The Birth of the Impossible 34x34 | Official Documentary

[@PrestonsPuzzles]

Sorry again for the terrible timing on the completion of my cube. If I had known of this cube at the time of completing mine, I defiantly would have held off so that you could have held the record before me. This was a killer documentary and it has been a pleasure getting to know you over the last few months. Respect!

[@MichaelAutism]

At least his cube will be remembered. Now this makes me wish to build a 64x64, but my parents dont have enough money to by 6 fdm printers

[@table5584]

I am making a 100x100 and currently have a basement filled with 15 fdm printers. I use a click and snap design similar to Greg’s 33x33x33 where the puzzle in the inner layers snaps to lock. This puzzle actually has to have magnets within the actual inner edge pieces, however this meant I had to scale the puzzle down either extremely small or extremely big to double combat the problem, which I chose extremely big, which I’m thinking of buying a warehouse to place it in, and it leaves room for movement between pieces. The 100x100 will come in 2050 and will be 10 ft long on each side, and require a special “belt” that will wrap around a layer to be able to block any corner cutting.

[@willoseb]

what happens when the lube on big cubes dries

[@fireblast_1000]

​@@table5584seriously?

[@TonyFisherPuzzles]

This should be shown on Netflix or one of those other streaming stations. Best puzzle related video of all time.

[@SpotErrOne]

Approved

[@gancuber11-wb5kc]

Sending application to Netflix as soon as possible :-) :-) :-)

[@nickroethel3326]

Wow. That is HIGH PRAISE coming from you. Ive followed you for like.. a decade. I don't have time to watch it right now. But I'm DEFINITELY going to watch it when i find the time. I mean I was going to anyway. But after seeing this one recommendation Its next on my list.

[@DipietroGuido]

To the best cubing video and documentary to have ever been made. Sincerely hats off.

[@Apol_catOfficial]

YOOOOOOOOO

[@Gianni5]

holy sh*t

[@pessimistic5579]

a whopping 5000 ars (approx 5 usd)

[@Apol_catOfficial]

@@DipietroGuido HOW THE DEUCE DID YOU DONATED 5 US DOLLARS!?

[@Apol_catOfficial]

@@pessimistic5579 does that mead 1 ars, 10 ars, 100 ars is 0.001, 0.01, 0.1, 1 us dollars?

[@jacobcubing606]

This has gotta be hands down the best cubing video I've ever seen in my 8 years of cubing. That ending gave me goosebumps. Also thanks for showing my 23x23 at 23:46

[@yellowmarkers]

Jacob Jacob, Jacob Jacog, Jacob Cubing, Jacob Cubinb, Jacog Jacob, Jacog Jacog, Jacog Cubing, Jacog Cubinb, Cubing Jacob, Cubing Jacog, Cubing Cubing, Cubing Cubinb, Cubinb Jacob, Cubinb Jacog, Cubinb Cubing, Cubinb Cubinb

[@SpotErrOne]

@@Autistic-Joseph-Shrayhan Lie. Such a direct lie.

[@yellowmarkers]

For some reason my comment only shows up in the Newest First tab

[@jacobcubing606]

@@Autistic-Joseph-Shrayhan So what? Why is that relevant?

[@jacobcubing606]

@@SpotErrOneI just mass reported Fynn-has-Autism for spam. Hopefully their comments stop

[@Longboost]

29:52 Decided to try and work this out because I was bored, lol. - The circle's diameter is equal to the width of the cube. - Draw a diagonal line between the opposite corners of the blue square. This is also the circle's diameter. Therefore, this diagonal line has a length of 1. - This diagonal line has split the square into two right-angled triangles, with the diagonal line also being the right-angled triangles' hypotenuse. - We can use Pythagoras to work out the side length of the square (let a equal the side length): 1² = a² + a² 1 = 2a² 1 / 2 = a² 1 / √2 = a - Now we can find 2x by simply subtracting 1/√2 from 1, and then we solve for x: 2x = 1 - (1 / √2) x = (1 - (1 / √2)) / 2 x ≈ 0.1464466094 Not as hard as I feared, nice! Edit: I realised that I forgot that 1 - (1 / √2) equals 2x, not x. My method has been updated to the (hopefully) correct solution, now.

