Detailed Hints for the First 2 Layers

The F2L (First 2 Layers) are likely the most difficult phase to master properly. Unlike the LL (Last Layer), where the algorithms are pretty much determined and "all you need to do" is to quickly recognize a fixed set of positions and memorize the algorithms, the F2L can be done in many different ways depending on which corner-edge pair you spot first. The F2L require more experience, insight, and ability to "see" the cubies on all sides, but the number of algorithms you need to memorize is minimal. The LL requires the opposite - a lot of memorizing and speed. It seems to be the case that mastering the LL is actually simpler and faster than the F2L. Many cubists report being slowed down by the F2L rather than the LL. Because I receive a number of requests for additional tips and hints on the F2L from frustrated cubers, I decided to provide another set of tips specifically focused towards the F2L.

First of all, do not sweat the cross. If you can do 3 edges and cannot figure out all 4, do just three and then, as you are inserting them, look for the fourth one. It is OK to use, say 10 moves, to do the cross instead of the optimal 7 or less. Do not try to memorize the cross at all price. Sometimes I am, too, forced to plan only three cubies if they are scattered in unfavorable positions, but I can still finish in teens.

F2L require a lot of experience but are lighting fast once mastered. Instead of memorizing the algorithms from my page, try to develop your own algs and then learn only those that you are having problems with. Do not just apply the algorithms from my F2L page blindly. Here is my advice:

1) There are two basic ways to bring a corner from the LL into the 1st layer (OK, with the exception when the white is on the bottom): the one that "picks up a distant edge along the way" (No. 18 on my F2L page) and the one that inserts the corner and the edge that is right behind it (No. 11 on my F2L page). Both moves are 3 (or 4) moves long. Now, 90% of the situations can be composed of combinations of those two basic moves. The remaining 10% are less obvious but only a move or two shorter than those "obvious" composed of those two basic moves. The secret is that you do not really need any special algs for the F2L. The savings you gain from using the shortest ones instead of the obvious ones make a small difference that becomes important only when you try to get below 20.

2) In the beginning right after the cross, you can use even simpler and shorter moves to insert the edges because you can use the fact that no corner-edge pair is in place. Again, those simple 3-move basic "algorithms" can do the job. Examples of this are given here.

3) Finally, when I finish the cross, I look at the last layer searching for pairs of cubies. In most cases, there is at least one pair already. Do that one and then repeat again. If you at some point do not see any pairs in the LL, you can do the following. Say, you have a corner in the LL and the edge is somewhere in the middle in some incorrect place. Then use the bottom layer to move the corner to the edge (or down-left or down-right see the examples) and do that 3-move to "pick up" that edge. You can always pick it up into one of the two positions that will only need one of those basic moves to insert the pair into the correct place.

You rarely, very rarely, need to remove an incorrectly positioned corner (or an edge) from the F2L in order to proceed. That will hold you back. Usually, you can do something like I described, which is fast, filled with two-side moves, finger shortcuts, and it is easy to see.

4) Now, I can't emphasize this enough: GO SLOWLY. The system works best when you do not apply the algorithms blindly but rather you try to minimize the number of moves, ignoring the time for a while. This is why I also always advice to forget about the time and, instead, start paying attention to the number of moves in the F2L. Then, _very_gradually_ increase the speed. This is the best warm up as well, by the way. Much better than twisting like crazy right from the moment you pick up the cube.

In the beginning, it may happen to you that you will be making unnecessarily many preparational turns to get the cubies into positions that correspond to the pictures from my F2L set of algorithms. If this happens, reread this article and go through the examples. There is a big space for improvization and improvement in the F2L. This is good because it means that one can develop individual approaches or discover new strategies, but it is also bad because the F2L phase is hard to explain in simple terms. To give you a better idea of how I approach the F2L and how I think out loud during solving, I prepared some examples of solving the F2L. Ron van Bruchem also gives a few examples taken from a video of his actual speedsolving.