This is a split board - You can return to the Split List for other boards.

Fiend colony (Maybe spoiler)

#11rickets987Posted 11/19/2012 2:32:32 AM
That's not really what I meant. Yes, there is a strong preference in the last ten years to implement RNG instead of either grind or skill-based difficulty. Game developers are taught in school that gamers will prefer a system where success is hurdled randomly than when they have to either invest hours into the system or genuinely be good enough at a difficult gameplay mechanic to succeed. It's sad.

What I was talking about is random seeding. This is when the block of code sets a random number (commonly based on the current time, down to far less than a second) and builds an algorithm that produces a string of pseudorandom numbers to generate an illusion of randomness. All games use this for RNG. Casinos that employ electronic gambling do this, too.

Old NES titles and GBA games have seeding algorithms which are not very complex. Because of this, a player who has mapped the system in Fire Emblem on the GBA can anticipate RNG so well that they know whether each attack is going to hit, miss or critical.

PS2 titles probably use much more complex algorithms. What is being called into question is whether the seeding can genuinely yield long strings of bad results. Say the max number is 100, for 100%: if it yields 1,5,18,60,3,23,43,etc., the results are strongly biased toward the negative side of things, while 78, 90, 83, 56, 3, 46, 97, 100, 82 is strongly biased toward the positive. Let's say you only get the Bushido Lore if the number is greater than 4%, so you only get the item if the result is a 97, 98, 99, or 100. In the former example, you have a far lower than 4% chance to get a Bushido Lore, but in the latter example, you will have been able to get it by the seventh try-- the chance is much higher than 4%.

When you're resetting the game, you're resetting the RNG seed. I'm questioning whether FFX-2's RNG seeding is actually so bad that it streaks in the way that I've explained. It doesn't seem to be, but it isn't as if I've thoroughly tested the system.
#12falconesquePosted 11/19/2012 8:07:25 AM
Thanks for clarifying. Probably the only person to look is pbirdman. Try a PM?

For myself, if I see that math the game is dead for me. I can't pretend I didn't see it. I know it's there, I know how to make practical adjustments as I need; that's enough.
---
Guides and other contributions: http://www.esque.com/slr/gamefaqs/
FFX, FFXII, KH, KH Re:CoM, KH2, Okage: Shadow King, Secret Agent Clank
#13timo4142(Topic Creator)Posted 11/21/2012 4:11:12 PM
rickets987 posted...
Has anyone actually proven that FFX-2 uses linear enough random seeding that a reset is required? It's an easy placebo, so..

I know for NES titles, the games really did run off predictable patterns of numbers which could or could not yield poor random number generation, but I have a harder time believing this is the case on the PS2.

In what little trial I've done for FFX-2, I couldn't see a difference between resetting and not for hunting chocobos on Kilika Island.


Well do you think for one thing that random seeding is based off of randis (Randis is random integers,randi is random integer). Some games still use random integers or rolling automatic dice scripts/labels to do this.

The kilika thing, I believe it is part of probability. If you have charm bangle, you have 0/100% chance to see a chocobo. If you don't have charm bangle or lure bracer, it's set to the default probability. Lure bracer I think might increase this probability (In some cases).
#14rickets987Posted 11/21/2012 9:22:30 PM
timo4142 posted...
Well do you think for one thing that random seeding is based off of randis (Randis is random integers,randi is random integer). Some games still use random integers or rolling automatic dice scripts/labels to do this.


I would lean toward no, but it's not as if I know. I used integers in my example just to simplify the explanation.

The kilika thing, I believe it is part of probability. If you have charm bangle, you have 0/100% chance to see a chocobo. If you don't have charm bangle or lure bracer, it's set to the default probability. Lure bracer I think might increase this probability (In some cases).


The Charm Bangle is having an effect on your chances to encounter a monster, not on the chances to encounter the chocobo itself. If a flag forced you into a random encounter, you could still encounter a chocobo. There is no reason why a Lure Bracer would increase your chances.