• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

Proposal to engineer and construct MBLD scrambler machine

abunickabhi

Super Moderator
Staff member
Joined
Jan 9, 2014
Messages
8,843
Location
Yo
WCA
2013GHOD01
YouTube
Visit Channel
Aim - Make a scrambling robot that can kept on a bench, take 70+ cubes kept on it, scramble them, arrange them back in proper configuration, cover setup with cloth, alert the user and automatically start video recording of the entire setup for the duration of the attempt from an independent video recording device.


Dreamy infeasible details - Some features that I can think of from the top of head:

1) It can check for logos and Bluetooth cubes and remove them. It brings each cube in WCA orientation before it executes the scramble.

2) The robot will also stack the 60+ cubes in your desired arrangement. Also once the robot is done scrambling and arranging the cube, it pings a video system that starts taking video of the entire attempt. The user need not worry about scrambling as well as video recording.

3) The prototype should be cost effective. In the electronics, it can use Raspberry Pi or FPGA or a 32-bit processor, that takes to scrambling part and sensory parts of the system.

4)Design a product that has one basic MBLD scrambler + 70 identical 3x3 cubes + UI that user can use easily to automate MBLD scrambling process.

5) For competitions of scale of Worlds, where there are 100+ people who can do 60+ points, we have a mega robot installed at the venue. It can scramble all cubes at once in 1 minute and there will be no delay in the MBLD event. Robot should be able to detect competitor making accidental turns during memorisation and also find out the number of DNFs post attempt.

6) Benchmark the product to have smaller footprint on the table and capable of scrambling more 3x3s(~200 cubes) at once.

7) User can practice custom attempts like only 7 alg scrambles for 60+ cube attempts and so on, without having to worry about logistics of scramble generation, misscrambles, duplicate scrambles and so on.

8) Software feature of instant reconstruction from all the cube attempts (User has to feed his algsheet otherwise recon from computer vision on video feed is super hard). Also the splits for the entire attempt is generated automatically, without the user having to go through the video feed.

9) Automatic harmonica holder over the user's hand once it detects user has started execution. The system also sets alarm for one hour limit so that the competitor can stop the attempt at the hour mark, and can accurately know the number of cubes solved over the hour without having to refer to the video feed.
 
Aim - Make a scrambling robot that can kept on a bench, take 70+ cubes kept on it, scramble them, arrange them back in proper configuration, cover setup with cloth, alert the user and automatically start video recording of the entire setup for the duration of the attempt from an independent video recording device.


Dreamy infeasible details - Some features that I can think of from the top of head:

1) It can check for logos and Bluetooth cubes and remove them. It brings each cube in WCA orientation before it executes the scramble.

2) The robot will also stack the 60+ cubes in your desired arrangement. Also once the robot is done scrambling and arranging the cube, it pings a video system that starts taking video of the entire attempt. The user need not worry about scrambling as well as video recording.

3) The prototype should be cost effective. In the electronics, it can use Raspberry Pi or FPGA or a 32-bit processor, that takes to scrambling part and sensory parts of the system.

4)Design a product that has one basic MBLD scrambler + 70 identical 3x3 cubes + UI that user can use easily to automate MBLD scrambling process.

5) For competitions of scale of Worlds, where there are 100+ people who can do 60+ points, we have a mega robot installed at the venue. It can scramble all cubes at once in 1 minute and there will be no delay in the MBLD event. Robot should be able to detect competitor making accidental turns during memorisation and also find out the number of DNFs post attempt.

6) Benchmark the product to have smaller footprint on the table and capable of scrambling more 3x3s(~200 cubes) at once.

7) User can practice custom attempts like only 7 alg scrambles for 60+ cube attempts and so on, without having to worry about logistics of scramble generation, misscrambles, duplicate scrambles and so on.

8) Software feature of instant reconstruction from all the cube attempts (User has to feed his algsheet otherwise recon from computer vision on video feed is super hard). Also the splits for the entire attempt is generated automatically, without the user having to go through the video feed.

