• 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!

Want help using Cube Explorer correctly and calculating cases

Status
Not open for further replies.

Athefre

Member
Joined
Jul 25, 2006
Messages
1,248
Since I started solving the cube almost four years ago I've been into the ideas and figuring out things more than speed.

My two biggest problems have been:

1. Calculating the amount of cases for steps. In my time using the forums, somehow I've never seen anyone describe how they do it. I've always turned to Bruce Norskog to do it for me. Anyone here willing to tell me how it's done?

2. Using Cube Solver, ACube, and Cube Explorer. I've been stuck finding my own sequences for the steps I think of. I usually end up with ones that are very short and very comfortable but there are usually some very bad ones that are 16 or 17 moves long.
-So I've tried using Cube Solver but it seems to take forever, I never get anything out of it.
-I've tried figuring out ACube many times but I just don't understand it. It says something like putting stuff in "command prompt". That's the "Run" thing right? I've tried typing what it says but it just doesn't work for me. I understand how to put the cubies in the program and get it to find something, that's simple. The problem is getting it to find the type of sequences I want.
-I just downloaded Cube Explorer, it's very nice. But, I'm having about the same problem as ACube. I set up the cube (or put in the moves) and get it to find a solution. The solutions it gives are almost completely half-turns though.

I like Cube Explorer the most so if anyone can, please tell me how to use it.
 

miniGOINGS

Member
Joined
Feb 27, 2009
Messages
3,049
1. Cube theory. Ask a question and I'll let you know how to figure it out to the best of my ability.

2. I either use CubeSolver, manipulate other algs, or figure it out on my own.
 

4Chan

Premium Member
Joined
Jun 21, 2008
Messages
2,984
Location
Lumbridge
YouTube
Visit Channel
I know that Acube can be manipulated for results you want.

This post may help.
http://games.groups.yahoo.com/group/zbmethod/message/26

I've found Acube to be more helpful finding RU and RFU algs

Personally, I've made QUITE a few algorithms with cube explorer, so I prefer it. :p

Sometimes AUF can help find better algorithms.

Fiddling around with the algorithm also makes it rather nice.
 
Last edited:

Athefre

Member
Joined
Jul 25, 2006
Messages
1,248
I'll try to be specific. I feel like I know what I'm doing with a cube, I've been solving and visiting the different forums for four years, but I know I'm not a master or even close. I know I can take a solution that the program finds and change it to be more comfortable.

Cases - The best example I want to have explained is the amount of cases for orienting four corners. There are 7 cases, I know that. But how would I be able to figure that out myself?

Solvers - Currently reading the page I was shown. Easy to read.

EDIT: I'm further than ever with ACube. I found "Run", and put in "Command" (I didn't even know this black and white box existed). Now I don't know what exactly to type from there.
 
Last edited:
Joined
Jan 17, 2009
Messages
377
Location
Portland, OR (where the sun don't shine)
EDIT: I'm further than ever with ACube. I found "Run", and put in "Command" (I didn't even know this black and white box existed). Now I don't know what exactly to type from there.

First off, you can add stuff to the end of the run command instead of the $!. a means you want a massive list of solutions. s means you want solutions to be seen as optimal in slice-turn metric. q means you want solutions to be seen as optimal in quarter-turn metric.

Before you enter the state of the cube, you can enter the octal number corresponding to which generators you want to use. Easy way to calculate: RDFBLRESM is the order. Now divide it into 3 parts.

First digit: RDF - R=4, D=2, F=1. Just add up the total of the generators you want to include.

Second digit: BLR - B=4, L=2, R=1. Just add up the total of the generators you want to include.

Third digit: ESM - E=4, S=2, M=1. Just add up the total of the generators you want to include.

Now, if you want an alg consisting of only the group <R,D,L,E,M>, you get the number (4+2) (2) (4+1), or 625. It's very handy and very simple once you understand how to work it.