[@DipietroGuido]

This is my favourite documentary ever and the 49x49 existing simply makes this story PERFECT. I love the attention to the smallest of details and the irony all over this. This is just all so insane to be true and I love it.

[@ProfessorBorax]

Ironic because of the completely different aproach?

[@ProfessorBorax]

Ah I hadn't watched the last couple minutes of the video yet! True, it gave the video a more spiritual twist :)

[@NobleCubes]

Soo my take away from all this is... a 100 x 100 x 100 is actually going to happen??

[@ConnorsCubing]

Shhhhhh 🤫

[@RoweJennings-p4n]

me **stops working** hey how did you find out

[@TYDRAX]

someone made a 49x49

[@SpotErrOne]

yes, but after 11-70 years

[@NarsehVR]

My mind was blown when I saw how recent this was posted! 40 minutes in and I am enthralled!

[@Cubersome]

YO NARESH! didn't expect you to be here are you a cuber?

[@NarsehVR]

@@Cubersome Not really, KZbin just blessed me with the original and had to see this 👀

[@dbrokensynth5920]

bruh the slicer support bug isn´t really a bug but no one seems to care, the added material apears beacause theres a given airgap to avoid the part fusing to the support, oviously in real life the gap doesn`t exist as the material drips until it touches the support (filling the air gap but not fusing to the support) its just a matter of doing what you did to your design but in the slicing process to compansate for the air gap caused error but yet no slicer seems to care enought

[@S0larisPrime]

And since you built this, Preston's Puzzle made a 49x49...

[@TheLadderman]

The 49x49 has been finished for months actually, he just took a long time to make the video.

[@SpotErrOne]

49x49 project was started before 34x34 (or before 33x33)

[@rubicade]

But this one turns so much better

[@fallen3424]

this one turns actually well though

[@dirp_dcb9531]

Oh my days I hope that people don’t just discredit the effort put into the 34x34x34 because of the 49x49x49

[@YouTuber-Playlists]

This guy has so much patience and is just a straight genius. And to see his excitement at the end! Keep up the good work man!!

[@tarinacj03]

The 39x39 may not have ever held the world record but i really hope you can take immense pride in the fact that this is one of the highest quality large NxNs produced. Hands down this thing is a masterwork of engineering and the work youve done here is really going to pay dividends on making larger and larger puzzles. Truly spectacular.

[@dottedboxguy]

just saw a video showing a 49x49, truly crazy how far we've come

[@InexorableTash]

Awww, now I feel extra special to have a printed yummyyummypbj puzzle! Great to see the behind-the-scenes of the epic stop motion.

[@JunJun238]

even if there were bigger cubes since then, yours probably turns the best out of all of them

[@jhudson_tiedye]

if? there is factually a 49x49

[@nintendomal9161]

And I'm happy to be here for this moment 😊

[@emmaginn]

Please let me know when you plan to release your 100x100 so I can upload my 101x101 right before.

[@SpotErrOne]

Just start to doing 101x101 right now.

[@catchmeramesh]

Bro this is INSANE even though I’ve already seen it, it’s just so crazy! Good job with this!

[@damianwilding]

Im so glad the 34x34 wasn't your end goal 😮

[@creativebuilders1117]

42:20 (note: i have not tried many of these filaments so I am only going off other people's experiences) ASA seems to be a very good alternative to ABS, as it warps less and is more uv-resistant than ABS. The only caveat being the price so it might be more practical for smaller projects. PCTG also seems very promising with it having better mechanical properties that PETG and it prints almost as good, if not better than PLA (once again more expensive so bettef for small projects). Zack Freedman has made some videos covering almost every filament on the market and he covers a lot of PLA alloys that cover its weaknesses so those might be something to look into.

