Optimal Rubik's Cube Solver Written and Copyright © 2008-2009 by Andrew Brown <firstname.lastname@example.org>
- A C compiler w/ standard libraries, I use gcc
- python to run the cube_convert.py script
- Ensure you have a C compiler installed as well as standard library headers. You will also need the "make" program. No other fancy libraries are needed.
- Run the "make" command in the source directory to compile all the files. This will produce a program called "main" which is used to generate the heuristics tables and to solve the cube
RUNNING THE PROGRAM
First you will need to generate the heuristics tables. This could take a while.
- Run the compiled program with "./solver"
- Select one of the three table generation options.
- Repeat 1-3 for the other two tables
Verify the table generation worked with these md5 signatures:
Next obtain a scrambled cube to solve. The solver expects a sort of cryptic format, so I have provided a script where you just enter the colors on each side and it spits out the "cube string".
Run the cube_convert.py script. It will give you this template:
1 2 3 4 R 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 B 22 23 W 24 25 G 26 27 Y 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 O 45 46 47 48
It will ask you to enter in the colors of each face in the shown order. Note
that you don't enter the center faces, those are fixed, so skip over those.
For example, a solved cube would be entered like this:
rrrrrrrrbbbwwwgggyyybbwwggyybbbwwwgggyyyoooooooo Which gets translated to:
(except all on one line)
Finally, to solve a cube: Run the compiled program "./solver" and choose the solve a cube option. Paste in your translated 120 character cube string all on one line and press enter.
Wait for solutions!
I was re-writing how the heuristics tables are computed when I last worked on this. Specifically, I changed how a cube configuration was mapped onto an index into the table. You can see the new version in corner_table.c, but I never got around to changing it for the two edge tables.
Also, this was like a year ago, so I don't remember if my new version was extensively tested. If you find it doesn't work, try checking out the version tagged "v1.2", which I'm fairly sure works. It's been so long though I don't remember what works and what doesn't.
I wrote this program completely from scratch, using ideas from Richard Korf's papers and other various internet resources. Any ideas, interesting comments, or suggestions for improvements please email me! -Andrew Brown <email@example.com>