Blockout II Randomizer
Posted: Fri Jul 13, 2007 4:53 pm
@jp_38: i have noticed that success in those
3x3x[6|10] extended modes depends strongly on the luck getting a nice random sequence of blocks. one \"just\" would need 100 games to have 2 or three games with an easy, playable sequence.
now, in 2d tetris there is a lot of randomizer theory: http://www.tetrisconcept.com/wiki/index ... _Generator
the simplest to implement random generator is the bag randomizer:
you fill a bag with all possible pieces (depends on mode: flat, basic, extended - in classic 2d tetris 7 tetrominoes),
and then shuffle that bag.
now you take pieces out of that bag until the bag is empty, then you start over filling the bag, shuffling and so on.
the big advantage of this is that you cannot get any unfairly easy sequences by just playing over and over again.
i think this bag randomizer would make scores more comparable, giving better competitions on the highscore list. also i think that it wouldnt break compatibility with the original blockout.
ah and i would suggest making the bag twice the size of the number of different pieces - this gives an intermediate solution between purely random (with the probability of getting 10 easy pieces in a row) and the sequence of stricktly all possible pieces coming in a run, but in shuffled order.<br><br>Post edited by: Herc, at: 2007/07/13 18:56
3x3x[6|10] extended modes depends strongly on the luck getting a nice random sequence of blocks. one \"just\" would need 100 games to have 2 or three games with an easy, playable sequence.
now, in 2d tetris there is a lot of randomizer theory: http://www.tetrisconcept.com/wiki/index ... _Generator
the simplest to implement random generator is the bag randomizer:
you fill a bag with all possible pieces (depends on mode: flat, basic, extended - in classic 2d tetris 7 tetrominoes),
and then shuffle that bag.
now you take pieces out of that bag until the bag is empty, then you start over filling the bag, shuffling and so on.
the big advantage of this is that you cannot get any unfairly easy sequences by just playing over and over again.
i think this bag randomizer would make scores more comparable, giving better competitions on the highscore list. also i think that it wouldnt break compatibility with the original blockout.
ah and i would suggest making the bag twice the size of the number of different pieces - this gives an intermediate solution between purely random (with the probability of getting 10 easy pieces in a row) and the sequence of stricktly all possible pieces coming in a run, but in shuffled order.<br><br>Post edited by: Herc, at: 2007/07/13 18:56