Author Topic: Rare Item RNG Patterns Published  (Read 34154 times)

FF12GrandMaster

  • Guest
Re: Rare Item RNG Patterns Published
« Reply #30 on: August 18, 2009, 08:34:44 PM »
I'm still working on it and have made more discoveries which will be published soon in an updated guide. :)


Chaos

  • Global Moderator
  • Order of Ambrosia
  • *****

  • Offline Offline
    • View Profile
Re: Rare Item RNG Patterns Published
« Reply #31 on: August 18, 2009, 11:27:18 PM »
Has anyone else actually gotten the pattern I mentioned in the other thread?

"Most people think time is like a river that flows swift and sure in one direction, but I have seen the face of time, and I can tell you - they are wrong." ~ Prince of Persia

LordBalthier

  • Moppet
  • *
  • Mood: Energetic Energetic

    Offline Offline
    • View Profile
Re: Rare Item RNG Patterns Published
« Reply #32 on: August 19, 2009, 03:28:21 PM »
Did some testing with the ps3 last night, and so far I've gotten the chest to spawn but I can't seem to come across the right combo to get the spear....I was wondering, does getting hit by the abysteels change the rng?? I'm thinking that it does, and say if I get hit twice in the beginning by them how would I go about getting myself back on track for the 5-hit combo??? casting curaga 3-4 times???

---And so began the story of the wanderer, the vagrant.


FF12GrandMaster

  • Guest
Re: Rare Item RNG Patterns Published
« Reply #33 on: August 19, 2009, 09:29:55 PM »
There are different RNG patterns, quite a few of them, and you need to be able to identify them so that you know whether you're in the right one or not. Casting cure/curaga may well put you on the right track but you need to know where you are as far as the patterns are concerned. Without that knowledge, casting cure is meaningless as you don't know how to identify the good pattern. Many hours of research would be needed to identify all the patterns that there are and then experimentation would be needed to find out how to shift to the right one.


mkotha

  • Guest
Re: Rare Item RNG Patterns Published
« Reply #34 on: August 21, 2009, 11:40:59 PM »
Hi,

I have recently tried to work out how the game's RNG behaves, and got some result. First it is found that the main RNG is based on the algorithm called Mersenne Twister, with seed 4537. Since the output sequence of the RNG is completely known, now i can understand for example why a 5 hit combo leads to a rare item in a chest. I also established a method for identifying current RNG position with 2 or 3 damage samples.

I published a text describing what I found, in Japanese:

http://www.kotha.net/ff12random/random-1.zip

Is anyone here interested in this kind of information? If there's some interest, I'll be happy to translate the text into English and publish it.


FF12GrandMaster

  • Guest
Re: Rare Item RNG Patterns Published
« Reply #35 on: August 22, 2009, 09:38:32 AM »
I would love to see it translated. I am very interested.


pprincess

  • Knight of the Round
  • *******
  • Mood: Smart Smart

    Offline Offline
    • View Profile
Re: Rare Item RNG Patterns Published
« Reply #36 on: August 22, 2009, 09:42:20 AM »
I am very interested also. Welcome, welcome, share anything you would like. I am very happy to meet you and look forward to learning more about your discovery.


AstriBlair

  • Vanguard
  • ****
  • Behold,the truth and lies that it keeps behind..
  • Mood: Relieved Relieved

    Offline Offline
    • View Profile
Re: Rare Item RNG Patterns Published
« Reply #37 on: August 22, 2009, 12:59:25 PM »
I too am Interested.Hope more news will come soon.



Life is never easy for Game-addicted humes

flak76e

  • Rear Guard
  • ***

  • Offline Offline
    • View Profile
Re: Rare Item RNG Patterns Published
« Reply #38 on: August 22, 2009, 08:09:11 PM »
Me too. Lets just hope there will never be a way to get drops 100% Because that will ruin everything + chaining is fun.


Relapse

  • Guest
Re: Rare Item RNG Patterns Published
« Reply #39 on: August 29, 2009, 07:55:00 PM »
Here's a non perfect translation of mkotha's guide, hope it helps:

For observations FF12

Overview

FF12 because you are dealing with random numbers to investigate what is inside, I learned writing.

Contents

Overview
Conventions
Note
Introduction
A characteristic of the generator
A common usage pattern of the series
A process using a series

        * Gun
        * Unarmed
        * Ax, hammer, attack damage calculation Handibomu
        * Kearu, Kearuda
        * Potion, Haiposhon, Ekusuposhon, ethers, ether-high
        * Opening of the Treasure
        * Load saved data

A series of random actions do not consume
Applications

        * Locate random
        * 5hits principles of law
        * "Plateau Serobi / Hara months crossing" Treasure of law POP

Can not know

Conventions

In this article follow the rules below.