Initial turns are weird. Just put a ~ before whichever turns you feel like allowing to be turned before the solution is found. This is kind of hard to explain. If you entered ~U, ACube will go through the solution process in your entered state and then each of the other 3 positions possible from executing a U turn. Initial turns do not count as taking up a move as far as optimal solutions are concerned. DO NOT USE INITIAL TURNS WITH ALL SEARCH ON!!! You will get waaaaaaaaaay too many results.

Now enter your cube configuration:
UF UR UB UL DF DR DB DL FR FL BR BL UFR URB UBL ULF DRF DFL DLB DBR
This is asking you where the cubie in the designated position wants to go. Orientation matters! UF is different from FU. However, UFR is the same as URF. You must define each cubie before you can go to the next one.

This is where ! comes in. ! is a shortcut used to show that the cubie is in the correct place and orientation. -! means the cubie is rotated counterclockwise from how its solved state. +! means clockwise. @! is rather useless.

? means you don't know what cubie is there, but you know its orientation. Again, you can modify it with -, +, and @. @? is very useful, telling ACube you know nothing of that piece and that ACube should ignore it.

!!, ??, -!!, @??, and the rest mean that ACube should repeat this marking until the end of the piece type (edges, corners).

Q or q instead of a cube state gets you out of the program.

I always use -? or -! for unoriented edges. Just a habit.

Some other notes: ACube solves in reverse order. Longest solution to shortest. It also gives a conformation solve, usually around 25-35 moves long to let you know something will happen. I have had the program hang on "depth 0...", and which point you have to quit and get back in.

Examples:

F PLL - block to front, allowing U and D rotations before solution, restricting generators to <R,U,L,D>: ~UD 630 ! UL ! UR !! ! UBL UBR !!

Some dot OLL, no restrictions: -? -? -? -? !! -? -? -? ? !!

CMLL case, T orientation, restricted to <R,U,L,F>: 530 @? @? @? @? @? ! @? !! UBR RUF BLU UFL !!

Hope that's enough info on ACube. If you have any questions, just ask - that's why this forum is here! (that and showing off...)
 

Athefre

Member
Joined
Jul 25, 2006
Messages
1,248
That seems very helpful. So do I really need to use "Run" to get the type of sequences I want?

If so, I still need a step-by-step guide on how to do that. I know this so far:

1. Open Run
2. Type "command"
3. This is where I have to stop. The page I was linked to said to go to the ACube directory. What exactly do I type to do that?

As a note so I don't get accusations from some people (not directed towards you). I have read the ACube readme many many times since I first downloaded it 3.5 years ago. I'm not a programmer so I just didn't understand the readme. Well, only how to get ACube started the way I want.

EDIT: My own example:

1. I opened ACube
2. I wanted to find a better sequence than R2F2RUL'U2RU'M'BL'B2RB'R'B2RLx'. It is the sixth case on this page. It's already comfortable, but it is way too long.
3. So I put ~U 531 ! ? ? ? ? ? ! ! ! ! ! ! URB UFR ULF UBL DRF DFL DLB DBR. I wanted it to permute the corners and place the correct edge at DF while keeping the other edges (UB, UL, UF, UR, DR) oriented. UB, UL, UF, UR, and DR don't have to be permuted correctly, they just have to keep their orientation. DR could end up at UL for example.
4. It only found one solution, which was terrible. (U) L2 U2 F2 U L2 U2 F2 U' L2 U2 F2. It's 11 moves, which is great, but it is almost completely half-turns. R, L, F, and M moves only is my goal. But B moves would be fine and I don't want D, E or S. They almost never turn out comfortable.
 
Last edited:

Oliver Wolff

Member
Joined
Sep 22, 2008
Messages
7
Btw I made an Excel sheet with pictures and optimized (concerning lenght) algs for all 3915 LL cases years ago but never published it yet.

Maybe you (or others) can find nice algs out of this. It has about 4.5 MB and I never had or took the chance of making it accessible for the community.

Is this possible on speedsolving.com or is anybody else willing to do that?

Can I simply zip it and attach it to a post?
 