9) Automatic harmonica holder over the user's hand once it detects user has started execution. The system also sets alarm for one hour limit so that the competitor can stop the attempt at the hour mark, and can accurately know the number of cubes solved over the hour without having to refer to the video feed.
That sounds like quite a project. I don't know how feasible it is for an event as small as multi
 
5) For competitions of scale of Worlds, where there are 100+ people who can do 60+ points, ...
I'm sure when you put this in, you were trying to think optimistically towards the future. But considering no one has ever actually gotten 60+ points in competition yet, and only one person has ever gotten over 50 points officially yet, I think an expectation that there will someday be 100+ people who actually attend a single World Championship who can do 60+ points is pretty extremely unlikely, at least in our lifetimes. There are several people who "can do 60+ points", even if no one has actually officially done it, but several is a lot smaller than 100!

I suppose I know better than to underestimate abilities of people to improve. But I do predict that by the time we have 100+ people able to do 60+ points in multiBLD at Worlds, it will probably also require a sub-6 average to even make it to 3x3x3 speedsolve second round at Worlds. :-)
 
Troll Post ;)😅🤣

ANNOUNCEMENT: Final mass producing the robot scramblers in collaboration with Boston Dynamics!!!!

We have had the prototype since 2025, and now we are happy to tie up with the company Boston Dynamics to mass produce an amazing design for MBLD scrambling.

Since 2023, scrambling for MBLD event has been an issue, as many people are 60+ points, and at the venue, where there are 10 60+ MBLDers, it becomes a tough job for the scramblers, U S' U2 S' L S2 L' U (eg Michigan cubing 2029).

The prototype that I tested in 2025, can scramble 60 57mm cubes in 1 minutes. We have come a long way since then, l S' D' S M' D L'. There are some of the improvements in the mass produced one that is going to be released:

1) It can check for logos and Bluetooth cubes and remove them. A big help for the delegates. No more rampant cheating possible with Bluetooth cubes, reference to the incident in 2026 local comp, where a person had execution time of 7 seconds per cube even when completely blindfolded, E' R D M2 U' M U M D' R' E.

2)Since the new product is a walking robot too, the scrambling happens 1km away from the solving station (precisely), and the humanoid robot runs with the 60+ cubes to the venue and arranges it on the solving station using robotic arm. We kept the distance 1km as the company Boston Dynamics wanted to show its cool walking gait of the robot. The robot will also stack the 60+ cubes in your desired arrangement. And, as always, a pod will be separated from it when the attempt starts, and it will take videos from 5 different angles for the MBLD attempt, and record the solve time up to femtosecond, U' M2 F' L S L' F L' M2 S L U. Now, how cool is that!!!!
3) The prototype until now was extremely expensive and only the WR holder could use it ($10,000). Now we ship it for ($1,000) so that many people can benefit from it.

4) For people who still use audio for MBLD and are not good, we have a starter's pack for free (yep no money to be paid), basic MBLD scrambler + 15 GTS26 cubes free + free shipping.

5) For competitions of scale of Worlds, where there are 100+ people who can do 60+ points, we have a mega robot installed at the venue. It can scramble all cubes at once in 1 minute and there will be no delay in the MBLD event (Organisers can install it in venue for $800k + insurance charges) Finally we can avoid the mishap of Worlds 2027, when the MBLD event was delayed by a whole day because of so many duplicate scrambles, and time taken to scramble all the cubes.

6) Lastly, for the current MBLD WR holder who does an insane amount of cubes in an hour, we are happy to announce our MBLD scrambler pro, which took a team of 100 robotics engineers 4 years to perfect, and over $10 million into the project. It can scramble 200 cubes in 30 seconds without melting the plastic. It is a technological marvel, and we are happy to give the best product to the best Cuber. It is no joke that 200 points is one of the toughest barriers in MBLD, and the WR has been for 199 points for 2 years now. Congrats Chong So Lee, on getting the Scrambler Pro installed in your house. Now, you can do multiple 200 cube attempts in a day, R F' U' L2 S' L' S L' U F R'.
P.S. The scrambler pro also livestreams your attempts on many platforms in 16k 1000fps, so expect more twitch subs Chong So Lee. Again, I would like you say you are great, and the entire cubing community wishes that you reach 200 points soon.

Stay tuned to our company website for more updates. We are still working on MBLD simulator using brain computer interface. Hope get the some exciting news on that product soon, U' F R2 U R E' R' u' R2 F' U.
 
Back
Top