Show HN: Rubiks Cube Solver | Hacker News Reader

Show HN: Rubiks Cube Solver

speedcube.com.br

23 pointswozzp16 comments

Speedcube is an open-source platform for speedcubers featuring a Rubik's Cube solver, competition timer, algorithm library, and AI-assisted cube recognition directly in the browser.

Built with React, TypeScript, Rust, and Python, the project aims to become an all-in-one platform for cubers—from beginners to competitive solvers.

http://github.com/williamisnotdefined/rubiks-cube-solver/

Open articleView on HN

Note to the site author: Using Chrome's translate tool seems to break your UI. Upon first visit, I was prompted to translate from Portuguese to English and accepted. Subsequent visits required I click the "Translate this page" button on the right side of the URL bar. (Edit: Chrome 1490.7827.201 on Windows 10).

When translated, clicking the Solver drop-down (default 3x3x3) displays:

Unexpected error

Something went wrong. The current screen broke unexpectedly. Please try again or switch routes to reload it.

Interesting, but the site has support for 9 languages, what language were you using?

- instead of giving a solver, solve the harder problem

- make a tool that teaches me how to visualize a rubiks cube so that i can solve it myself

- make it something like this https://visualrambling.space/neural-network/

I will add it to my backlog, thank you very much.

that's a completely different project

Does it effectively achieve God's Algorithm (minimum theoretically possible sequence of moves to solve each position)?

God's algorithm is not computationally feasible on consumer hardware so I'd assume not although there are many algorithms that can get pretty close (either matching or 1-2 moves off the optimal solution) which are much faster to solve. If you're curious, look up Cube Explorer which is an app that's built for this.

God's algorithm has worked fine on consumer hardware for decades. Look up Korf. Even Cube Explorer has an optimal solver.

Not in practice. Computing the absolute minimum solution for every possible position is computationally infeasible for a web-based solver. This uses Kociemba’s two-phase algorithm instead, which produces very efficient solutions, usually close to optimal, without requiring enormous amounts of time and memory.

Why infeasible? cube20.org says "a good desktop PC" can optimally solve 0.36 random positions per second. And that's from the year 2010. I don't see why a web-based solver 16 years later should be much slower.

And apparently vcube optimally solved six cubes per second eight years ago already: https://github.com/Voltara/vcube

I think that "Randomise" button could be helpful. Or may be i haven't found it

do you want a scrambler?

yep

the scrambler is on timer page, there is given to the user a scrambler and a timer to practice. If you want to scrambler the solve page cube, go to the timer page, copy the scrambler (icon copy click) go back to solver page and paste the scramble alg, this way you will able to see the scrambled cube and see the "most likely best solution". please, let me know any feedback you have and thank you.

HN hitting new lows when slop like this makes it on my feed. This is neither original nor inspiring. Props on the umpteenth Rubik’s cube solver, I guess.

Your -3 karma is pretty self-explanatory. Hope you have a better day tomorrow.

Reply on news.ycombinator.com