Athefre

Member
Joined
Jul 25, 2006
Messages
1,248
I think you can attach a zip file. When you click "Post Reply", below the text field in "Additional Options" you can click "Manage Attachments".
 

4Chan

Premium Member
Joined
Jun 21, 2008
Messages
2,984
Location
Lumbridge
YouTube
Visit Channel
Btw I made an Excel sheet with pictures and optimized (concerning lenght) algs for all 3915 LL cases years ago but never published it yet.

Maybe you (or others) can find nice algs out of this. It has about 4.5 MB and I never had or took the chance of making it accessible for the community.

Is this possible on speedsolving.com or is anybody else willing to do that?

Can I simply zip it and attach it to a post?

Yummie~

I am very interested in this list.
I'm also interested in how the pictures are, mayhaps they're better than the other 1LLL list I've seen.
 
Joined
Jul 11, 2008
Messages
437
Location
Tennessee
Btw I made an Excel sheet with pictures and optimized (concerning lenght) algs for all 3915 LL cases years ago but never published it yet.

Maybe you (or others) can find nice algs out of this. It has about 4.5 MB and I never had or took the chance of making it accessible for the community.

Is this possible on speedsolving.com or is anybody else willing to do that?

Can I simply zip it and attach it to a post?

Yummie~

I am very interested in this list.
I'm also interested in how the pictures are, mayhaps they're better than the other 1LLL list I've seen.

Yes, please post those soon.
 

Athefre

Member
Joined
Jul 25, 2006
Messages
1,248
To explain my case calculation troubles better:

Number of Cases for Orienting the four corners on U-layer. I know there are 7 cases but I can't figure out myself why there are 7 cases. I try to do....3 possible orientations for the first corner multiplied by 3 possible orientations for the second corner....and I'm already over the limit. I know there are cube laws (can't swap just two edges or just two corners, things like that).
 

miniGOINGS

Member
Joined
Feb 27, 2009
Messages
3,049
To explain my case calculation troubles better:

Number of Cases for Orienting the four corners on U-layer. I know there are 7 cases but I can't figure out myself why there are 7 cases. I try to do....3 possible orientations for the first corner multiplied by 3 possible orientations for the second corner....and I'm already over the limit. I know there are cube laws (can't swap just two edges or just two corners, things like that).

The first corner has 3 possible orientations, same with the second and third corners. That totals 27 possible orientations because the last corners orientation is fixed. There are 4 possible U face orientations, but the H case has only 2 because it is symetric and the solved state only has 1. 1 + 2 + 4 + 4 + 4 + 4 + 4 + 4 = 27.
 

Athefre

Member
Joined
Jul 25, 2006
Messages
1,248
I think I see now, I see the 7 cases in those numbers (all six of the 4s and the 2). But I still don't think I would be able to figure out other things on my own.
 

Athefre

Member
Joined
Jul 25, 2006
Messages
1,248
Thanks Stefan. Almost understand it, I'm just trying to find the 28 in the corner orientation example.

1 + 2 + 4 + 4 + 4 + 4 + 4 + 4 = 27

Fixed points:

1. Rotation through 90 degrees.
2. Rotation through 180 degrees.
3. Rotation through 270 degrees.
4. The identity.

27/4 isn't 7. Or did I read the page wrong?
 

Forte

Member
Joined
Jul 12, 2009
Messages
1,151
Location
Waterloo, Ontario
WCA
2009SHIN02
YouTube
Visit Channel
ummm . . .

A. Rotation by 90º: 1 solved = 1
B. Rotation by 180º: 1 solved, 2 H = 3
C. Rotation by 270º: 1 solved = 1
D. Identity: 27

(1+3+1+27)/4 = 8

Which is 7 OLL cases and one solved. I believe I did that correctly but I'm not sure :p
 
Last edited:

Forte

Member
Joined
Jul 12, 2009
Messages
1,151
Location
Waterloo, Ontario
WCA
2009SHIN02
YouTube
Visit Channel
Double post, whatever.

