Comments

• 

  3 days ago by aschnell | Reply

  So the IDA* algorithm looks more clear now. It finds the solution by iterating the "game tree" but concentrates on the "best" path on every iteration. This is done using a heuristic.

  BTW: Back then test research papers were considered part of artificial intelligence so I'm also doing AI after all.

×
Edit Comment 8857
# A First Level Header
## A Second Level Header

Use one asterisk to *emphasize*

Use two asterisks for **strong emphasis**

- Use hyphens
- for unordereed
- lists

This is an [link to example.com](http://example.com/)

This is an image ![an openSUSE geeko icon](https://en.opensuse.org/images/d/d0/Icon-distribution.png)

This is a user link @hans

This is a project link hw#some-cool-title
More Complex Markdown Help
Formatting Help
• Edit
• Preview
So the IDA* algorithm looks more clear now. It finds the solution by iterating the "game tree" but concentrates on the "best" path on every iteration. This is done using a heuristic. BTW: Back then test research papers were considered part of artificial intelligence so I'm also doing AI after all.
 
×
Reply to aschnell
So the IDA* algorithm looks more clear now. It finds the solution by iterating the "game tree" but concentrates on the "best" path on every iteration. This is done using a heuristic.
BTW: Back then test research papers were considered part of artificial intelligence so I'm also doing AI after all.
---
# A First Level Header
## A Second Level Header

Use one asterisk to *emphasize*

Use two asterisks for **strong emphasis**

- Use hyphens
- for unordereed
- lists

This is an [link to example.com](http://example.com/)

This is an image ![an openSUSE geeko icon](https://en.opensuse.org/images/d/d0/Icon-distribution.png)

This is a user link @hans

This is a project link hw#some-cool-title
More Complex Markdown Help
Formatting Help
• Edit
• Preview
 
• 

  2 days ago by aschnell | Reply

  I have looked at various programs implementing IDA* based solvers. Many are unfortunately unfinished, broken and undocumented.

  The best so far is https://github.com/benbotto/rubiks-cube-cracker. Among other things it properly implements the index to the pattern databases while other programs seem to simply use a hash. It also gives speed measurements: It is really slow (several hours and more). The Korf IDA* solver is apparently still so slow.

  That made me think what the online solver at ruwix.com uses since it usually finds a good solution (never had more that 20 moves) in less than a minute. Maybe it uses another approach called the "Two-Phase-Algorithm" (https://kociemba.org/cube.htm). That also used IDA* but combines it with some group theory.

×
Edit Comment 8941
# A First Level Header
## A Second Level Header

Use one asterisk to *emphasize*

Use two asterisks for **strong emphasis**

- Use hyphens
- for unordereed
- lists

This is an [link to example.com](http://example.com/)

This is an image ![an openSUSE geeko icon](https://en.opensuse.org/images/d/d0/Icon-distribution.png)

This is a user link @hans

This is a project link hw#some-cool-title
More Complex Markdown Help
Formatting Help
• Edit
• Preview
I have looked at various programs implementing IDA* based solvers. Many are unfortunately unfinished, broken and undocumented. The best so far is https://github.com/benbotto/rubiks-cube-cracker. Among other things it properly implements the index to the pattern databases while other programs seem to simply use a hash. It also gives speed measurements: It is really slow (several hours and more). The Korf IDA* solver is apparently still so slow. That made me think what the online solver at ruwix.com uses since it usually finds a good solution (never had more that 20 moves) in less than a minute. Maybe it uses another approach called the "Two-Phase-Algorithm" (https://kociemba.org/cube.htm). That also used IDA* but combines it with some group theory.
 
×
Reply to aschnell
I have looked at various programs implementing IDA* based solvers. Many are unfortunately unfinished, broken and undocumented.
The best so far is https://github.com/benbotto/rubiks-cube-cracker. Among other things it properly implements the index to the pattern databases while other programs seem to simply use a hash. It also gives speed measurements: It is really slow (several hours and more). The Korf IDA* solver is apparently still so slow.
That made me think what the online solver at ruwix.com uses since it usually finds a good solution (never had more that 20 moves) in less than a minute. Maybe it uses another approach called the "Two-Phase-Algorithm" (https://kociemba.org/cube.htm). That also used IDA* but combines it with some group theory.
---
# A First Level Header
## A Second Level Header

Use one asterisk to *emphasize*

Use two asterisks for **strong emphasis**

- Use hyphens
- for unordereed
- lists

This is an [link to example.com](http://example.com/)

This is an image ![an openSUSE geeko icon](https://en.opensuse.org/images/d/d0/Icon-distribution.png)

This is a user link @hans

This is a project link hw#some-cool-title
More Complex Markdown Help
Formatting Help
• Edit
• Preview