New Rubik's Cube World Record Gets Set At An Absurd Three Seconds

June 14, 2023 5:14 AM

Puzzle maven and speedsolver Max Park beat a super-fast solve time that was set in 2018
posted by Etrigan (45 comments total)

Imma need a 240fps slow-motion replay on that one, Etrigan.
posted by clawsoon at 5:26 AM on June 14, 2023 [2 favorites]

I read this last night and saw the video and holy shit......

🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲🎲
posted by Fizz at 5:41 AM on June 14, 2023

Imma need a 240fps slow-motion replay on that one, Etrigan.

See the video at the bottom of the article where a YouTuber breaks down the solve. If you've ever played with a Rubik's cube, this is one of those moments where you realize there are people approaching something you thought you understood on a level so far beyond you that it may as well be magic.
posted by The Bellman at 6:09 AM on June 14, 2023 [7 favorites]

Interesting that it's almost all kids, and they seem to solve the cube in their head before actually starting. They know what they want to do before starting, it's just how fast they can execute their plan.
posted by COD at 6:43 AM on June 14, 2023 [4 favorites]

This may as well be magic as far as I'm concerned. The previsualization and the hand-eye coordination to execute the solution at that speed are amazing.
posted by EvaDestruction at 6:50 AM on June 14, 2023 [5 favorites]

Park is one of the two main characters in the documentary "The Speed Cubers," which is definitely worth a watch on Netflix.
posted by key lime guy at 6:57 AM on June 14, 2023 [7 favorites]

This is great. The previous record of 3.47 was somewhat controversial because there wasn’t great video evidence of it. This is a big jump and may last for years. I took up “speed” cubing a couple years ago but as an old person I’ve never gotten better than about 105 seconds which is too long for anyone to want to pay attention. You can use the beginner method to get down to under a minute with just proficiency, but below that requires a very impressive mastery of a dizzying array of algorithms. The sub-10 folk are just on another level.

Max is kinda old for a cuber at this point, good to see he’s still competing at the highest levels.
posted by skewed at 7:04 AM on June 14, 2023 [2 favorites]

“Yes, [there was luck involved]. However, there was a lot of skill and planning that went into this solve which shows how great of a solver Max Park is”

Kudos to Mr Park! But the word 'luck' bugs me. Does that imply some cubes are more scrambled than others? Is there a maximum possible entropy?
posted by JoeZydeco at 7:18 AM on June 14, 2023 [3 favorites]

See the video yt at the bottom of the article where a YouTuber breaks down the solve.

Interesting that having a slightly-luckier-than-normal starting position helped. I'm guessing(?) that they have some kind of rule in place to make sure that whatever random starting position they get, it isn't too easy?
posted by clawsoon at 7:18 AM on June 14, 2023 [1 favorite]

Few people know this, but I’m actually the world leader in cube slow-solving. I started on May 18, 1986, 17:35:33 and still haven’t completed it yet!
posted by slogger at 7:53 AM on June 14, 2023 [45 favorites]

I've known a couple speed cubers and yes they all know exactly the moves they'll do before they start. You couldn't get fast enough if you were thinking as you go. This also means that basically any serious speed cuber can solve blindfolded or behind their back, which is uncanny to watch even if you know how they do it.
posted by potrzebie at 7:54 AM on June 14, 2023 [1 favorite]

See the video yt at the bottom of the article where a YouTuber breaks down the solve.

I watched that and now I need an explainer video to explain the explainer video.
posted by kirkaracha at 7:58 AM on June 14, 2023 [7 favorites]

My new pandemic skill (instead of sourdough or cutting my own hair) was learning how to solve a Rubik’s cube. I know that in theory any cube can be solved in a small number of moves, but that’s beyond me - all I was able to learn was step by step, here’s how to get the top to look like this… here’s how to get the sides to match like that … until it all looks right. I know from watching this and slower videos that these speed solves are possible, but based on how I know how to solve one, this just looks like cheating. Or like EvaDestruction says, magic. I can’t even understand how to understand what I’m seeing as the same cube I have in my house. Just wow.
posted by Mchelly at 8:00 AM on June 14, 2023

At competitions like this are cubes randomized so they always take roughly the same number of moves to solve? Or can someone get lucky? Any cube can be solved in 20 moves. This page tells us most cubes require 18 moves to solve. Do competition cubes always take at least 18 moves? Or can someone luck out and get one that needs 15?

I may be thinking about this wrong: 20 moves is a theoretical bound for the optimal algorithm. In reality I imagine cube solvers use simpler algorithms that are easier to plan and execute.

