Intuitive State Reduction

An LSLL solving method idea that I've had for a while is to use an algorithm to reduce the LSLL to one of a set of simple states. These would be states that have the fastest algorithms. This is essentially the algorithm combination concept seen in Petrus 270 or Duplex applied to LSLL. It isn't a new concept. However, an alternate idea is to reduce to states that are easily solved intuitively. This way there is no recognition step after the first algorithm application and there are just a few remaining turns. Because this would require recognition of the majority of the LSLL pieces, we can start by making use of ZBLL or 1LLL to reduce to these states.

A few example states include the following. They are setups to the example states, so execute each forward on a solved cube. With additional AUFs, the number of possible states is increased. Additional desired end states can be added in to this system.

• R

• U R

• U' R

• U2 R

• R U R'

• R U' R'

• R U2 R'

Upon reaching the final slot, place a final corner and edge pair that sets up to one of the ending state that involves that kind of pair. Or build in this kind of pair into the F2L during inspection. The major advantage of taking care of this during inspection is that you can reduce or remove the setup moves for the pair that would otherwise have been involved during the final pair. This can also remove any blind spots early in the solve. This can also be performed at other points of the solve if advantageous. The examples images below show some of the pair types. Then, after the application of ZBLL or 1LLL relative to the pair type, the final intuitive turns can be executed.

This can also be incorporated into the normal pieces involved during inspection. Instead of a usual cross edge, to form the desired end state early, an offset cross edge can be planned with the matching pair type.

It may seem that recognition would be difficult. With Straughan recognition or Polar recognition, pseudo ZBLL and 1LLL recognition is now much easier. Depending on the configuration of the corner and edge pair, recognition slightly differs. You should know the kind of pair you created, so you will be prepared and know the recognition type before arriving at that step.