Genuine Atomos Pictures now with Taj

[Senoska]

Taj's List of Win:

(Please correct any mistakes)

-Entering Nyzul before release
-Entering HazhalmTesting Grounds
-Locking people's items into bazaars
-Force disconnecting other players
-Accessing Windurst debug NPCs

I wonder how large the list will grow.

Edit: Corrected.

Hoe checking a GM when hoe checking was cool was the very first taj accomplishment on /bg/ iirc

[Taj]

Taj's List of Win:

(Please correct any mistakes)

-Entering Nyzul before release
-Entering HazhalmTesting Grounds
-Locking people's items into bazaars
-Force disconnecting other players
-Accessing Windurst debug NPCs

I wonder how large the list will grow.

Edit: Corrected.

Hoe checking a GM when hoe checking was cool was the very first taj accomplishment on /bg/ iirc

Nein, I believe that was after Nyzul. Really shortly after, though.

[Secksi]

North Al'Taieu?

I thought you went to the limbus zones first too, weren't they called something else?

[Taj]

North Al'Taieu?

I thought you went to the limbus zones first too, weren't they called something else?

Yeah, just Ou'Hpat Obelisk. I didn't know how to get into Lower Promyvion at the time.

[Tricen]

Taj's List of Win:

(Please correct any mistakes)

-Entering Nyzul before release
-Entering HazhalmTesting Grounds
-Locking people's items into bazaars
-Force disconnecting other players
-Accessing Windurst debug NPCs

I wonder how large the list will grow.

Edit: Corrected.

Hoe checking a GM when hoe checking was cool was the very first taj accomplishment on /bg/ iirc

Nein, I believe that was after Nyzul. Really shortly after, though.

Wait you did what >_>?

[Rocl]

Taj's List of Win:

(Please correct any mistakes)

-Entering Nyzul before release
-Entering HazhalmTesting Grounds
-Locking people's items into bazaars
-Force disconnecting other players
-Accessing Windurst debug NPCs

I wonder how large the list will grow.

Edit: Corrected.

Hoe checking a GM when hoe checking was cool was the very first taj accomplishment on /bg/ iirc

Nein, I believe that was after Nyzul. Really shortly after, though.

Wait you did what >_>?

He went into Nyzul long before it was every released.

[Tricen]

I mean he hoe checked a GM <_<.. How well did that go over?

[Mayareira]

wow someone found that old ss lol yeah that was from when they copied accounts for ballista royale, i killed vilma and taisaijin for shits and giggles then ran to behemotdom to check yellow KB and got stopped D:

http://i20.photobucket.com/albums/b2...r/img_0039.jpg

One day I'll own this ring. Its a matter of pride!

[mikazuki]

640/4 means faf in this case* has a DMG of...
137
(137/9 +8 etc.)
that's pretty lame tbh >.>
it's pretty easy to work out his attack and str from that but probably more sensible on an uncompeted mob.

*making all kinds of assumptions about mobs formula being similar to that of players

also to whoever said 6 attacks~ didn't prove faf couldn't crit on the GM, the rate should be 1/4~1/3, it gives around a 10% chance for no crits to happen, i'll be satisfied when i see a screenshot of any mob critting a GM.

the judge gear might only have the description of def+50 and not actually have the mechanic of changing the GM def value - which is likely set to a naked lv1, or simply 1. it doesn't really make sense for their def to actually be calculated since it's never a relevant stat.

[Alvinho]

lol Maya. Keep trying.

[divisortheory]

640/4 means faf in this case* has a DMG of...
137
(137/9 +8 etc.)
that's pretty lame tbh >.>
it's pretty easy to work out his attack and str from that but probably more sensible on an uncompeted mob.

*making all kinds of assumptions about mobs formula being similar to that of players

How would you go about working out his Attack and STR?

[Awoir]

640/4 means faf in this case* has a DMG of...
137
(137/9 +8 etc.)
that's pretty lame tbh >.>
it's pretty easy to work out his attack and str from that but probably more sensible on an uncompeted mob.

*making all kinds of assumptions about mobs formula being similar to that of players

How would you go about working out his Attack and STR?

I think for attack you would need to find the highest amount of defence needed to have him still hit you for 640(?) in normal(un-raged) conditions, and the lowest amount of defence you need to have him hit you for 0 under normal (No-JA's bar sentinel, no phalanx) conditions? I could be totally wrong though so wait for Aurik or someone who knows a bit more to chime in.

[DuracellPhoenix]

derail... I ran across this guy on Phoenix with Vilma's Ring April last year

http://i31.photobucket.com/albums/c3...4-18-06-35.jpg

[Nagamaki]

Before trying to figure the offensive stats of Faf, you'd want to start by figuring out if there are other differences between the player vs mob and mob vs player systems beyond the (theoretical) modified pdif cap.

So you'd want to use a reasonably low level mob, that is still high enough to hit full cap on a lv1. Something EM to 37 would probably be perfect, you could then use a lv1 job/mule for the cap test and a 37 job or whatever for EM. You'd want to have a job slightly above/below the mob's level as well to check for possible level correction.

You use your lv1 mule/job to figure how much the Lv37 mob hits it for, you're looking for a capped number. From that number you can find out their weapon damage and fSTR cap (assuming the same rules apply). Then, you switch to your level 37 job (should have as much range in def gear/buffs and VIT as you can manage) and you have the mob beat on you until you know the mode number.