    * With ? x ?, x represents the greatest integer less of.
    * x% y in, x to y represents the modulo. x % y = x - y * ?x / y? x% y = x - y * ? x / y ?

Note

In this article, FF12 and observe the rules of randomness in which describes how to control it. Play like this is probably not the intention of the developer, you can use to casual game off balance. I have read this sentence in FF12 as well as the unfit liability.

This sentence is so poorly written verification on the whole, well aware that there may be an error.

Introduction

Computer games in general, if the random numbers required in the course of treatment, instead of random numbers used to calculate the value derived in accordance with certain procedures. These are called pseudo-random number. Pseudo-random number, random, so the steps are made (the disordered) and are not designed to have no problem forgiving enough regularity on the fact considers the random. What you want more information googled.

FF12 is that they have at least three pseudo-random number generators, they operate independently. One of which is particularly important, damage calculation and determination of the attack hit, and determination of the contents of the treasure, which regulates many of the events of the player-friendly touch. A generator make sure to call it. A deal is a little less than the generator only.

A characteristic of the generator

A generator is less than 0 or 2 32 (approximately) one can generate random integers at once. Each time you generate the internal state of A is an integer that will change constantly, and immediately after power-up, and immediately after the hardware reset, A is the internal state of a certain value. By this, A generates the same sequence of random numbers every time. This random sequence can be calculated as a specific sequence, the first section, for example, 1820836235, and 1372235862 is the following. In other words, to generate random numbers for the first time after power-up, A will always generate 1820836235.

Therefore, A If you want to predict the future output, it is after power A for "many pieces of" what I knew enough of the random. In other words, A is fixed to the inside (of infinite length), and it has a table of random numbers, behave as if you're using one from each of the top reading is required each time a random number. Also, if you ever had A n generates a random number of pieces, the next generation of the series on A (n +1) is equal to the second. That this current position on the table of random numbers to believe that n, n a (for series A) to be called a random position. In addition, if the position of k-random process that went through that process, "the random number k in total consumption," and expressing.

A difficulty in reading the following series Ue be calculated without knowing how or in concrete, but to write and keep people interested, A is the Mersenne Twister (mt19937) is calculated by Ru. Initialization routine, however the latest version of the year rather than 2002, using 1998 versions of those years. seed in 4537 (4357 I do not ya!).

A common usage pattern of the series

A 0-2 series is a value ranging from 32 -1 to be, in fact, often a smaller value is required. For example, n if you choose one at random from each choice, FF12 are widely used in the method is to use the remainder. Specifically, it generates a random one, it takes a modulo n. Then, from 0 to n-1 obtained in almost equal probability to the integers.

As an application of it, "k% chance of doing something" in that process. This is one of the random numbers generated (v to) and, v% 100 to do it only if k was less than. v% 100 from 0 to 99 because it takes up to a value almost equal probability, v% 100 <k k% probability will be met and about. This procedure is used so extensively, "(k of accuracy)% determined" to be called by name.

 K% accuracy of decision:
   A single generating a random number series, v Tooku.
   v% 100 <k determine if success, otherwise fail decision.

A process using a series

The following series of actions in handling a variety of attacks and magic and A is used to write about or how the random numbers. Regarding the survey, the state is absent around the enemy, do not remove all the effects related to the investigation status on who has to perform an action for friendly forces. If you act against the enemy in combat is not necessarily just might not fit.

Gun

And gun attacks, if you hit it, eight random numbers to individual consumers. Breakdown is as follows.

Knock-back rate calculation (2)

    Now generate a random number 2, v0, v1 and. Knock-back rate is less.

     Knock back knock back rate = value of arms (10) + v0% level of the attacker - v1% level of the target (negative result when the calculation is 0)

    This is slightly different from that found expression in Batoruarutimania, probably right here.

Knock-back Tests (1)

    % Accuracy and to determine the rate of knock-back. If your attack is successful even if you do not knock back, but the decision is to be performed.

Non-decision errors (1)

    Accuracy and error rate% for non-decision. Non-normal error rate is 100, determined to succeed, but always, a random number is still consumed. Fails to determine if this is not done the following decision to quit just four random consumption.

Critical Decision (1)

    Critical factor (5)% and accuracy to make decision.
 
