When did pre-orient become the basis for every new method? People seem to think that 2gen pairs are the greatest thing ever, when most cases are either already 2gen after rotating, or just much nicer 3gen (R U' R' (U+d) R' U' R remains my definitive example)
I also don't get what is so great about WV. Most of the cases seem like a lot of moves for not much work. People seem so opposed to doing regular OLL these days and I don't get why.
In any case, if you insist on combining ZZ and Roux, here's another idea I've had not long after having the one you mentioned:
EOLine as usual. Only this time, the Line in EOLine is DR and DL. Then Fridrich those 4 slots. Use (slightly more advanced) phasing during the last slot to permute 2 opposite edges. Then a 1-look last step.
It seems like you would be wasting the formation of the line though. With normal ZZ getting the other 2 D-layer edges in place is really easy, it seems needlessly complicated here.
I think doing EO first would make the first 1x2x3 much slower than normal. and regular EO for roux only take about 1.5 seconds anyway (faster than ZZ EO).
Major high-five.
Time to split this up and take it apart.
Yu-Korick Method (YK)
Descrition 1.0
STEP 1. EO
This step orients all edges on the cube. The definition of oriented is that the edge can be moved to the correct location without using quarter turns from F or B (F2 and B2 are OK)
During inspection of the cube, try to determine which pieces are oriented. The main method for this is to do this.
If 2 edges are misoriented.
Place 1 in the F or B layers, make a F or B turn, now 4 edges are misoriented.
I'm pretty sure setting up and doing F R F' etc. will always be a better way of orienting 2 edges.
STEP 2. 1x2x3 on the Left Side
This is basically the first step of the Roux method. Nothing is solved, but we have all edges oriented. The idea is to create a 1x2x3 brick without ever using quarter turns from F or B.
A warning! If you do M or M', the F or B colors have now become the U and D layers, so you can't turn those quarter turns. For example, M U M' will disturb orientation, and M U2 M' will preserve it.
A good idea is to use the R slice as a working area, where you can pair up pieces and then use the U layer to transport them to the L slice. Remember never to use quarter turns from F or B.
Move-restricted blockbuilding kind of sucks. The advantage of blockbuilding vs CE pairs is the lower move count of the blockbuilding. Using R as a "working area" to make pairs also defeats the purpose of blockbuilding, although its a result of pre-orienting.
STEP 3. Line
This solves DF and DB. It should be done intuitively, and without messing up edge orientation.
Some hints:
If the edge piece is in the U layer, put it on top of where it needs to go.
For DF, now do M' U2 M.
For DB, now do M U2 M'.
If it's in the R layer, do a R layer turn to bring it to U and do a U insertion.
Now we come to the major downfall of this method. Why would you split up the EO and the line? Its like making your cross with 2 flipped edges and an adjacent swap! You're taking EOline, replacing it with a step that takes virtually the same number of moves but does half the work, then going back later and finishing with more pointless moves. Not to mention the extra moves caused by having pieces from your first block located in DF/DB.
STEP 4a. Finish F2L
This is basically the Petrus F2L step. Using R and U turns only, finish the entire F2L. This can be done by blockbuilding, or by a simple Fridrich system where you add a cross piece and then do Fridrich F2L to solve. This is not the only option, you can also do a method that will guarantee an OLL skip, see STEP 4b.
STEP 4b. Finish F2L
This is a different way of doing step 4a. Use blockbuilding methods to solve a 2x2x1 brick with the DBR corner, RD and RB corners. Then, use WV (Winter Variation) to solve the last two pieces in the F2L and orient all corners at the same time, therefore creating an OLL skip (since you preserved Edge Orientation).
Most of what I said about step 2 applies here as well.
I also don't know why everyone thinks WV is so great, CLS would probably be better since your edges are already oriented.
STEP 5a. OLL/PLL
This is probably the easiest method to use. If you know 2-look OLL with Fridrich, you know 1-Look OLL for YK since edges are already oriented. If you know full PLL as well, you have a 2 look LL. If you know 2-look PLL, you have a 3-look LL.
STEP 5b. PLL
This applies only if you used step 4b, and not 4a. You now have an OLL skip, so if you learn PLL, you have a 1-look LL.
STEP 5c. ZBLL
This is the hardest method to learn, because of the amount of algorithms. Solve with Step 4a, and then do ZBLL to give you a 1-look LL. It's much easier to use WV instead for a 1-look LL.
For 5b and 5c, you may have a 1-look LL, but you have a 2-look Last Slot.
Okay, what you just outlined is nothing more than a broken form of ZZ. You waste moves and time by not combining steps 1 and 3, and step 2 suffers as a result. There's really nothing original here, and I definitely believe blah came up with this exact thing when he went through his phase where he played around with new methods.
What Stachu described is a bit more efficient, I would describe it as pre-orient Roux with WV+CPLL instead of CMLL. However, the blockbuilding still suffers from the pre-orient, and I think COLL would be better than WV+CPLL.
I don't think either method really stacks up to its unaltered parent, and don't see any clear advantages over any method being used widely right now.
