

For my retired chocobo, the only way to raise its receptivity attribute is to use Chocobo training tokens, which cost 100 chocobucks. Based on the above information I came to the terribly wrong assumption that these 100-chocobuck tokens also raise attributes by 5 points.

Unfortunately, after entire days of "botting" free runs to accumulate chocobucks, I found out that these tokens are actually less effective than the aforementioned 80-chocobuck training programs available to "active" chocobos. Not even seven RCP tokens--which I thought would raise receptivity by 35 points and thus raise my chocobo's receptivity from a "Poor" rating (0 points) to "Substandard" rating (35 points)--were sufficient to raise its receptivity from "Poor" to "Substandard." Who the hell knows how many points they're worth? Two points? Three points? Fuck that.

Considering the time and gil cost of accumulating chocobucks, to offer 100-chocobuck training tokens that are more expensive and less effective than the 80-chocobuck training programs is classic FFXI "development team" timesink/gilsink bullshit. I even wonder if the 90-chocobuck training tokens are similarly less effective, but there's no way in hell I'm going to check.

My hope was that I wouldn't have to raise another chocobo just to place in the Crystal Stakes months down the line. Instead, I would rather eat the time and gil cost now of "botting" hundreds of free runs a day to bring my chocobo up to competitive form rapidly and get a fucking inventory-1 trophy within the month.

Yet not even I am masochistic enough to throw away millions of good gil after bad. At least I can take solace that I can accumulate chocobucks for other purposes without actually attending to the game at all, one motherfucking four-minute free race, 3.5 chocobucks (on average) at a time.

Motivation behind this blog

I consider most Final Fantasy XI-oriented "blogs" utter trash. No one wants to read about some FOBs waxing solecistic or fat, narcissistic bitches preening about some loot they leeched or be inundated with screenshots demonstrating that they lootwhored such and such.

Now, I don't have any grandiose aspirations for this blog of mine other than to document some esoteric issues that I come across (rather, inflict upon myself) in the day-to-day playing of FFXI. But I do fancy the idea of building a reasonable alternative to those garbage "blogs" that shed no light whatsoever and serve only as monuments to grotesque self-importance.

Why bother discussing things of very limited interest anyway? Frankly, no one I know in FFXI actually cares about any of the issues I take up, so I don't bother bringing things up with them, but I hope that someone out there might care and call me on some mistaken assertion or learn something that is worth knowing to them. I'm an optimist that way.