Damage calculation (2)

    First, the reference value is given, and nine times higher than the reference value 1 / 8 procedure twice to get a random value less than rand98 defined as follows.

     rand98 (reference value):
       Reference value 0, and 0 without spending a random result.
       Otherwise, each one generates a random number, v and. Provided the following results.
       Result = threshold + (v% ? threshold * 12.5 ?) / 100

    In short, in the range from 0.01 Possible integer multiples of the results and pick one at random.

    Rand98 gun damage is calculated using the following.

     rand98 (attack value) is executed, the results and r0.
     rand98 (value attack) to try again, with the result r1.
     Determine the damage as follows.
     Damage ratio ? = ? r0 * r1 *

   Where "ratio" is the weather on the HP and attack power attributes fill that UP is multiplied by all the options and other compensation.
Tests added additional effect (1)

    As% accuracy rate to determine additional effects occur. Unchanged even if you consume the random bullets that are not equipped with additional effects.

To tell the character of gun attacks, and to cancel the attack and immediately entered a motion to change equipment attack, only four random numbers are consumed.

Unarmed

The attacker is unarmed, is consumed in a single random number usually 10. If you experience a sequence n hits of the hammer, if you hit all consumers will total 9n +12.

Now the 10 consumed is divided into each of the five halves. If you experience a hammer sequence is repeated only a few minutes sequence determination of total 5??late. The following is a breakdown.

Knock-back rate calculation (2)
Knock-back Tests (1)

    This is the same for two guns.

Unknown (1)

    But one that consumes a random number unaccounted for.

Hammer binary decision (1)

    % Accuracy and to determine the sequence??. ??sequence is calculated by the following formula.

     Ren Ren = ?????* 0.7 ? (if not equipped with a forearm of Genji)
     Ren Ren = ?????* 1.8 ? (if equipped with a forearm of Genji)

    If successful in this decision, and M to determine the accuracy and 11 times (number of successful +2) hits a sequence of hammer occurs. 11 but also to determine the success of all 12 hits occur only once. M is defined as follows.

     M = 8 (? max HP / 4 ? ? now HP)
     M = 16 (? max HP / 8 ? ? current HP <? max HP / 4 ?)
     M = 32 (? up here HP/16 ? ? HP <? max HP / 8 ?)
     M = 64 (current HP <? up HP/16 ?)

    If you encounter a sequence n hits of the hammer, the determination of the second part is repeated n times, during the second half and the second part of (n-1 times in total), and time-consuming four unaccounted for random numbers.

The second part follows.

Decision guard shield (1)

    Shield to avoid% accuracy rate to determine. If successful, the following conditions exist.

Decision guard weapons (1)

    And avoidance of arms% accuracy rate to determine. If successful, the following conditions exist.

Decision Parry (1)

    Parry% accuracy and to determine the rate. successful, the following conditions exist.

Non-decision errors (1)

    Accuracy and error rate% for non-decision. f it fails, not less.

Damage calculation (1)

    Damage calculation procedure is as follows.

     rand98 (attack value) is executed, the results with v.
     Determine the damage as follows.
     Damage = ? (v - value defense) * magnification ?

    Where "ratio" is a basic ratio (force * (power + level) / 256) multiplied by the ratio of various correction.

Ax, hammer, attack damage calculation Handibomu

First, the real number x is given as a reference value, 0 or x procedure to generate a random value less than rand11 defined as follows.

 rand11 (x):
   x <0.01 at 0 and the result, without consuming a random exit.
   Otherwise, each one generating random numbers (v and), and provided the following results.
   100 Result = (v% ? x * 100 ?) / 100

Using this definition, the damage these weapons will be calculated in the next step.

 rand98 (attack value) is executed, and a result.
 c = (a - the value target defense) and provided. c <0, then damage is 0, exit without consuming any more random.
 Otherwise rand11 (ratio) is executed, the results with r. Damage determined as follows.
 Damage = ? c * r ?

Where "ratio" means (force * (energy + level) / 128) multiplied by the correction value basis of various magnifications.

Kearu, Kearuda

Only one individual consumes a random number used to calculate the amount recovered. Computational procedure as well as unarmed damage calculation. But instead of the value of magic attack power, defensive value of 0 instead of using the basic magnification (magic * (magic + level) / 256 + 2) as calculated.

Potion, Haiposhon, Ekusuposhon, ethers, ether-high

Consume only a piece of random numbers. Whereabouts are unknown.

Opening of the Treasure

You examine the treasure, Gill made the first decision (accuracy varies depending on the Treasure). Following process is followed.