[@speedcubingdotorg]

This may be the best cubing video ever made, it is really impressive that you can both make such a massive puzzle and also make excellent content around it, hopefully the algorithm promotes this to the extent that non-cubers I know start asking if I saw the news about the 34x34!

[@SpotErrOne]

1:39:39 bro really said "world record"

[@gusbert]

This beautiful video demands a wider audience! Your sheer joy at completing this Herculean task make me so happy. I know that I would be considered a twisty puzzle nerd (I have well over 1000 in my collection) but this is so entertaining I'm sure more people would love to watch it. I cannot believe the creators of the slicing software did not know about the 0.25mm bug! OMG! When I saw you use 3D filament as guide pins I realised how stupid I was. In the past I used cocktail sticks or bamboo skewers for this, but of course these things are not dimensionally accurate, so drilling or sanding was required. I wish i had known this years ago.

[@malalalalalala]

What an amazing accomplishment! The patience and dedication that went into this project, not only with the cube but with this documentary is unfathomable.

[@robertgcode965]

Support issue is not a bug. You need a separation distance so the supports don't fuse to the part. Maybe try reducing the separation gap and change the dense support to be easily breakable. Or get an MMU to use PETG supporting layer with 0mm separation gap.

[@caledoncuber7502]

Doesn't a nonzero separation distance cause the next layer to "sag down"? Sure, it prevents the part from fusing to the support, but at the cost of dimensional accuracy as Bahner describes here. Would it not make more sense to raise the nozzle slightly when printing regions above support material, rather than to leave a gap beneath the very sections which need support? I think this approach would have a chance at producing convenient breakaway support, without sacrificing dimensional accuracy.

[@SpotErrOne]

Matt Bahner should cook spaghetti for you

[@LuProch]

You are a mad man 😲 And I love it, this is 100% up my alley - twisty puzzles, 3D printers and new technology in general, your filmmaking, crazy style and humor, everything. Absolutely worth watching. You made something incredible, even if you don't have the record. Looking forward to more content from you. Now to watch the stop motion assembly again 😀

[@janaki3829]

These "biggest nxn" videos remind me of that post where someone commented something like "but that's going to take 5 years" and someone replied with "the time will pass anyways"

[@DirkCuber]

The thumbnail changed to a scrambled cube??? That means a solve is coming.

[@ProfessorBorax]

This is inspiring dude, letsgooo!!!

[@Nehimsupacac]

What a beautiful video Matt, love and talent drips from second of it. This is the kind of perseverance that will get the 100x100 to life.

[@very_not_emo]

this video is gonna be so peak i can feel it

[@androidlogin3065]

My congratulations, finally proved, nothing is really imposible, only is needed the knonledge and effort. I belive in you, you can build them and i knew that. I am really happy to see you being so happy.

[@willoseb]

what will you do when the lube dries on this cube, it seems unlikely that you will try to break it in

[@laiscube]

