Meteorswarm wrote:Naurgul wrote:Meteorswarm wrote:I think the problem with first-person tetris is that it's still tetris, and I hate tetris. Too stressful. And it's been proven that, even with an ideal strategy, you must lose eventually, assuming random blocks.
Who the hell proved that? And how?
In other news, I've been replaying
Canabalt these days. It's
awesome unfair how it sometimes presents you with impossible obstacles.
For any size game, there is some number of R or Z tetraminoes that will fill the level, since you cannot clear anything with ONLY those. Any random sequence of sufficient length will contain that number (and more) R or Z tetraminoes. The key is getting only R or only Z for long enough to fill the level.
Disproof by counterexample.
Consider a game with only Z-tetraminos (my diagrams use S-tetraminos because I read the post wrong, but the same idea applies).
First arrange them like so (state A):

- tetris_a.png (278 Bytes) Viewed 3647 times
Notice that the second row gets destroyed, producing the following (state B):

- tetris_b.png (264 Bytes) Viewed 3647 times
Now add another bunch like so (state C):

- tetris_c.png (381 Bytes) Viewed 3647 times
Notice when it's collapsed you return to state B and thus, an infinite loop:

- tetris_b.png (264 Bytes) Viewed 3647 times
ALSO!
Robot Unicorn Attack is pretty much amazing and you should all play it.