Determine if successful Gill
    If a fixed amount that is set to Treasure, Gil get a random value without consuming it. Otherwise, we generate a random number each one (v and), (1 + v% maximum) to get the gil.
Gill failed to determine if
    To determine non-rare. Accuracy, if a diamond bracelet that features a 90, if not 50. Where "non-rare" and refers to what is written in the left column of the table described Shinarioarutimania Treasure.

Load saved data

By definition, random positions of power is always zero, but is consumed by hundreds of random numbers can be manipulated to load the saved data. This time, even the same data multiple times and try to load completely at random position is not certain when it becomes operational. "Plateau Serobi / highway rest station" I tried saving the data in random locations were within the range of approximately 570 ± 10. But untested, this behavior may vary depending on model of the body. The PS2 is used in this study SCPH-70000.
A series of random actions do not consume

    * Opening the menu, change equipment, replacement of members battle to obtain a license but we also do not consume any A series of random numbers.
    * Na bra, Bokyaru, each Desuperu magic is not consumption but also run a random one. Also, (as many as tested) all actions are random and do not consume a complete stop before running charge. Meanwhile, according to Batoruarutimania, to calculate the charge time is a random number that is used. Therefore, if my description of the Ultimania, when determining the time-series A charge will be non-random numbers are used.
    * Character, when you attack with a gun may emit a slogan. There are several types of the slogans, slogans or any issue that seems to have been chosen at random (but most likely not be the same slogans issued twice in a row). However, when consumed in a gun attack A random series of eight pieces is known to all applications, the decision of the slogans that are estimated by systematic random numbers are used.

For those who are interested, put an A-series we know about the pseudo-random number generator has been found except two. Are also using the linear congruential generator either. Recurrence equation of the linear congruential "x n = (x n-1 * a + b)% m" and you can write, one of which (a, b, m) = (6364136223846793005, 1, 2 64) at , for each term x n (x n / 2 32)% 2 31 to the output. The initial value is probably 1. The other (a, b, m) = (7 5, 0, 2 31 -1) in the output sections intact. The initial value is perhaps 305,420,679.

Applications

Locate random

During play, if you can identify the exact location of that point random numbers in any way, by adjusting the position of intentionally random, can be controlled to some extent the attitude of random numbers that follow. For example, if you wanted to advance the position of 100 random pieces, a gun and shot 12 times for them (on the attributes taken to prepare to die and absorption), the better it Kearu?Ere four times then.

The estimation of random location, convenient to use the numbers will recover the amount and the amount of damage. A calculation is given in random sequence so completely, for example in advance for each random position "when the amount recovered in Kearuda He advocated that position," to every table to calculate the actual two or three times chanting Kearuda recover the amount from the table to find the sequence of it. "All random position" to make a table covering is of course impossible, in fact, almost 10 suffice if you make a table of random numbers less than the range of positions worldwide.

The idea is simple but this work, creating a table to calculate by hand is impractical because a large amount of computation. To assist in this task simple JavaScript included so that the tool you want to try if you're interested.


Relapse

  • Guest
Re: Rare Item RNG Patterns Published
« Reply #40 on: August 29, 2009, 07:56:15 PM »
5hits principles of law

5hits law and was originally discovered in IZJS approach, which is a high chance that you can get rare items from the treasure by following the following broad steps.

   1. Near the treasure to save to a hardware reset.
   2. Load and the Treasure of the target zone change until POP repeated. However, even if you do not repeat the start some POP.
   3. Repeatedly beat him. In this case, a shield or guard to prevent disconnect, be equipped with a cameo belt.
   4. If you encounter a sequence of hammer 5 hits, you want to open the Treasure equipped with a diamond bracelet.
   5. As a condition to being attacked by the enemy must not open the treasure.

To sum up, this technique, a random position at the time of opening the treasure that 1445 has been fixed to work with. The A series has been around as follows.

 1444: 2,510,889,783
 1445: 1,861,786,791
 1446: 1,379,767,799
 1447: 80,686,868

Positions 1445 and 1446 under two-digit value of 91 and 99 respectively, and both great. This is a fairly large percent accuracy that a decision to fail. In fact, when I opened the 1445 random locations in the Treasure State, rare and non-decision decision Gill maximum accuracy, so both 90 and failed to ensure, as a result of getting rare items. However, IZJS transparent armor will determine the accuracy of the 95 non seems to be rare, compared to 99 so you still fail.

5 hits consecutive hammer itself is a random position 1388 occurs when you attack with bare hands.

