Petrus is like, so dead now.

Do you use Petrus (or even a variation of it)


  • Total voters
    173
Personally, I find Pi and H to be the ZBLL sets with the hardest recognition and the worst algs. Forcing them every solve doesn't sound like a good time, but maybe you could make it work if you have better algs and a better recognition system than me.
I find the easiest way to learn recognition is just practice and let your subconscious do the work.
 
I use a variation of the Petrus corner method as 2-look CMLL in Roux (and any other time I need to solve corners). Permuting first enables 2-look CMLL with only 4 algorithms. Here's my prior post about it, with a flowchart:

 
Roux is actually very similar to Petrus, and I really see it as an improved version of Petrus. This is because the last layer in Petrus is the same as the last steps of Roux, but less convenient than Roux. In both methods, you solve the last layer’s corners first and then solve edges. But Roux gains an advantage by leaving 6 edges unsolved until the last step, rather than solving the DF and DB edges early (too early).

The Allan algorithm that the Petrus method uses to solve the last 4 edges is a conjugate of the same 4-move commutator that Roux uses to solve the last 4 edges. In Petrus, the DF and DB edges are already solved so we have to perform and undo setup moves. In Roux we avoid this by leaving the DF and DB edges unsolved.

So Roux and Petrus are really the same, except for the DF and DB edges. In Petrus you solve them early, which restricts your moves without any substantial benefit, and also requires you to break and restore them in the last step. So, just do Petrus but leave these 2 edges unsolved until the last step. This is Roux.
 
Back
Top