Once (if) you find the mode, you then tweak your VIT up and down as much as possible (since low lvl gear offers some range but nothing insane, ideally you have a BRD there) and see what happens to the mode. You're trying to figure out the mob's STR by seeing at what pace the mode fluctuates through reverse application of the fSTR comparison, until the mode is equal to what you calculated as the mob's weapon dmg.

Once (again, if) you figure out the mob's STR, you repeat the same variation of gear/buffs with def until you find what number makes the avg dmg per hit taken (non-crit) the same as the mode. Once you have both attack and STR figured out, you come out on a different job slightly above the mob's level and use the numbers you found try calculate what the avg dmg should be around against your new VIT/def, then test what it is and see if it looks like there's a level correction applied.

If all of that works, you've just confirmed that mobs use the same type of fSTR and pdif to hit us as we use to hit them, but with a different cap.

Now do the same thing with Fafnir.

Its possible all this was already tested by studio gobli or something, if it is all truly confirmed then you could skip the rest and go straight to Faf testing... if people feel like letting you do that with Faf anwyay.

For that sort of test on HNM types, what you'd probably really want to use is crap nobody gives a damn about, like the ToAU kings. You're not going to get exact matches to the numbers of real useful mobs, but at least it should give you a range, and then you could try to look for modes while fighting the real stuff.

edit: an extra note, since you can find mob def with /c and vit using your fSTR, you'd probably want to figure the exact value of those, to use as starting value for your def/vit tests. Its not likely to be the same, but at least it'll give you a clue.

[Priran]

derail... I ran across this guy on Phoenix with Vilma's Ring April last year

Ring also drops from Rose Garden, which under the best of conditions pops twice a day.

[Sidanoir]

Can't you just get a bunch of data points for one VIT and defense set up off of a Faf...

Then repeat with a different VIT/defense set up.....and take the differences between the 2 to figure out his attack stats? The same numbers should be popping up for the most part...or do mobs have variable pdif too? Or is there still many unkowns to solve?
I guess there is 3? fstr? attack and damage?

I know with Faust i was able to floor how much I was getting hit for on NIN to be the same everytime pretty much...can't that be used in those equations? Never really looked into FFXI math too much ...

[mikazuki]

,_, you can work out the str and atk mainly by exploiting the region for atk/def which gives pDIF = 1, check what vzx did for working out the vit of ulli and byakko. in this case you know the def and vit, so you can discover the atk and str.

dmg+((str-vit+x)/4) * f(atk/def - 0.05*lvdif)

dmg is approximately 137: if this figure is inadequate then you have to just test by varying VIT.
vit is known: in this example, 100.
x will be around 4.
def is known: in this example, 500.
lvdif is zero in this case, so ignore it.

137+((str-100+4)/4) * f(atk/500)

now we imagine that faf happens to have a range of atk that lets pDIF = 1, if he doesn't we need to increase (or decrease) our def value until he does. obviously when pDIF = 1, all normal hits are for the same damage.
if it gives a result of 160 (just for example) then you would say:

160-137 = (str-96)/4
23*4 = str-96
str = 192

note this is a slight approximation but the error range can be discovered.
you can see the maths is easy - testing to get pDIF = 1 might be extremely difficult on fafnir.

now you can work out the exact atk!
if mobs followed the same rules about cRatio as players (see note below) then you would already have a range of values for atk, because you know where the min/max is in regard to your def. but let's assume you want to know more, well conveniently you can now set pDIF greater or lesser than 1! (less is preferred, because it means you don't have to raise for every data point).
damage = base_damage * pDIF
where pDIF has a maximum and minimum which varies with atk (def being static). for this test you have to get hit plenty of times to show maximum and minimum pDIF values.
let's say...
damage_max = 163
base_damage = 160

163/160 = 1.01875
now you would say fmax(atk/800) = 1.01875
800 is just an example, probably need a brd/blu if you want to produce a number like that.

damage_min = 20

20/160 = 0.125
fmin(atk/800) = 0.125
you would now apply this into the cRatio => pDIF formula...

1.2*(atk/800 - 5/12) = 0.125
0.4 + 1.2*(atk/800) = 1.01875

atk = 416
atk = 412.5

at this last stage we come to a problem - that being, since i made up all the numbers, and there are rounding errors, the solution is only close to 414, oops :D
nonetheless it's evident that you can determine to within 1% the atk of the mob.

DISCLAIMER
all this relies on the mob dmg formula being basically similar to that of players - it's easy to tell if it isn't and you can make some educated guesses about how to go about changing it.

,_, takes sooooo long to explain...
also sharp eyes will notice i worked backwards on some of that, it doesn't change the validity of it.

edit: nagamaki posted some details of the procedure above, i bet he had as much fun typing it as i did.

[Sidanoir]

That was a very good explanation thanks..

I suppose as you said pdif of 1 on Fafnir is...um ya not going to happen.....too bad we can't use some other sort of known point to solve or can we? I don't feel like taking out pen and paper and making it like im back in college <.<

[mikazuki]

you can fairly easy get pDIF = 1, ask your native brd/blu about their minne+cocoon special!
i'm prepared to bet mobs don't have atk over 999 (or even close to).

the problem is doing it on fafnir, i can't imagine anyone would care enough to try testing it on that mob, however you should get similar ranges of atk for various wyrms.

edit: yeah i think you understood the mob was the problem rather than the statistic so i'll just uhh apologise^^

[Therin]

I've seen dozens of those rings around on Fairy, so they can't be that rare. >_>