Tetris gives you random tetrominoes, right?

Originally published at: https://boingboing.net/2019/09/04/tetris-gives-you-random-tetrom.html


Oh that’ll be unpopular alright. I don’t think masters of tetris need the game to be made easier or more predictable. There are people who play insanely well at instant drop alright.


Also, I don’t think there are theoretically unplayable sequences. Otherwise you wouldn’t be able to have an algorithm that did this:

Unless you mean the theoretical possibility of literally getting the same piece forever. I don’t know if we consider possibilities in the 10^-40 range.


Determinism could work if it was on a long enough timescale to where you’d really have to be an obsessive to notice the pattern.

That said it’d lead to being able to “solve” tetris which is not what anyone wants here I think.

1 Like

The OP suggests unplayability is theoretically possible but vanishingly unlikely, that it’s really just about how unpleasant it makes the game.

Even so, it’s not clear that the character-drawing version of tetris is dealing random pieces – and it’s using a large playfield. I figure the confined playfield of normal Tetris contributes markedly to floods and droughts being a gameplay problem!


Right you are. I added to my post:

A Tetris position should be analyzable in the same way as a position in Chess or Go. I admit that I might be halfway to inventing a tetromino board game that is obviously not Tetris, and is probably reinvented about monthly by someone already


Can’t seem to find it now, but there was a “hostile” version of Tetris that intentionally gave you the worst possible tetrominoe for the board given situation. A “left handed” hook when you need a right, etc. Getting more than three or four rows was a challenge.


Bastet! Linked in the post


Sorry, I missed it on the first reading. And I see why… I understand BB needs to have income, but some of those inline ads are very distracting. In my browser, the Bastet sentence appears immediately adjacent to one.


What? No XKCD links yet?

I’ll fix that :wink:


Off the top of my head, how about: for each column of the playfield, and for each tetromino, if any rotation of the tetromino can be dropped in this column without creating a gap, then add N copies of that tetromino to the bag, where N is the “depth” to which the tetromino will fall, or 1, whichever is greater; otherwise, remove one of that tetromino from the bag. Then remove a tetromino from the bag at random, and reset the bag.

This would mean there was no chance of a shitty tetromino, and a good chance of getting the piece you need, so it would be more satisfying and easier for bad players. But it would also challenge more advanced players, by making it less likely to get pieces that will fill in “undercuts”, and by making it possible to anticipate pieces.

Anyway I don’t think there’s much to be said for “randomness” per se, any more than Mario games would be improved by placing platforms at random instead of letting designers work out a really good place to put them. The tetromino-selection algorithm is pretty much the only degree of freedom designers have to shape the gameplay, so why not use it?

Random is just fine. Deal with it.
(says the guy who is enjoying daily use of an original Nintendo Tetris game on a CRT television)

" inventing a tetromino board game that is obviously not Tetris"

in a high school french class the teacher made us come up with board games to teach others french so I made a french trivia game where you had to get correct answers to move falling tetrominos.

turns out turn based tetris is amazingly bad.


I’m certain that I’ve come across references indicating that Game Boy Tetris will keep track of the number of times you rotate a given piece as an indication of the difficulty in placing that piece – and subsequently cause a “difficult” piece to occur more frequetly. But such references likely predate modern analysis methods.

Also, I guess we need this here for completion.

…Did you know that the spacecraft featured in the NES and Game Boy versions is not the American Space Shuttle? I had no idea. Makes a lot more sense.


I still need to make these tetromino-placing games I mocked up! !


cf. Atari Rampart


We implemented Bastet (dubbed “worst piece mode”) as part of Fire Tetris to keep the games short - it kicked in at the 3 minute mark. Basically this meant implementing a Tetris “AI” to solve the game for each possible “next piece” and figure out which harmed the player’s position the most when placed optimally. (It turns out it’s usually S or Z.)

One player still managed to play for 10 more minutes when he found a flaw in our AI (I wasn’t around that night so I don’t know the details unfortunately).



1 Like

Getting a good random function back in the 8bit days was a pain in the ass.

How about a bitcoin based on completed Tetris levels rather than power-wasting mining calculations?

1 Like

I would play the shit out of Invatris.