The analysis video shows the scrambling was 21 moves. I couldn't count the moves in the solution he demonstrates.
posted by Nelson at 8:12 AM on June 14, 2023 [4 favorites]

…I’m actually the world leader in cube slow-solving…
posted by slogger

Eponysterical?
posted by snofoam at 8:17 AM on June 14, 2023 [8 favorites]

I’m actually the world leader in cube slow-solving. I started on May 18, 1986, 17:35:33 and still haven’t completed it yet!

Pfft - amateur! I've never even started one since it was invented in 1974, so I've got you beat by a mile.
posted by Greg_Ace at 8:34 AM on June 14, 2023 [4 favorites]

This also means that basically any serious speed cuber can solve blindfolded or behind their back …

Or while juggling!
posted by TedW at 8:37 AM on June 14, 2023 [1 favorite]

Park is one of the two main characters in the documentary "The Speed Cubers,"
The Speed Cubers, previously on Fanfare.
posted by polytope subirb enby-of-piano-dice at 9:00 AM on June 14, 2023 [1 favorite]

TedW: Or while juggling!

I'm pretty sure that's what Shakespeare was talking about when he said, "There are more things in heaven and Earth, Horatio, than are dreamt of in your philosophy."
posted by clawsoon at 9:15 AM on June 14, 2023 [3 favorites]

I enjoyed The Speed Cubers, a documentary film that focused on Max Park and another person.
posted by lookoutbelow at 9:45 AM on June 14, 2023

I'm not a video game person but I once got hooked on one of those online Tower Defense games. I thought it was so fun to try and think up a little impenetrable wall. I showed friends and they got hooked too. I then randomly saw a YouTube video of someone posting insanely high scores. My jaw dropped. They were playing the game at such a high level and it looked completely different than what I was doing. They were doing something called "juggling" where they were taking apart defenses on one side so attackers would rush that way and the simultaneously building that back up and breaking holes on the other side before they got there. I would never get as far as they did using the approach that I think most would start with. I never played the game again. This feels like that to me -- both awe-inspiring at their skill but makes the way I thought about Rubik's Cubes feel so quaint and novice that it kinda removes the desire to pick one up and randomly try and solve it.
posted by This_Will_Be_Good at 10:12 AM on June 14, 2023 [6 favorites]

I took a look at the video The Bellman linked to and two things strike me:

1. man this guy has a weird accent that I cannot place
2. this cube is radically different from the officially-licensed Rubik's Cube sitting on my shelf, I physically cannot do things like rotate the bottom layer with my pinky because the cube tends to jam up. I'm noticing that the cubelets of his cube, and the one Park is using, have very rounded inner corners that I assume helps with this problem.

A little more investigation suggests that these cubes are, in fact, held together with magnets, which would sure explain why they move with so little friction.

fuck it, I'm ordering one to fool around with, it'll be a nice break from spending this week trying to cram all the shortcuts for a new animation program into my brain
posted by egypturnash at 10:15 AM on June 14, 2023 [2 favorites]

A little more investigation suggests that these cubes are, in fact, held together with magnets . .

F'ing speed cubes, how do they work?
posted by The Bellman at 10:20 AM on June 14, 2023 [8 favorites]

I don't want to derail the thread about Max's accomplishment into a thread about the documentary that focussed on him, but The Speed Cubers isn't easy for me to recommend.

I enjoyed it as a window into competitive Rubik's cube process, culture, and events, but I was caught off-guard by the extremely neurotypical outside perspective on Max (who IIRC was not interviewed and, whether by choice or not, never shares his perspective with the camera) as an autistic person. Reviews are thin on the ground, so for a more in-depth discussion of the documentary's flaws I would recommend reading this German review, through your favourite machine translation tool if need be.
posted by polytope subirb enby-of-piano-dice at 10:20 AM on June 14, 2023 [3 favorites]

There was only one record-setter but at least five or ten kids and adults were screaming and celebrating. It warmed my heart and made me smile to see the joy and excitement of everyone in the video.
posted by Taro at 10:28 AM on June 14, 2023 [4 favorites]

Max's record was set at was Pride in Long Beach 2023, a World Cube Association event, during the first round of the 3x3x3 Cube event.

At competitions like this are cubes randomized so they always take roughly the same number of moves to solve? Or can someone get lucky? Any cube can be solved in 20 moves. This page tells us most cubes require 18 moves to solve. Do competition cubes always take at least 18 moves? Or can someone luck out and get one that needs 15?

