Advanced techniques for Roux

Revision as of 17:43, 6 December 2016 by Shiv3r (talk | contribs) (I added a CMLL prediction link, so you guys can predict CMLL's easier.)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Blockbuilding

There are a couple of techniques which come to mind for improving you blockbuilding and influencing later steps and such.

Influencing the second block

Influencing the second block during the first block may not sound very specific or important, but it's crucial to get your times as low as possible. For example, if you find a second block pair during first block, just save it on the R face for when you finish First block.

Making the second block not too inefficient

The second block in Roux is usually where we all get frustrated with high movecount, and feel there's not much to do about it- some cases for the second block are really annoying. You can try to apply Multi-slotting to reduce cases. There also are many cases where it may look like it'll take many moves, but it won't. Here are some cases which may not be immediately apparent in solving.

Corners

Using a corner prediction system like Shiv3r's corner prediction can help you out a lot, because It decreases the time you spend recognizing cases. KCLL is what comes to mind here. It is basically influencing edge orientation while solving the U corners. Notice it's not necessarily solving EO, but influencing it. Sometimes solving EO may be more trouble than just influencing, say, a 4 flip then simply doing M'UM'. This is why it helps to know multiple algs for each case.

You can change EO by just inserting a slice move before an alg. A very obvious case is if you got a Sune case with UF and UR flipped. If you insert the M' before it, to make r U R' U R U2 r', you flip the edges, with literally no extra moves. This comes in handy as some EO cases take many moves to execute, like 2 flips or the 6-flip.

OLLCP can be applied here too. And if DF and DB are oriented you can solve Eo easily during CMLL.

LSE

Non-matching centres: allowing free centre orienation from the beginning of LSE

You were always told to orient your centres first before doing EO. But sometimes, it's useful NOT to, or to make centres free. You see, an edge that is not an L/R edge is oriented if it has the same orientation as the centres. Whilst the L/R edges will be oriented if you have your U/D colour on top/bottom. So this leads to some fun...

Non-matching centres: misorienting centres to ease EO

Here it's different. You orient your centres beforehand but then use the property to your advantage to ease EO. Just look at these tasty moos:


Visualcube.png Orientation for this is just M, how delightful!

Orientation for this is just M'/M.