- 最后登录
- 2024-11-21
- 在线时间
- 938 小时
- 阅读权限
- 40
- 注册时间
- 2008-1-6
- 积分
- 1960
- 帖子
- 1075
- 精华
- 6
- UID
- 17579
- 性别
- 保密
- 积分
- 1960
- 帖子
- 1075
- 精华
- 6
- UID
- 17579
- 性别
- 保密
|
page 1 THE 45-52 MOVE STRATEGY Introduction. Let G= <L,R,F,B,U,D> , G1= <L,R,F,B,U2,D2> , G2=
<L,R,F2,B2,U2,D2> , G3= <L2,R2,F2,B2,U2,D2> . The plan is to manoeuvre down through the chain
G=G0> G1> G2> G3> 1. One gets from Gi to Gi+1 by using moves in Gi only. In its shortest form
this strategy would be executed with the help of a computer, in which case I conjecture only
45 moves would be needed, but here I have sacrificed 7 moves in order that there should be no
need for a computer. With a few more pages of tables, the figure 52 could be reduced to an
intermediate figure of, say, 49. I intend to do this shortly! The indexes of the chain of
subgroups are 2048, 1082565, 29400, 663552, but these figures are considerably reduced by
considering symmetries. The reader may check that these indexes multiply together to give the
order of G. The accompanying tables are broadly classified according to corner positions, and
in detail according to edge positions. The words listed actually produce the positions under
consideration, so the restoring moves are the inverses of these. In order to be able to use
the tables it is necessary to understand the basic characteristics of the groups G1, G2, G3,
so the necessary facts are presented below. Getting into G1. This involves edge pieces only,
and is easy, for which reason no tables are given. An edge piece is BAD if in taking it home
an odd number of quarter-turns of U and D faces is needed; otherwise it is GOOD (note that
badness is well-defined). The reader may quickly work out a rule of thumb for deciding
whether a piece is GOOD or BAD. Now quarter-turns of either U or D faces convert BAD pieces
to GOOD and vice versa; other moves have no effect. Therefore to make all edge pieces GOOD,
move groups of them to U or D face avoiding quarter-turns of U or D, and then cure them by
performing a quarter-turn of U or D. For example, if all twelve are BAD, DBFUR'L'D will cure
them all!
第一页 45-52步策略介绍
令 G=<L,R,F,B,U,D>,G1=...,G2=...,G3=...计划这样降下来:
G=G0>G1>G2>G3>1. 从Gi进入Gi+1,仅使用Gi规定的操作。这种方法最简短的形式,用电脑来算的话,估计
45步足够了,但这里,牺牲7步,电脑其实不需要。使用几页纸的表,52可以缩减到49.子群链的indexs分
别是:2048,1082565,29400,663552,但显然,这些数字可以用对称来减小。读者可以检查,这些数乘起来
,正好是群G的秩。陪集表用角的位置来分类,用棱的位置来详细描述。列出的表是思考的过程,因此,还
原的顺序要颠倒过来。为了能使用这些表,要理解群G1,G2,G3的基本属性,因此,下面介绍必要的情况。
进入G1
这一步仅涉及到棱,并且很容易,因此,没有表给出。“错误”的棱是指需要奇数次的U与D操作才能返回
原位的;其它情况就是“正确”的。(注意:“错误”的方向有了明确的定义。)读者很容易发现确定一
个棱的“好坏”的准则。这样,90度的U或D旋转就用来把“坏”棱变“好”,“好”的变“坏”;其它操
作没有影响。因此,要把所有的棱方向变“好”,只要不使用奇数次的U,D旋转,把它们集中到U面或者D面
上,然后做一个1/4周的U或D面的选装。例如,如果所有的棱都是“坏”的,这个操作可以把它们都翻正:
DBFUR'L'D。
(现代的习惯,先限制F,B;也就是把G1定义为<F2,B2,U,D,L,R>。在定义“错误的棱向”的时候,用奇数
次的FB;翻转棱的状态时,是集中在F,B面,然后用奇数次的F,B翻转。也就是说:如果12个棱的方向都不
正确,翻转公式可以是:
B L R F U' D' B 。
在定义G1,G2,G3时,限制的顺序不同,就会产生“优先级”不同的感觉。先限制F,B产生的效果,同某些盲
拧方法对棱向的定义是一致的。
)
[ 本帖最后由 aubell 于 2011-2-7 01:20 编辑 ] |
|