Can't say how much I loved watching this video. I really love your (and Jacob's) videos and puzzles, did talk to Jacob about designing puzzles too, in especially over cheating the law of bigcubes and making the outer layer go below the 1/6.83 number. As much as Preston's 49x49 was impressive, I still prefer how the wide outer layer of the 34x34 looks over the curvy cuts of the 49x49 although a hybrid between the curvy cuts and the wide outer layers would be intresting too. It's not that hard calculate a height to which the cut line comes out of the cube slanted, and upon reaching the calculated height it changes to pure vertical, such that the outer layer will be perfectly parallel squares and rectangles, but all the inner layers have curvy cuts. I can't name a number for this as it would be a function of the outer layer and not a constant. But specific numbers can be named for 1/7 or 1/8 sized outer layers because we have plugged numbers into the function. Constants would be like a outer layer cant be smaller than 1/6.83 of the entire cube, or how the angle of a perfectly parallel corner stalk is 'inverse tangent square root -2' or 54.74 degrees. By that, I'm that math nerd who solved for x long before even watching this video. It's the 1/6.83, just that I prefer to present it reciprocal style to show how it relates to 6x6 and 7x7, and how 7x7 is actually the smallest impossible cube to build without tricks like flexible parts, curvy cuts or big outer layers. I'm another old cuber who initially learnt puzzle design from your old turotial. and my first few puzzles were SolidWorks/SLS (like the Maglev 5x5 and my first few 4x4s), but I went to onshape/FDM sometime in 2022 and made two maglev 4x4s, a 2x2x14 and some others. Also made an onshape-based 4x4 design turotial because even cubes are also my passion. Went through my method of making the AB allignment mech in onshape, and invented my own allignment mech separately which was used in the Maglev 4x4s. (acutally would rather deal with allignment mech than center caps). My 3d printer is still the a the Ender 2 pro (Florian Shengshou 4x4 of the 3d printers, great budget printer for it's time with mods, but better things have since came out), and financially/free time-wise I'm not in a position to upgrade, eyeing Bambu labs A1 mini (which would be the Aosu WRM in comparison) Speaking of support matertial, Episode 5 (still unlisted but made some time ago) of my 4x4 design series mentions my prefered oriention of printing most of the 4x4 pieces with no supports. You are the reason why if I'm designing object A to have a hole to fit object B inside, I intentionally make the hole 0.2mm bigger than the dimensions of object B. Literally learnt the 0.2 number from watching your videos. Object B is most commonly a magnet and Object A a cube piece or part of a core-corner magnetization kit. I have had a few kits ruined by support residue that caused a magnet to be glued in slanted. The wing switching operation is such an anxiety trigger. Made a lot of edits because I'm adding to this comment as I watch the video

[@nowymail]

More emotions and plot twists in that 1 hour 47 minutes than in cinema movies nowadays!

[@damiens4601]

Pissing off Greg was the best reason xD

[@MrEFynn]

This is an AMAZING documentary! Thanks for sharing how you built the 34x34

[@Blue-Cubing]

I agree!

[@SpotErrOne]

I agree!

[@MrEFynn]

@@SpotErrOne btw I think Matt Bahner blocked the annoying person because I don't see any of his comments anymore

[@SpotErrOne]

@@MrEFynn Which one?

[@MrEFynn]

@@SpotErrOne When he replied to Jacob Cubing

[@Bigcubefan]

Just a few days ago I was your first video on this cube, where you promise more will follow. Then I saw you hadn't uploaded in a while and thought the channel has died off or something and then this dropped. :D

[@rccubing907]

38:20 Im not an expert for designing and 3d printing but this came to my mind and you can correct me... If the prints requires support and gives .20 to .25 mm of error, could you or is it possible to reduce the size of the main parts thats requires support by .20 to .25 mm ?

[@TheZixOriginal]

Parity cases would be wild on this🔥

[@pilotboy217]

Y'all out here building 34x34's and I just want a mini pillowed 7x7 or 9x9.

[@iPhone18ToasterfrLOL]

Matt yeeting his 34x34 out the window after PrestonsPuzzles made the 49x49

[@Trapped_in_the_Dunya]

@1:38:19 this is very close to the reaction I had when I legitimately solved my first cube. What a feeling, and I'm so happy to see someone feeling that sensation!

[@ashokacelebucki4575]

i would definitely watch the 1-hour video explaining the design process

[@SmiteoPuzzles]

Epic project and documentary!

[@Unknown-am76]

How long would you think this would take to solve with help and without

[@Icetastesgood]

5 months is my guess

[@creativebuilders1117]

With help I'm pretty sure it would pop at some point even with such good turning 💀

[@Israel-Niles]

22:24 something funny I realized is 7 tiers of 7 is exactly 100! What a coincidence!

[@Israel-Niles]

You need to get revenge on Preston by making a 50x50

[@GaborRudas]

I need to ask, when you mention weeks or months, do you mean full working days? What is/ was your profession? Do you do this full-time? Thank you for making this. I've learned a lot from this and from your other videos as well.

[@nowymail]

Next already in progress? 101x101?

