I wasn't writting a design document. I don't see how you rest your case?
If I was. I would have detailed how.
Namely. It assumes the player makes no moves, which is how it gets more and more wrong the farther you look ahead. If the player has already locked in other characters, it processes their actions and stores them to enact between rounds. If they haven't locked in their moves, it assumes no motion on their part. The AI then 'plays' the next round like it would should the game be playing the next round. It holds the outcome and repeats for however many rounds. Along the way, it stores random values for random happenings, so should the same random calls be made when the player actually gets that far into the battle, it can repeat with that random value, making the actual future and the guessed future more consistant. When it reaches the target time-frame it 'ghosts' the new location of the enemy, and shows ghosted values over all the characters (PC and Enemy) for their new values. Also showing ghosted versions of the status effect values. Thusly, showing what the future might hold.
Best if used to look forward in the future by 1 or 2 turns when you have a PC and an enemy who are likely to continue attacking eachother in the comming battles. In this instance, it will correctly predict the next movement of the enemy. Say, the enemy is going to step back and heal himself, the player can chose to use that information to his advantage. It might, show that the enemy plans to unleash the strongest attack skill in the game, in which case inflicting 'silence' or something else that will interrupt the skill might be the best course of action. As might be retreat.
I DON'T just say things without investigating how one would do these things.
Do you want me to detail what kind of AI it is? That isn't important here, seeing as it can store the values of, and use before-hand, almost any 'kind' of AI.
If you were saying it isn't possible... that's silly.
OOOoorr... Were you saying a 'psychic' was done in another RPG? Which? I've played a LOT of RPGs...
If I was. I would have detailed how.
Namely. It assumes the player makes no moves, which is how it gets more and more wrong the farther you look ahead. If the player has already locked in other characters, it processes their actions and stores them to enact between rounds. If they haven't locked in their moves, it assumes no motion on their part. The AI then 'plays' the next round like it would should the game be playing the next round. It holds the outcome and repeats for however many rounds. Along the way, it stores random values for random happenings, so should the same random calls be made when the player actually gets that far into the battle, it can repeat with that random value, making the actual future and the guessed future more consistant. When it reaches the target time-frame it 'ghosts' the new location of the enemy, and shows ghosted values over all the characters (PC and Enemy) for their new values. Also showing ghosted versions of the status effect values. Thusly, showing what the future might hold.
Best if used to look forward in the future by 1 or 2 turns when you have a PC and an enemy who are likely to continue attacking eachother in the comming battles. In this instance, it will correctly predict the next movement of the enemy. Say, the enemy is going to step back and heal himself, the player can chose to use that information to his advantage. It might, show that the enemy plans to unleash the strongest attack skill in the game, in which case inflicting 'silence' or something else that will interrupt the skill might be the best course of action. As might be retreat.
I DON'T just say things without investigating how one would do these things.
Do you want me to detail what kind of AI it is? That isn't important here, seeing as it can store the values of, and use before-hand, almost any 'kind' of AI.
If you were saying it isn't possible... that's silly.
OOOoorr... Were you saying a 'psychic' was done in another RPG? Which? I've played a LOT of RPGs...