Crashing the game was in fact the goal. It was discovered by using a bot that the game would eventually bug out and start trying to read tile data from the RAM, which gave a chance for the game to crash. No human player had been able to reach this game crash until now, which is why it’s a big deal. It’s the first time someone has technically beaten Tetris, as normally every single player will eventually top out and lose either due to mistakes or bad luck with the pieces.
It was posted yesterday with a different video that goes more into details what this is all about, the old records and how these were achieved and what the “true killscreen” (this crash) is, etc. I didn’t think it would actually be that interesting, but I watched the whole thing and quite enjoyed it: https://www.youtube.com/watch?v=GuJ5UuknsHU
I saw the post yesterday, but didn’t watch the video at the time.
Your comment convinced me to watch it, and, boy, was that more interesting than I expected.
One question remains: what are those gloves they are wearing? I get that it’s something that helps with the rolling technique, but is it just for abrasion/chafing prevention? Or does it do something else?
I’m guessing the gloves provide low and consistent friction theoughout the game. It’s probably easy to get either sweaty or dry fingers during a long run, which could probably ruin some precise movements.
I haven’t read about the specifics of this Tetris crash, but usually what happens with these old games is that memory is very tightly packed, imagine you have a small version of Tetris that has 3 digits XYZ where X is the speed of the game, Y is the amount of lives and Z is the level you are in, so for example if you’re in speed 5 with 8 lives on level 7 the number would be 587, if you go up one level it becomes 588, now on that example if you’re on speed 9 with 9 lives on level 9, i.e. 999, and you go up one level the number becomes 1000, but because only the 3 last matter you’re now on speed 0, with 0 lives on level 0, since speed zero means nothing moves you crashed the game.
Again, this is not exactly what happened here, but probably something similar where increasing a number overflew to the next one in memory and that caused some weird behaviour.
Yes, I completely understand that those games that ran on a shoestring can easily crash when some values are exceeded. What puzzles me is that I would have expected the player to be annoyed at his game crashing (of course simpler games on dedicated hardware didn’t really get to crash all that much back then, so maybe that was seen as an achievement of sorts).
I suppose it’s my lack of exposure to the console side of things, having gone from 8 bit PCs to the ubiquitous intel machines without ever using one of the dedicated gaming devices.
There’s been a moving concept of the ‘killscreen’ in tetris. Pacman has a limit where the game isn’t any longer playable. Tetris only got to a certain speed and was too fast to progress until new tech for hitting buttons were discovered. Recently, someone found out that after a certain level some conditions would crash the game so people have been racing to meet those conditions.
The crash was known, it’s been reached by a TAS but no human had gotten far enough to trigger it. He was intentionally trying for the crash to be the first one to do it.
Tetris doesn’t really have an end. It just keeps going. So this is a very specific crash where if you get far enough into the game, it can’t keep up with the player any more. You “beat” Tetris by playing so well you make the game break.
This is similar to getting pacman to crash by beating level 255 at which point incrementing the level goes past what can be stored and the data gets corrupted.
It looks like the game just crashed. And he looks happy rather than annoyed.
Is that what beating Tetris is? I thought it would maybe run out of cubes after a while, or something.
Crashing the game was in fact the goal. It was discovered by using a bot that the game would eventually bug out and start trying to read tile data from the RAM, which gave a chance for the game to crash. No human player had been able to reach this game crash until now, which is why it’s a big deal. It’s the first time someone has technically beaten Tetris, as normally every single player will eventually top out and lose either due to mistakes or bad luck with the pieces.
That makes sense. Thanks for the extra info.
It was posted yesterday with a different video that goes more into details what this is all about, the old records and how these were achieved and what the “true killscreen” (this crash) is, etc. I didn’t think it would actually be that interesting, but I watched the whole thing and quite enjoyed it: https://www.youtube.com/watch?v=GuJ5UuknsHU
I saw the post yesterday, but didn’t watch the video at the time.
Your comment convinced me to watch it, and, boy, was that more interesting than I expected.
One question remains: what are those gloves they are wearing? I get that it’s something that helps with the rolling technique, but is it just for abrasion/chafing prevention? Or does it do something else?
I’m guessing the gloves provide low and consistent friction theoughout the game. It’s probably easy to get either sweaty or dry fingers during a long run, which could probably ruin some precise movements.
It’s just my guess, though.
I haven’t read about the specifics of this Tetris crash, but usually what happens with these old games is that memory is very tightly packed, imagine you have a small version of Tetris that has 3 digits XYZ where X is the speed of the game, Y is the amount of lives and Z is the level you are in, so for example if you’re in speed 5 with 8 lives on level 7 the number would be 587, if you go up one level it becomes 588, now on that example if you’re on speed 9 with 9 lives on level 9, i.e. 999, and you go up one level the number becomes 1000, but because only the 3 last matter you’re now on speed 0, with 0 lives on level 0, since speed zero means nothing moves you crashed the game.
Again, this is not exactly what happened here, but probably something similar where increasing a number overflew to the next one in memory and that caused some weird behaviour.
Yes, I completely understand that those games that ran on a shoestring can easily crash when some values are exceeded. What puzzles me is that I would have expected the player to be annoyed at his game crashing (of course simpler games on dedicated hardware didn’t really get to crash all that much back then, so maybe that was seen as an achievement of sorts).
I suppose it’s my lack of exposure to the console side of things, having gone from 8 bit PCs to the ubiquitous intel machines without ever using one of the dedicated gaming devices.
There’s been a moving concept of the ‘killscreen’ in tetris. Pacman has a limit where the game isn’t any longer playable. Tetris only got to a certain speed and was too fast to progress until new tech for hitting buttons were discovered. Recently, someone found out that after a certain level some conditions would crash the game so people have been racing to meet those conditions.
The crash was known, it’s been reached by a TAS but no human had gotten far enough to trigger it. He was intentionally trying for the crash to be the first one to do it.
Tetris doesn’t really have an end. It just keeps going. So this is a very specific crash where if you get far enough into the game, it can’t keep up with the player any more. You “beat” Tetris by playing so well you make the game break.
This is similar to getting pacman to crash by beating level 255 at which point incrementing the level goes past what can be stored and the data gets corrupted.