[@youtubehandlesux]

Just installed 0.3mm nozzles onto my dual extruder printer. Time to print some cubes! Hope all the extra money I spent perfecting its accuracy means I can make smaller cubes that turn well. I'm also going to use PETG which has lower friction, and dissolvable supports would make overhangs silk smooth.

[@matt.bahner]

Are you on the puzzle maker discord? If not email me and I can send you a link [e-mail]

[@JenniferStreetArt]

A 100x100 cube would be awesome, a Google x Google Cube would be awesomer. Great Video ❤

[@AnimilesYT]

I just saw a video about a 49x49. But it is in the category of cubes which are hard to turn, so you still hold the record of biggest usable cube

[@trendygaming795]

I can't imagine parity on this thing

[@redsgxd]

You may not have the world record for the largest nxn but at least I think it's cool you have the world record for the largest even nxn which are harder to make. The previous one before that was the 22x22 which is just only one layer bigger than the biggest massed produced nxn, the 21x21

[@ZeroKage69]

I got an 8x8 years ago and my first time solving it it broke apart and all the pieces fell to the floor around me and I didn't know what to do lol. I put them all into a box and they sat there for like months while I spent my time solving my 6x6 and 7x7 that I got at the same time. But eventually I tackled the arduous task of reassembling the 8x8. I divided all the pieces by shape and found a video online for how to reassemble an 8x8 and then took a long time to finally get it all back together. And I still have it today. But I had to build it before I could even solve it for the first time because of that.

[@CandyCuber1]

MATT! Huge question: I know you’ve been designing puzzles for a while, so you have more experience, but I want to learn from someone who knows what they’re doing and from not trial and error. What do you use for tolerances (like in mm)? And do you take that amount away from every internal surface or only the parts of the mechanism that pull out?

[@jeremiaszchec]

so now you can make a giant ball 34x34?

[@annachojnowska8750]

Z3 cubing should do this on his challenge: 1 turn per mile

[@SpotErrOne]

Hey! Could you please post tutorials on how to make edge and corner turning puzzles? (like helicopter, dino, skewb)

[@yuriymatsevych5249]

THIS VIDEO NEEDS TO BE TO BE SEEN BY ALL CUBERS, I hope it gets more views

[@PuzzleWorks]

“All men can see these tactics whereby I conquer, but what none can see is the strategy out of which victory is evolved.” - Sun Tzu, The Art of War

[@TheMadScientistOfLuton]

It's like watching a brain being built out of plastic

[@JakePark-s2d]

The thumbnail gave me a heartattack

[@EggBoyMAGOLOR]

Cubers with NxN cubes larger than 3: Once you pair the edges and match the centers, you can solve it like a 3x3! 34x34: 🗣️🗣️🗣️

[@AnimilesYT]

I'm no expert, but is it necessary to have the pieces go all the way to the core of the cube? Can't they hold on to each other in some other way without becoming unstable?

[@akeustlom]

It is possible to make big cubes without the pieces extending so far down; in fact, Matt used a much shallower mechanism when he made his 11x11 ball. In theory, the same approach could be applied to a cube of any size, but having the pieces extend towards the core helps in general because it becomes less likely for a piece to tilt out of place.

[@jayalthr9081]

Matt, your are one of the GOAT of the twisty puzzle making game!! Right next to Tony and Oskar and other puzzle makers. This video is amazing, entertaining, educative and inspiring. The level of detail you use to explain the process, the engineering, the methodology is simply great and I will probably watch it another time. The dedication, time sacrifice, passion you have forces respect and your nice personality makes this long video on a somewhat niche topic one of the best video I ever watched!!! I loved it!

