Breaking down the key strategies in Cheez's world record 100 line sprint NES Tetris speedrun. Cheez's channel: / @cheeztetris 100 Line Sprint Leaderboard: www.speedrun.com/tetrisnes My Twitch: / agamescout My discord server: / discord
I also pushed down on a few pieces to save a few frames, but that probably saved less than a second overall lol. Something to consider if the runs get super optimized though
@@EebstertheGreat I'm pretty sure it's one frame per pushdown. That means if you pushdown on one of the right frames ONCE per piece you save 250 frames, which is a full 4 seconds. You could also just stack higher, but that adds an extra luck element to the run, which a Tetris speedrun definitely does not need
I have long suspected that shooting for tetrises specifically was not critical. Unlike for scoring, where a triple scores significantly less than a tetris, for line clears, a triple is obviously 75% as good as a tetris.
In lower level speedruns, people usually score singles, doubles, and triples, but people can’t tap fast enough to make up the difference on 19. Only Cheez did it because he can roll
Line deletion takes time. This animation takes constant time regardless number of lines cleared. So with singles you have 100*x on animations when with tetrises you have only 25*x. At 4:39 he sais that animation time difference in these games is near 6 seconds.
@@QwDragon If you have a tree, Tetrises are less optimal, because you'll have to spend time to rebuild the plateau. Not to mention the RNG involved. It's better to eat the animations than the RNG. I don't think there's anyone fast enough to accommodate the RNG with tree strats... yet
N = level. Single = 40 x (n +1). Double = 100 x (n +1). Triple = 300 x (n + 1). Tetris = 1,200 x (n + 1). A triple is not 75% as good, it is 25% as good.
it's pretty rare that you see a new technique introduced in any game or hobby that is as dominant as rolling. it's crazy to see cheez burning through all these records so quickly and i can't wait to see where we go from here
I have literally ZERO clue of how these people even hypertap that fast?! And cheez is rolling the controller like a superhuman... I wanna be good at tetris one day, but it feels like such an impossible task.
The talent pool has widened considerably with classic tetris over the past few years so more and more outliers are emerging and rising to the top. Don't worry about comparing yourself to the best in the world - figure out what goals you'd like to accomplish and work from there. Many online tournaments have different tiers for players of all skill levels, and a maxout is still very attainable with only DAS.
There's also one extra thing with number 6 point - management of "tree". There were few setups (e.g. 5:05) where Cheez was forced to do singles because (at the same time) he needed to gap a holes that were created, especially cuz doing it with a L/J/lines could end with a destruction of "tree". If he'd save that and also had a manageable time - he was forced to do it more carefully, by doubles and triples.
Here is a potential video idea: Why is tetris 10 lines wide? Personally I think it should be 7, because if the goal of the game is to make tectrices, and there are 7 varieties of pieces which all fall at the same rate then the only width which allows the column dedicated for the I piece to fill at the same rate as the other 6 lanes together
but what if you turned the big regtangle on 12 row so when you turn it to the right you have a bit more time to turn it because now its only 1 block tall instead of 4
So I did a little math a couple weeks ago on why Cheez's ended up so much lower than PixelAndy's, and here it is again, though cleaned up a little: Cheez spent most of the game (about 240 pieces' worth) saving 8 frames per piece over PixelAndy due to basing his stack at 8 high rather than 4. [note: though Scout said 7, it's at 8 for most of the run] That's 1920 frames saved, or 32 seconds. He lost a good chunk of this by not going for tetrises as much as Andy, and thus not building as high off his base as Andy did, so in practice he was probably more like 1000 frames ahead with this metric. (their AVERAGE heights of placement were by no means 4 tiles apart) A portion of that 1000 is lost by having to take more singles, doubles, and triples, but it's not really that impactful. Line clears are 17-20 frames each, while at a height of 8 and above, the spawn delay* is generally 14-16, with 18 too rare to really be worth factoring in. Even if he only went for singles, worst case he'd only lose 600 frames, but on average he'd lose 450 frames. Given his healthy inclusion of doubles, triples, and tetrises, it was probably more like 200-300 frames lost. On speedrun dot com he's 720 frames ahead (unsure why there's no millis) so my estimates are probably not too far off. *Line clears override the spawn delay, and the spawn delay is only 14-16 because it increases based on how high the previous piece is placed. That's why in practice, a single only adds 3-6 over the spawn delay at that height.
The talent of pro Tetris players is incredible. The new methods they find are amazing. I honestly want to see the day when one player is just rubbing the controller on his forehead really fast whilst the other is using the ball of his foot on the d-pad whilst facing the wall or something.. AND THEY ARE JUST BREAKING RECORD AFTER RECORD 😆
I mean, TASes were using trees in 2004, so it was known to be optimal for a long time. Of course, having strategies in a TAS is radically different to actually doing them, but they've been around for ages.
I got 42 Tetris in a row to start the game in Tetris DX (no other line types) after grinding for hundreds of hours listening to podcasts I would be shocked if someone could get close to getting 34-36 in a row , it's super sacky
Cheez definitely has the potential to win the next ctwc, but there are so many rounds and good players nowadays it's impossible to presume a single likely winner. Anyone can be taken out by a misdrop at a bad time, like Joseph last year.
@@clarkkent1521 yeah, consistency really is king... Just look at greentea, he put up a good fight against joseph using das, and honestly if he wasnt busy with tetris effect, i could see him getting in the top 8 of ctwc last year, and the thing with rollers and tappers is that they take more risks, more risks means either more points or losing. Being consistent is still a better tournament strategy than being risky.
@@crudojoshuaricofort8795 I disagree? Using a faster playstyle doesn't equate to taking more risks, you just have to take risks so you don't lose pace... it's a different time in terms of scores. The thing people don't understand about tournament nowadays is that it's mostly RNG unless you're good at killscreen, which nobody is extremely consistent at. We're close to the full potential of the stacking/adjustment meta from the past couple of years, so you have to just hope your opponent doesn't get luckier than you so you can get a slightly higher maxout-1.1. You can't really get by at the top level with anything under consistent maxout. You are probably able to get a 1.1 with decent RNG and no misdrops, but anything higher than that is good pieces or killscreen. So there really isn't any hope of getting a really high score without something rare. However, the faster you are, the less luck-dependent you are, and the more potential you have on killscreen. You can stack higher, meaning you'll be able to wait longer for a longbar, saving burns in most cases. DAS can only get you so far nowadays because of the mobility. It's rare for a DAS player to get into masters event, and there were only a couple of hypertappers in the 2019 championship. Greentea is definitely up there though. If he played more he would probably be one of the few DAS players to make it into masters.
@@CheezTetris in that sense speed can lower risk, but what the person means is rolling can be more prone to misdrops than other strategies, and sure if you practice rolling enough you can probably get it to be just as consistent but right now it kinda looks like you misdrop a tiny bit more often than other absolute top players, which i see as the only thing holding you back from dominating practically every competition. however idk if i should be arguing because you obviously know much more than me and maybe i'm completely wrong. anyways good luck with everything!
why does no one use minecraft clicking techniques on the backside of the controler? it would be pretty easy to drag-click on that surface, wich would geberate up to 50 Hz
How responsive are the controller buttons though? FWIW I remember in GCSE maths trying to get to 100 on my calculator and my best was generally around 11 taps per second but not sure how responsive the calculator was.
I admit, when Cheez first made it to the quarter-finals in CTWC 2020, I thought it was a fluke. Now I'm convinced he's the best Classic Tetris player in the world, bar none.