I tried doing this counting reflections as the same thing.
I'm not sure if I did this right, but here goes:
A. Rotation by 90º: 1 solved = 1
B. Rotation by 180º: 1 solved, 2 H = 3
C. Rotation by 270º: 1 solved = 1
D. Reflection through horizontal midline: 1 solved, 2 H, 2 pi, 2 T, 2 U = 9
E. Reflection through vertical midline: 1 solved, 2 H, 2 pi, 2 T, 2 U = 9
F. Reflection through one diagonal: 1 solved, 2 L = 3
G. Reflection through other diagonal: 1 solved, 2 L = 3
H. Identity: 27

(1+3+1+9+9+3+3+27)/8 = 7

Which is the solved case and the 6 corner orientations (because sune and antisune are the same).
 
Joined
Jan 17, 2009
Messages
377
Location
Portland, OR (where the sun don't shine)
That seems very helpful. So do I really need to use "Run" to get the type of sequences I want?

If so, I still need a step-by-step guide on how to do that. I know this so far:

1. Open Run
2. Type "command"
3. This is where I have to stop. The page I was linked to said to go to the ACube directory. What exactly do I type to do that?

As a note so I don't get accusations from some people (not directed towards you). I have read the ACube readme many many times since I first downloaded it 3.5 years ago. I'm not a programmer so I just didn't understand the readme. Well, only how to get ACube started the way I want.

EDIT: My own example:

1. I opened ACube
2. I wanted to find a better sequence than R2F2RUL'U2RU'M'BL'B2RB'R'B2RLx'. It is the sixth case on this page. It's already comfortable, but it is way too long.
3. So I put ~U 531 ! ? ? ? ? ? ! ! ! ! ! ! URB UFR ULF UBL DRF DFL DLB DBR. I wanted it to permute the corners and place the correct edge at DF while keeping the other edges (UB, UL, UF, UR, DR) oriented. UB, UL, UF, UR, and DR don't have to be permuted correctly, they just have to keep their orientation.

First, you want to enter this on your command line. Keep in mind that you have to specify the path that you have saved all this to. So before the ACube3.jar you have to give the file extension:

@javac -target 1.1 -O -d cls src\*.java src\pack\*.java src\transform\*.java src\prune\*.java; @jar cf ACube3.jar -C cls; java -cp ACube3.jar ACube $!

That will do everything at once and launch the program. The $! placeholder at the end should be used the first coupla times, as when you enter that it will give you the options you have available. Otherwise you just have to tinker with the program a lot.

To answer you issue with all half turns, just place an 'aq' (just the letter part) after java -cp ACube3.jar ACube. This gives you (1) all solutions so you have more to choose from and (2) quarter turn metric so you can restrict half turn moves.

The last few entries when doing this:

Code:
depth 16...
F U F' R' F' U' F' L' F' L U L' F L R U' . (16q, 16f, 16s)
F' U' F L F U F R F R' U' R F' L' R' U . (16q, 16f, 16s)
L F' L' U F' R F U F' R' F U F' U F U . (16q, 16f, 16s)
L F' L' U2 F U L' U L U' F2 U F U . (16q, 14f, 14s)
L R F R' U R F' R' F' U' F' L' F' U F U . (16q, 16f, 16s)
L' R' F' L U' L' F L F U F R F U' F' U' . (16q, 16f, 16s)
R' F R U2 F' U' R U' R' U F2 U' F' U' . (16q, 14f, 14s)
R' F R U' F L' F' U' F L F' U' F U' F' U' . (16q, 16f, 16s)

I have no idea how to come up with the number of possibilities other than pen and paper and some logic. According to me there are 501 ZBLL cases, but other than that I'm usually pretty close.

How I count:
Number of corner orientations then permutations
Number of edge orientations then permutations
Cancellations

Cancellations are hard to explain, but using AUF is my default example. The others should be pretty obvious once you know how to look for them, and that comes with practice.
 
Last edited:
Status
Not open for further replies.
Top