------=_Part_2640_10337955.1209583908202
Content-Type: multipart/alternative;
boundary="----=_Part_2641_30484632.1209583908203"
------=_Part_2641_30484632.1209583908203
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
When I am solving a 3D rubric cube I use a modification on the
layer method. The layer method involves solving one side of the cube, then
move on the next layer, until you get everything solved except the top side.
Then you need moves that can solve the pieces on the top of the cube. What I
do is I leave out one corner or edge piece in each layer. This creates a
hole that I can use to build up each layer, and later can be used to start
on the top side. When you are done with this you will have all the edges and
middles on the cube solved, and will be left with 4-5 corners. This is where
corner logic comes into play. I define a corner move as a move that
exchanges three of the same kind of pieces without messing up any other
pieces on the cube. This simplest version of this is to find a piece on the
cube that slides into place by turning one of the sides of the cube, then
take the piece out of the side, turn the bottom side so there is another
unsolved piece in the hole, then undo the moves you did to take the piece
out of the side. You will now hopefully have swapped three pieces. This same
logic will work on corners, edges, and middles.
When I solved the 4D cube for the first time I attempted to
follow something akin to the layer method I used on 3D cubes, however what I
ended up doing was solving two adjacent sides to start with rather than just
one side. I found that I could do this and still have the degree of freedom
required to solve the cube. I have attached a log file of me half way trough
solving the cube. I found I could use 3D logic in someplace on the 4D cube
without messing up other parts of it.
The question is whether corner logic will work on a 4D cube. The answer I
found was a resounding yes. First of all the first part of a corner move
where you take a piece out of its original side is a lot easier in 3
dimensions than in 4 dimensions. It is easy to take 3 or 2 pieces out of a
4D side but taking one individual piece out takes some more thought. (A 3D
corner move usually takes 7 moves but a 4D corner move takes between 12 and
18 moves) The other thing is there are more ways to mess up a 4D corner move
than a 3D one. A 3D corner piece can only be twisted three ways which means
there are only two ways to put a piece in a side incorrectly. In a 4D cube
there are many more so it is really easy to get a piece in the right side
but twisted incorrectly. Also since there are so many ways you can turn a
cube it is difficult to keep track of which three pieces you are exchanging.
I also noticed that 4D corner moves do not lend themselves to becoming
macros do to the large number of variations of moves that are possible. Even
on a 3D cube there are at least 6 different ways you can take a piece out of
a side.
My original goal of solving the 4^4 was to try to solve the parody problem
but that hasn't happened yet.
------=_Part_2641_30484632.1209583908203
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
When I am
solving a 3D rubric cube I use a modification on the layer method. The layer
method involves solving one side of the cube, then move on the next layer,
until you get everything solved except the top side. Then you need moves that
can solve the pieces on the top of the cube. What I do is I leave out one
corner or edge piece in each layer. This creates a hole that I can use to build
up each layer, and later can be used to start on the top side. When you are
done with this you will have all the edges and middles on the cube solved, and
will be left with 4-5 corners. This is where corner logic comes into play. I
define a corner move as a move that exchanges three of the same kind of pieces without
messing up any other pieces on the cube. This simplest version of this is to
find a piece on the cube that slides into place by turning one of the sides of
the cube, then take the piece out of the side, turn the bottom side so there is
another unsolved piece in the hole, then undo the moves you did to take the
piece out of the side. You will now hopefully have swapped three pieces. This
same logic will work on corners, edges, and middles.
When I
solved the 4D cube for the first time I attempted to follow something akin to the
layer method I used on 3D cubes, however what I ended up doing was solving two adjacent
sides to start with rather than just one side. I found that I could do this and
still have the degree of freedom required to solve the cube. I have attached a
log file of me half way trough solving the cube. I found I could use 3D logic
in someplace on the 4D cube without messing up other parts of it.
The question is whether corner
logic will work on a 4D cube. The answer I found was a resounding yes. First of
all the first part of a corner move where you take a piece out of its original
side is a lot easier in 3 dimensions than in 4 dimensions. It is easy to take 3
or 2 pieces out of a 4D side but taking one individual piece out takes some
more thought. (A 3D corner move usually takes 7 moves but a 4D corner move
takes between 12 and 18 moves) The other thing is there are more ways to mess
up a 4D corner move than a 3D one. A 3D corner piece can only be twisted three
ways which means there are only two ways to put a piece in a side incorrectly. In
a 4D cube there are many more so it is really easy to get a piece in the right
side but twisted incorrectly. Also since there are so many ways you can turn a cube
it is difficult to keep track of which three pieces you are exchanging. I also
noticed that 4D corner moves do not lend themselves to becoming macros do to
the large number of variations of moves that are possible. Even on a 3D cube
there are at least 6 different ways you can take a piece out of a side.
My original goal of solving the 4^4
was to try to solve the parody problem but that hasn't happened yet.