Main

EOLR for Roux: In 2012 I proposed both EOLR and EOLRb, and developed all of EOLR and half of EOLRb. This is a technique that has become commonly used among Roux users, even to break world records.

DFDB recognition for Roux: DFDB is a recognition method that solved the Roux step 4c edge 3-cycle recognition problem and is also commonly used. This has possibly been used to break world records.

LEG-1: I developed LEG-1 in 2012 and it has since become something that is occasionally used by 2x2 solvers. It has even been used in world record averages.

Group reference addition to the standard notation: Using Singmaster notation we typically reference a corner or edge by its intersecting layers (UFR for example). In 2010 I had the idea to use notation to reference groups of pieces or stickers. For example: dFR is the intersection of the wide d layer (bottom two layers), the F, and the R layer. This intersection identifies the F2L pair at the front right.

Minimum sticker recognition concept (Straughan recognition): This allows for recognizing CLL using just four stickers total, ZBLL using just six, and a large sticker reduction for many other steps.

Nautilus method concept: Nautilus started as a 3x3 method, but has since proven to be an interesting concept shape of its own - solve all layers below the top layer, leaving out all front layer pieces to the right of the left layer. This takes care of most blind spots on any puzzle and creates an interesting move set that includes the use of the upper layer and all layers to the right of the left layer. Nautilus has been adapted to several other puzzles.

Solved the pseudo corner recognition problem: This is one of my first developments. In 2004 Gilles Roux published the idea of starting with the LR color orientation, then checking the U sticker pattern (NMCLL 1). But this turned out to be too slow for speedsolving because of the second step. I solved this problem in 2010 with the development of NMCLL 2, making pseudo corner recognition possible in speedsolving. In 2021, the NMCLL 2 recognition method was simplified to require checking just two stickers in the second step. This created the first corner recognition method that uses five stickers (ATCRM) - one fewer sticker than the typical U sticker based recognition.

ACMLL, AOLL, ACLL and others: In 2022, I developed the concept of partially solving groups of 3x3 pieces then correcting them during an algorithm of a later step. This is like EG, but applied to 3x3. It is also similar to the undeveloped PEG idea by Kenneth Gustavsson. In the case of ACMLL, the two blocks of the Roux method can be solved imperfectly then the CMLL algorithm will include the correction. This reduces the overall solve move count, improves ergonomics during the second block to more often consist of all R and U turns, and allows for deeper inspection to fully plan the two blocks. I have also created AOLL for CFOP to allow for always solving the final pair using R and U. This can also be applied to Petrus to eliminate the EO step, still allowing for an all R U right block, but requiring a two step last layer or large number of 1LLL algorithms.

Pattern Corner Permutation: In methods such as CEOR, 2GR, or CP Nautilus, the U and R corners are placed into a <U, R> solvable orbit during the first steps. The recognition methods all involve the use of tracing - looking at each corner and following a path to determine the permutation state and any necessary swaps. PCP is the first ever early corner permutation recognition method that doesn't need tracing. Instead it is similar to corner recognition in CLL or CMLL where sticker patterns are checked.

The addition of unions to the multi-state solving concept: In the early 2000s Lars Petrus developed a concept of using a combination of two short algorithms from a single small algorithm pool to solve ZBLL without having to learn all 493 cases (Petrus 270). In 2012, Thom Barlow applied this concept to 1LLL (Duplex). In 2020 I expanded upon this with the concept of unions and the ability to solve to any desired end state. Any number of algorithms, not just two, can be joined in a union to solve a current state to any other desired state. This also solved the problem of how to solve or affect any random piece or set of pieces from a larger group rather than having to always solve a specific piece or set of pieces. An example of this, and where it was first developed, is CLL+1. Unions can be used to solve any U layer edge along with the corners and greatly reduce the case count, rather than having to learn the algorithms for all corner cases multiplied by the eight possible positions and orientations of the specific edge that is used every time.