 1387: 382,760,416
 1388: 2,267,807,233 attack here; one knock-back probability calculation
 1389: 1,886,692,943 knocked back two probability calculation
 1390: 2,131,861,130 Decision 1391 Knock back: 2952945076 (unknown)
 1392: 2,341,692,401 hammer binary decision (accuracy 3) ? success 1393: 3,404,601,602??increase in the number sequence determination (accuracy 8) ? success 1394: 2,203,834,536??increase in the number sequence determination (accuracy 8) ? fail 1395: 961,042,754 increase in the number??sequence determination (accuracy 8) ? fail 1396: 2,005,317,402??increase in the number sequence determination (accuracy 8) ? success in 1397: 2,981,387,111??increase in the number sequence determination (accuracy 8) ? fail 1398: 2,833,431,843??increase in the number sequence determination (accuracy 8) ? fail 1399: 3,060,953,428 sequence determined increase in the number??(accuracy 8) ? fail 1400: 3,953,158,966??increase in the number sequence determination (accuracy 8) ? fail 1401: 3,263,761,202??increase in the number sequence determination (accuracy 8) ? success 1402: 161,153,939 increase in the number??sequence determination (accuracy 8) ? Fail 1403: 1,977,869,617??increase in the number sequence determination (accuracy 8) ? fail; 5 hits occurred in the 1404 serial hammer: 3808337799 determine Guard Shield (Accuracy 0) ? fail 1405: 3799551709 ... (snip) ...

5 hits the hammer for 57 consecutive random numbers so that individual consumption, the location is just 1445 random sequence will??.

The hammer does not occur if consecutive, random position every 10 inch forward beat yourself. Therefore, the position has reached the 5 hits in 1388 in order to impose the rank of just one of eight random positions seem to beat him not to start in the state. However, in fact, one digit is 0 or 1 there is also a good thing. If the rank of 10, position 1070 occurred in 4 hits in consecutive hammer, hammer at the end of 1118 so the sequence is located, since 1338 after 27 strokes to reach the self. If one is in place, the position of the hammer 751 in 4 hits generated sequence of random numbers are 39 positions total consumption will be 790. Even after 1070 times of 26 strokes to reach the self.
Thus, random position by 1388 is relatively easy to get shots repeated itself, 5 hits rarely causing a sequence of hammer that has a very good position to open the treasure after??random sequence addition, good and combine the properties of a rare position, 5hits method has the advantage of this special nature.

"Plateau Serobi / Hara months crossing" Treasure of law POP

Investigate whether the procedure will be determined based on the laws involved in how the random events, to introduce through examples. As an example, "plateau Serobi / original cross-months" are determined to find out whether or how the emergence of each of the Treasure.

The Shinarioarutimania p.122, according to whether the treasure will be determined to appear when you visit the zone. So, first, try to examine what is consumed when you enter some random number of months in the original cross.

First, after reset, then load the saved data highway rest stations, using random numbers to identify the location of the gun damage. Months and then crossed into the original, random re-locate, count backward random position after entering the zone. Random position by taking the difference between the two months when he entered the original cross was found that 16 of the total consumption is random.

The road back to a resting place, random measure the consumption of the same procedure at the entry again. Then and now only 6 random pieces that were not consumed. This is possible but is omitted from the emergence of Treasure decision does not meet the conditions for the reemergence of Treasure. Treasure months HARAUCHI occurrence of crossing points is 10, according to Ultimania, the first random number of individual consumption of 16 to 10 make sense to think of each other had been spent to determine the occurrence of each treasure.

Does the consumer think about how 16 was used to determine which of the emergence of the treasure which a single random number. According to the Ultimania is 10 incidence of the Treasure of all places is 50%. Therefore, the determination accuracy of 50% emergence of the Treasure the moment and try to assume that the decision has been made. Consumed about 16 random numbers each, the position accuracy of 50% and if you arrange the outcome of the decision made was as follows.

 nynnnnnynyynnyyn (y is a success, n is a failure)

Then, turning to actually cross the original months to see if any positions have emerged Treasure. Treasure was the fourth one to appear, in particular, with the number and called the Ultimania Treasure, 55, 57, 59, was 60.

Total consumption of 16 random numbers, random numbers were used to determine location of the treasure will appear on the assumption that explore sequentially. That is, assuming no consumption between independent random occurrence of Treasure decision. Target of 16% of the acceptance of the decision columns, length 10, just four columns of which one part of that success is only two ways.

 n [ynnnnnynyy] nnyyn and
 nynn [nnnynyynny] yn

If it is correct to assume here that most likely corresponds to the part of the decision of either of these treasure.