Per the WCA guidelines:

1h+) RECOMMENDATION Competitors in the same group should use the same scramble sequences. Different groups should use different scramble sequences.

If Pride in Long Beach went with 1h+, everyone in the same round has their puzzles scrambled the same way.

The WCA regulations get into how scrambling takes place. Scramble sequences are generated with an official scramble program, TNoodle-WCA-1.1.2.

In any event, this particular format averages 5 different solves, so the effect of one particularly favorable scramble would be minimized.
posted by zamboni at 11:24 AM on June 14, 2023 [7 favorites]

An ex of mine was into Rubik's cubes as an adolescent. At age 11, he could solve one in 11 seconds. He had a collection of off-brand cubes, and it got to the point where the only thing that was slowing him down was the mechanism itself. He used to take his cubes apart and lubricate them. Some of his friends used butter for this, but he used the vegetable cooking spray Pam. He bought his own can and kept it on his closet shelf.

One day when his mother went into his room to put away his laundry, she found the Pam. When he came home from school he saw that his can of Pam was sitting on the dining room table, and his mother said, "Come in here, young man, I want to talk to you."

He went to her, and she said to him, "Are you doing drugs?"

He said, "No Mom, I'm not doing drugs! If you found drugs in my room, I didn't put them there!"

His mother believed him, but with that explanation for the Pam ruled out, she looked at her 11-year-old son in a puzzled way, wondering what he could possibly be doing with vegetable cooking spray, until suddenly her look of puzzlement morphed into a look of horror, and she said, "OH MY GOD!!! Go to your room! I don't want to talk about it! Go to your room!"

He went to his room, got his cubes, came back to his mother, and showed her how he'd been lubricating them with the Pam. His mother said, "Ohhhhh," looked tremendously relieved, and gave him back his Pam.

I used to tease him about how I'd never measure up to his first girlfriend... Pam.
posted by orange swan at 11:29 AM on June 14, 2023 [31 favorites]

I have an acquaintance who got one of these competition cubes to practice with and while as far as I know they never achieved any high level of competitive speed, I learned that a magnetic 3x3 Rubik's Cube is a *fantastic* fidget toy. I've never successfully solved one but they feel entirely different from the standard 4x4 ones everybody knows and are absolutely hypnotic to interact with.

Also, I want to say how much I appreciate Kotaku just starting out the article with a gif of Park doing the thing they're talking about. A+ formatting. Love to see it.
posted by Phobos the Space Potato at 11:53 AM on June 14, 2023 [4 favorites]

This is an extreme version of what my friends and I would do: we’d look at the cube then make any number of moves behind our backs, look again, repeat. Winner was the fewest looks. I don’t remember my look count exactly, but I think it was around seven. This guy gets it in one. Amazing.
posted by sjswitzer at 3:03 PM on June 14, 2023

I get a 404 on the German review of The Speed Cubers.
posted by holborne at 3:12 PM on June 14, 2023 [1 favorite]

orange swan, in AJ Jacobs’ most recent book on puzzles, he mentioned a special lubricant that you can order off Amazon for puzzle cubes/boxes. Might have led to a different awkward conversation but at least there wouldn’t be any suspicion of cooking spray abuse.
posted by dr_dank at 3:55 PM on June 14, 2023

This is an extreme version of what my friends and I would do: we’d look at the cube then make any number of moves behind our backs, look again, repeat.

This is basically equivalent to a separate WCA event, blindfolded solving.
posted by zamboni at 5:18 PM on June 14, 2023

According to https://ruwix.com/blog/max-park-rubik-single-record-313/, which has a widget thinger you can use to step through the solve, Park's solution was 33 rotations.

And I think people treated it upthread, but 20 is God's number---whatever deities have time to play with twisty puzzles can unscramble any scramble in 20 moves or fewer. Some individual scrambles are definitely lucky, and my understanding is that records like this are where "chance favors the prepared mind"----you need to be able to capitalize on that lucky scrambles when you happen to get one in a sanctioned competition. And you need a little more luck to make sure this doesn't happen https://youtu.be/qt8zMZ8V148?t=17.
posted by adekllny at 6:20 PM on June 14, 2023 [1 favorite]

My now about to graduate from high school son was into cubes for a while 5 years ago or so. I asked him today, "did you see that dude who solved the cube in 3 seconds?" He is of course on TikTok, and he was like, yes, I've seen it, yes that was crazy.
posted by Windopaene at 8:29 PM on June 14, 2023

Man, when I was a kid I thought that being able to solve it in under three minutes made me the shiznit. (I don't even remember the method that I used then.) This is some straight up Reed Richards stuff.
posted by Halloween Jack at 9:38 PM on June 14, 2023

When I was in High School (Class of '87) I made a bunch of extra cash solving other people's Rubik's Cubes. I could do it in about 60 seconds doing what is now called the "beginner method".