[@melissazoethout3097]

Please do a time lapse of you solving it

[@getjaketospace]

I used PLA when I made my world's largest pyraminx and I don't think I had a single failed print (if there was it was because I did something dumb externally). It was insane compared to when I only used ABS way back in the day

[@CandyCuber1]

Bro you gotta make some designing tutorials bc I’ve been using onshape for a while and my brother dropped my 9x9 so I designed a new corner for the one he snapped, but it took me FOREVER to design

[@damianwilding]

Can i have that extra wing you accidentally made? For the bit? I'll pay 🤣

[@stephencole9289]

Havent watched it all yet, but does it say why it was a even number rather than an odd number design ? (even is usually based on an internal odd number one greater)

[@stephencole9289]

This is dicussed at 17:00-25:00

[@minecraftr298]

Are you gonna go for the record again? I'm halfway through video atm so you might answer that later lol idk

[@damiens4601]

Someone said the 0.25mm thing is not a bug and it's the support distance and you have to adjust the thingy

[@renzostc7938]

Amazing

[@christophergeorgia9670]

well done

[@leviphillips4941]

Sad that the record is already broken

[@LostGuyStudios]

Wait it was? When

[@jhudson_tiedye]

@@LostGuyStudios49x49 and technically it was finished before the 34x34 but the video was only posted a few days ago

[@diamonddogie]

​@@jhudson_tiedyedid you mean 34 was finished before 49 but 49's video was posted 5 day before? Otherwise your comment make's no sense.

[@jhudson_tiedye]

@@diamonddogie You didnt watch the end of THIS video. I knew before the video but the 49 was finished a month before the 34 but not posted. 34 was never a wr nor did he ever claim it was. Do your own investigation before saying I make no sense

[@jhudson_tiedye]

@@diamonddogie 1:43:48

[@Xz12_358]

Why's this guy jus got GameCube controller laying on the floor? 47:48

[@Peaceful_Zen_Life]

Imagine dealing with parody on this 34 x 34!

[@Jupiterninja95]

*parity

[@Peaceful_Zen_Life]

@@Jupiterninja95 lol thanks. I wasn’t quite sure if I spelt it wrong or not.

[@Jupiterninja95]

@@Peaceful_Zen_Life all good! Weird Al makes parodies, parity refers to whether something is even or odd

[@Peaceful_Zen_Life]

@@Jupiterninja95 the funny thing is is now I do a total facepalm about it! From me, my reference for parody is movies. I can remember like 2% of my brain was like hey idiot remember parody movies that’s not how you spell it.

[@ThePotatoMemer]

now think about that on a 49x49.

[@Jacktheawsome]

Can you add magnets

[@SpotErrOne]

19:36 I actually used this formula before

[@YuutoGaming]

This is age so well

[@pedropiata648]

1:38:31 imagine his neibors 💀💀

[@Supermario64fangd]

Can't wait for the 69×69×69. You are the johnny sins of cubing

[@BlåhajTheGodShark]

I know you probably won't see this, but I was thinking of making a slideshow with wacky rubiks cube concept art. I will edit my comment if/when I make some art by putting a link to the slideshow. If I make it, I recommend you take some ideas but you don't have to. Shower thoughts be wilding huh Sincerely, Andrew.

[@jimlavogiez-sf6wb]

So... the 34x34 was just a warm up for the 100x100 ?!

[@kidredglow2060]

welp.. NEW WORLD RECORD: 49x49x49

[@CatnipMasterRace]

TBH wouldnt be surprised if you could get more than $150,000 for this...

[@sk1ller_604]

38:00 nah, i think no one can fix this, sandpaper it is. for chicken nuggies tho🤔

[@DirkCuber]

13:46

[@ChickenScratch]

Dude I love your energy so much. Thank you for making this!

[@ThePotatoMemer]

Me when the 49x49...

[@billybobjoe-gv7jx]

Bruh. Who said you will last another decade. You must build yourr 100x100 tomorrow before its too late

[@jow_angelow]

1:14:38 excuse me sir, what are you doing with 1000 bottles of lube? 🤔

[@jiroscop]

The real WR as the other barely work

[@SpotErrOne]

then you have (even)x(even) cube

[@HarryLarsson-b2n]

unfortunate timing

[@SpotErrOne]

1:45:05 ok 1:39:39 wt

[@Regian]

Blink and you miss five hours of editing

[@LâmHuynh-h9b]

99×99 is my dream