Hi, everyone. I haven't posted here in over a year. Glad to see there is still activity here. I'm learning about the math behind tetromino distribution (how the game chooses which tetrominos to give to the player). According to the Tetris Wiki, all games after 2001 use the Random Generator, except for the "Tetris Grand Master" series, which uses its own system, called the TGM randomizer. So, what about the games made before 2001, like the Game Boy and NES versions? How do their engines work? Are there any good sources that I can read about them? Thank you in advance.
I've often wondered this myself. It seems like in my experience that the I piece is slightly less common than the others. It is far more often the smallest number in my stat count at the end of a game than it is the largest number. Also never in history can I recall getting 4 in a row of the I, though I know I've had 4 in a row multiple times of every other piece available. The highest I've ever had an individual piece's stat reach while still having another piece at 0 is 15.
The NES randomizer is documented in the wiki here. (For future reference, the Tetris Concept and Hard Drop copies of the wiki are both more up-to-date and ad-free. The Tetris Wikia is a stale copy of what was originally written by members of the forums.) There was some discussion of the Game Boy Tetris randomizer in the "Randomizer Theory" thread, starting here. It is followed with a bit of an info dump I posted on theoretical NES stats.
Interesting stats. I wonder how they compare to the engine's intentions. I want to analyze Sega Tetris in a similar way but I'm not sure how. All I know so far is that it definitely does not use the "bag of 7" system that the modern games use. That thread is just what I was looking for. I don't know anything about bits or bytes, so most of the posts are gibberish to me, but I think I get the gist of it. So, "I" was intended to be the rarest tetromeno, but it ended up being "L" due to a programming error, and you guys were the first to discover this? That's amazing. What really intrigues me is the probability distribution posted by Xkeeper. Would it be possible to make something similar for every version of Tetris?
For GB, I think they intended the pieces to be weighted roughly equally, but failed on all counts due to a programming error and a bad choice of "randomness" source. For NES, they had a second chance and more or less succeeded -- it's not perfectly flat across the board, but it gets the job done well enough.
Hi, not sure where to post this but it seems partly related to this topic. I have been playing the original UK Game Boy Tetris since its release and I have several queries about it, including its randomiser. (I see that the previous post says that it is useless.) This forum and wiki seems mainly geared towards the NES version. Is there a thread/ forum that deals with the original Game Boy version or should I post my queries on this thread? Thanks.
