#INSTRUCTIONS FOR ROCK PAPER SCISSORS HOW TO#

To take advantage of how the AI thinks, we can easily get around it by finding out how far it thinks ahead. Another thing to keep in mind is that the AI increases number of turns if it is losing, or so it seems. This seems counter intuitive but helps with shortening the formula.ДЄlso keep in mind that the AI thinks ahead further than 6 turns its just that the pattern repeats. You might also notice that the player turns (and AI turns) are losing hands to their previous respective turns. > pa -> -> ra -> -> sa -> ДЄs you might notice there is a repeating pattern which makes short work of thinking ahead. It is limited to 6 turns because after the AI thinks 6 turns ahead on the 7th turn is essentially the 1st turn. How far the AI thinks ahead ranges from 1 - 6 turns. The AI indeed thinks many times ahead but in a sense this may go on forever so there is a limit to how far it thinks ahead (More on this soon). If AI won with Rock, it might not only thinks player will choose Paper, but also think the player knows the AI will choose Scissors and will choose Rock to beat Scissors. However this seems far too simple and short sighted so the AI also thinks further. If AI won with Rock in a round, the AI will think the loser (Player) will choose Paper in the next turn, solely because Paper beats Rock (the previous winner), and thus choose Scissors to beat the player. The AI will think what the loser of the previous round will choose, in its logic: the hand that beats the previous winning choice. The AI decides by choosing who won, and in its own head thinks in turns (between player and AI) which hand will beat the previous hand. So logically to win a round you always pick the best one, however there is no best choice, I guess the developer knew this and instead of making the AI choose a completely random hand they added logic to the AI. This is a very well balanced game where all choice has an equal chance of winning, as there is always 1 way to win and 1 way win (and 1 way to tie, but from my testing and in terms of game design there's no tie, instead the AI always win: AI will cheat and always beat player's choice). Rock beats scissors, which beats paper, which beats rock. The most simple way is to explain something we already know. Now let me tell you about how AI decides its logic. However the second round the AI makes a decision based on who won, what choice won and how far it is allowed to predict (this depends on few factors, I could be wrong though). To my understanding the initial first round is completely random (or so it seems) and is impossible to guess what the AI will choose. The AI in essense thinks a head and decides which r/p/s to choose depending on the out come of the previous round in a session of the mini game.

#INSTRUCTIONS FOR ROCK PAPER SCISSORS HOW TO#

To get to grips of how to almost always win against AI opponents you just need to know how the AI thinks and apply it into simple algorithim. or just raw brain power) that the AI plays the mini game in a logical manner. In D:OS the mini game of rock paper scissors seem apparently random at first but if you were like me you would have discovered (purely by chance. AI plays 1 to 3 rounds of the same hand, but then changes hands on the next hand (player need only be 1 step ahead), pattern repeats player only need to play x step ahead, no matter who wins (that includes 0 steps: AI plays same hand always) If player wins, player needs to increase 1 step more starting from 1 on the first win Pattern varies depending on character charisma If tie the AI changes to a winning hand (behind curtains) denotes theory, hasn't been tested or wouldn't be able to test. Once you've understand how my 'step' method works let's move on to patterns that AI almost always seem to follow. AI won with Rock, the very first step is Paper. Player won with Rock, the very first step is Scissors. The first 'step' is in fact what the player might need to play to beat the winning hand that won the previous round (if player won), or against the hand the AI might play to beat the previous round (if player lost). Let's discuss how we find the very first 'step'. Doesn't make sense? Bear with it for the mean time. One 'step' is the hand that would lose to the previous 'step'. Now, I'll have to explain a term that I use for myself: 'step'. However if you just want to skip all that and get right to the point this is the section. The explanation to how I reached this conclusion can be read further down. I'm pretty sure AI has a pattern that with only a few tries, or in one go, can guess what the pattern is.

Instructions for rock paper scissors