I picked up a couple of speed cubes a few years ago and tried again, now I can do 30-60 seconds (2-look OLL and PLL).

But yeah, the under-10-seconds people are on a whole different level.
posted by mmoncur at 9:56 PM on June 14, 2023 [2 favorites]

A little more investigation suggests that these cubes are, in fact, held together with magnets, which would sure explain why they move with so little friction.

It's even more interesting than that -- the magnets aren't to hold them together, that's done with screws and springs that can be adjusted to the desired tension. What the magnets do is twofold -- first, when you turn a face, as it passes the 45° mark, the next magnets along pull the corners forward, so you don't need to put as much initial force into the turn. That's part of how the cubers can do a turn with just the flick of a pinky, because the magnets do a lot of the turning work.

The other thing they're for is, they snap the cube into alignment, so the cuber never has to manually adjust it to keep everything squared up, which can be a problem with a heavily-lubed nearly frictionless cube.
posted by rifflesby at 11:07 PM on June 14, 2023 [5 favorites]

I get a 404 on the German review of The Speed Cubers.

Here’s the right link for the German article.
posted by ellieBOA at 1:52 AM on June 15, 2023 [1 favorite]

Magnetic cubes are not held together with magnets.
The magnets are used to aid in alignment, for accuracy. You can, with practice, flip an edge and it just goes to where you want it.
Apparently, the added weight helps too.

I have a GAN model that's kind of on the low end of speed cubes as far as price goes. It is super smooth, so much so that I have a hard time doing normal cubes, and I've never lubricated it.

I have a couple of under 40-second times, and don't expect to get much faster because 1- I'm old; 2- I'm not interested in memorizing 135 positions; 3- I can't see more than 3 moves ahead. I think I could get a little better if I worked on some finger techniques. I went from a 2-minute average to a 50 second average using Roux
posted by MtDewd at 3:31 AM on June 15, 2023 [2 favorites]

OK here's the thing I don't get. Why isn't the looking at the scrambled cube and working out a solution part timed? Isn't that a critical part of solving a Rubik's cube?
These guys would still be solving it thousands of times faster than me, it just seems like solving a Rubik's cube should be just as much mental as dexterity.
posted by MrBobaFett at 2:30 PM on June 15, 2023

It is timed in a sense -- you get a certain number of seconds to look, and if you go over that your final score is penalized. I don't know why the looking time isn't just part of the final score, though.
posted by rifflesby at 4:57 PM on June 15, 2023 [1 favorite]

My understanding is that even among elite competitors, they’re not pre-solving the cube during the inspection period, they’re just planning a few moves in. In standard competitions, everyone gets a 15 second inspection time, and that’s how it’s been since the first world championships in the early 80’s. Back then, record times were around 23 seconds, and I think it was felt that the inspection time helped reduce the effect of getting a lucky scramble with an easy and obvious opening sequence. I imagine if they were starting over, they’d probably have a reduced inspection period given today’s records, but to maintain continuity, they wanna keep the same rules.

In blindfolded competitions, there is no inspection period, the clock starts when the cube is revealed, and the player just puts on their blindfold when they’re ready, so a player has to basically solve before they pick it up.
posted by skewed at 5:13 PM on June 15, 2023 [3 favorites]

I can't see more than 3 moves ahead

To clarify something that’s been said a couple of times here, a “move” isn’t a single face rotation, but rather a combination of rotations, seven or so, that’s been entirely committed to muscle memory.
posted by sjswitzer at 9:05 AM on June 16, 2023

I was using 'move' to mean a combination, like a 'sexy move' or a 'sledgehammer'. I would like to be able to see further. I easily lose track of what was supposed to be my next object.
There's one I do in finishing the top corners in a 14-move sequence that I could write down only with difficulty, and I usually mess up if I try, but my hands can do it in 2 seconds as long as my brains not paying attention.

... I learned that a magnetic 3x3 Rubik's Cube is a *fantastic* fidget toy.
I realized this only in the last year, but now I will describe what I'm doing in those terms.
posted by MtDewd at 2:48 PM on June 16, 2023 [2 favorites]

He once had motor skill challenges. Now he's the world's fastest Rubik's cube solver
posted by Etrigan at 8:20 AM on June 21, 2023