Situational Compound Method

Since the late 90s or early 2000s, the most popular 3x3 speed solving method has been CFOP. Over time, CFOP has gradually incorporated steps and techniques from other methods. Rather than ask "Is it still CFOP?", I will state that the method that's being used is no longer completely CFOP and that top level solvers are now using a situational compound method, a melting pot of steps from various methods.

• Cross: This comes from the Layer By Layer method of the late 1970s.

• Edge control: Petrus is the original speed solving method that orients edges early, providing advantages in ergonomics and last layer case reduction. In the early 2000s edge control was added to CFOP so that it could have those same advantages.

• Block building: The Petrus method is primarily about block building for efficiency. CFOP incorporated this as XCross and beyond to the full on F2L-1 or F2L block building that top solvers are now using.

• ZBLL: Bernard Helmstetter first identified this as a useful last layer reduction and developed the algorithms in the year 2000 for the Petrus method.

• CLL + ELL: Guus Razoux Schultz developed this in the early 80s. Yiheng apparently sometimes uses this.

• Last Three Corners: Feliks used this sometimes, where the solve would end with two corners on the last layer and one on the bottom layer.

• ZBLS: From what was called the ZB method in the early 2000s. A joint creation by Zbigniew Zborowski and Ron van Bruchem.

• Full EO: This comes from ZZ in 2006, was then changed by the community from EOLine to EOCross, and has been seen in some top level solves.

• Extended Slot: Presented by me in 2006 and used in Nautilus and APB. Teodor Zajder's 2.76 world record uses this and other solvers are using it as well.

Looking at the above list, none of Cross, F2L, OLL, or PLL are 100% used in every solve, meaning the entire acronym is no longer always used and doesn't describe what is happening solve to solve. Edge Control altered OLL and PLL to become what's called ZBLL. Then we saw LSLL experimentation, such as L3C, change up the F2L letter of the acronym. Cross is the last of those steps to change with Extended Slot now being used. I think the top level solvers are seeing the same benefits that I did: freedom of movement, ability to solve more pieces at once, can manipulate other pieces, and it maintains good visibility and look ahead.

Someone may say that this is simply the natural evolution of CFOP. But we could say the same for all other methods. If Petrus had been the one to take off in the early 2000s, it would have evolved too. Solvers may have started adding in the last bottom layer edge as part of the 2x2x3 to improve ergonomics and look ahead. Maybe they would have delayed EO to the last slot for better control, or pulled EO further back to the beginning like ZZ. And maybe extended slot would have eventually naturally appeared.

A method may be good, but it doesn't mean that all of its steps are good and that the method should be used as a whole. A method may have all good steps, or just a single good step. So we are just in the natural evolution where the best steps of each method are being used. What really is a method anymore other than one person's idea that their specific set of steps is good? A method is just a group of steps or states that are often interchangeable with other methods, and we don't know in a group of methods that share steps among each other whether one set of steps is the best of the group. But what we are more naturally able to do is identify steps that are good.

What's really happening is that steps from various methods are being incorporated into this situational compound method that's continually evolving, whether those added steps were known or not known to already exist in other methods. We can relate this to 2x2 solvers using various methods or steps and it has just taken longer for 3x3 to get to that point because of the length and complexity. All of us in the method development subcommunity have been for years putting many hours of effort into developing self contained methods that we think could be better than CFOP. Little did we realize that what we actually were doing was devising steps that would eventually be put on trial and deemed worthy or not to be accepted into the situational compound method.

What will be added next? Will Roux FB become a start that is used in a notable percentage of solves? Will solvers start learning Last Extended Slot algorithms? What about some sort of Last X Edges from multiple layers?

It may be time for a new name or acronym that describes this situational method that's being used, one that uses the best steps for the given scramble. Preferably a name that doesn't focus on layers because, as seen above, the method is even moving beyond solving layers or ending in a last layer. It would probably be best if the name is applicable to all puzzles since it is happening in 2x2, Pyraminx, and maybe others. Situational Compound Method is the name that I've been using for the concept.

For related discussion, see Joseph Tudor's thread about method convergence, and my follow-up to that. This third article that you're reading is the examination of method convergence and what's happening in high level solves to identify the concept at the core. This being that high level solvers are moving away from CFOP and toward a method or solving concept that is a combination of various methods.