Data just because it is unlikely that any further say anything, save once / collect similar data for another case and reset. Repeat this a total of five times of data collection.

 0: nynnnnnynyynnyyn occurrence: 55, 57, 59, 60 (data first)
 1: nynynynynnnynyyny occurrence: 52, 53, 58, 59
 2: ynynynnnynyynynny occurrence: 51, 55, 57, 59, 60
 3: nynnyynnynynynynn occurrence: 51, 52, 57, 60
 4: ynyynnnnyynynnyy occurrence: 51, 53, 56, 60

First as a surprise to the consumer at the time of entering the original random cross 16 months if there is not total of 17 total. In this case, either before or after emergence of the Treasure of the portion determined to assume that there is an additional random consumption.

Under the above assumptions, found that only one can explain how consistently the results of random numbers (this check is made with the computer, I do not think you can not even human). Specifically below.

   1. The random numbers to consume one or two pieces (unaccounted for, one or two conditions are unknown or pieces).
   2. Random numbers to consume 10 pieces, used to determine the appearance of one each of Treasure. Determine the order of, 55, 53, 51, 52, 54, 58, 57, 56, 60, 59.
   3. The random numbers to consume five pieces (unaccounted for).

This total of 16 or 17 will be a random number of individual consumers.

Above, Treasure of the original cross POP months was able to make a hypothesis about the law, the following to verify this. I actually tried to reset again, the POP as had been mostly hypothetical situation (the original is probably more serious and should be verified, because suitable work was simply tired of the loose again). A conclusion to the Treasure of the ribbon to adjust the position of POP will be random, really close to get a ribbon.

The map of the original cross-months will be written as follows: determine the order of appearance of each treasure.

Map of the original cross months. Determined by the order from west to 9, 10, 6, 7, 8, 1, 5, 2, 4, 3

Can not know

The current situation is that few people know that immediately come to mind following points uninvestigated.

    * Batoruarutimania p.79, according to the state of the Kura Miya will receive twice the rate of compensation rate and Parry guard. However, try to attack with bare hands and run to the darkness who actually determine the guard shield weapon guard decision, simply fold the accuracy of two to three simple Parry seems to be unable to determine correctly predicted .
    * "Plateau Serobi / Hara months crossed" not certain because of the random number when you enter the consumer. It's kind of investigation is not linked to the weather.
    * For monsters drop items.
    * Random consumption of the city.


Larsa_Ferrinas

  • Hedge Knight
  • **
  • I JUST LOVE GAMES, LARSA RULES!!

  • Offline Offline
    • View Profile
Re: Rare Item RNG Patterns Published
« Reply #41 on: September 05, 2009, 04:11:08 AM »
WOOOOOOOOWWWW

humans will soon realize, after gold is theirs, after all trees is gone, after all desires is fuldilled, then after all those... they will realize that money cannot be eaten....

b4bodo

  • Moppet
  • *

  • Offline Offline
    • View Profile
Re: Rare Item RNG Patterns Published
« Reply #42 on: September 28, 2009, 12:17:51 PM »
I, am beginning testing for the PAL 60gb model ps3. should be interesting haha, seeing as you guys and gals have put so much effort into this, I'm going to have a crack at it and maybe shed a whole lot of light on this for fellow ps3 users. wish me luck, and patience. :-\ but first... i have to get 2 more espers on my ps3 data to access the special area in henne mines. easy as cake. i'll just level over night using the jelly exploit and should be able to take 'em out in no time. but i digress. again, wish me luck.

edit: btw this is the chechc02 model of the 60gb ps3 that im testing on.

« Last Edit: September 29, 2009, 04:20:55 AM by b4bodo »

Ergunno

  • Moppet
  • *
  • Love Games, Love Anime

  • Offline Offline
    • View Profile
Re: Rare Item RNG Patterns Published
« Reply #43 on: October 11, 2009, 03:19:03 PM »
RNG is 1 of the biggest helps I could've gotten in my Perfect Game. Got 6 Zodiac Spears way faster then without the method. Got lucky though. 5 runs for my first Zodiac Spear and I got a great combo. The others went quite smoothly also. Big props to FF12GM for discovering this :)


FF12GrandMaster

  • Guest
Re: Rare Item RNG Patterns Published
« Reply #44 on: October 13, 2009, 05:07:07 PM »
*Takes a bow*

Thanks, Ergunno, glad to be of assistance to everyone  ;D