The Meyer Thread (4x4 Reduction Sub-Method)

[shadowslice e]

This method is amazing! Previously I used Hoya, but the more I use Meyer the better I like it.
I also got my new 4x4 and outperformed all my previous 4x4 solves, doing an average of 1:17.34 (with a 1:07.93 single best). I was around 1:20 - 1:40 previously.
So, thanks!

Great to have another convert! Let's prove that this method can and will be fast

 

[Berd]

Has anyone tested this for 5x5 and above? My 4x4 is broken right now

 

[TDM]

Has anyone tested this for 5x5 and above? My 4x4 is broken right now

Edge pairing is horrible and 3x3 stage is also very bad. This method is great for 4x4 but I can't see it being useful for bigger cubes.

Edit: just thought of how to do edge pairing better (don't use freeslice!), will do another solve which will hopefully not be 3 minutes.

E2: 3:07 down from 3:16, so no

 

[WayneMigraine]

Edge pairing is horrible and 3x3 stage is also very bad. This method is great for 4x4 but I can't see it being useful for bigger cubes.

Edit: just thought of how to do edge pairing better (don't use freeslice!), will do another solve which will hopefully not be 3 minutes.

E2: 3:07 down from 3:16, so no

I think Richard Meyer uses something like Meyer on 5x5? I found a video here where I think he builds a Roux block, then does centers and edge pairing, then finishes the solve.

 

[shadowslice e]

Has anyone tested this for 5x5 and above? My 4x4 is broken right now

I reckon the problem is in the roux 333 stage because m-slices get too hard to do efficiently on the bigger cubes :-/

The redux stage is still nice and ergonomic (more so than Yau5 or freeslice I think)

 

[PenguinsDontFly]

I reckon the problem is in the roux 333 stage because m-slices get too hard to do efficiently on the bigger cubes :-/

The redux stage is still nice and ergonomic (more so than Yau5 or freeslice I think)

ross can M slice big cubes.

 

[WayneMigraine]

ross can M slice big cubes.

Kaijun lin also does it very well (as is seen here)

 

[shadowslice e]

ross can M slice big cubes.

Well, so can I but it's still nowhere near as fast as non slice moves.

 

[Berd]

I can just about do slices on 5x5 but not any higher.

 

[tukra]

I'm playing around with shadowslice Meyer. I like that more of the solve is direct. I'm wondering how people like to pair edges for this. I'm doing 1-3-3 but is there something better? Also what's a good way to deal with 2 edges left? What I'm doing is pretty slow (r l D2 l' U2 l D2 l' U2 r' and variants). I don't see any way to do the flipping thing reduction first solvers do without disturbing the solved blocks.

 

[PenguinsDontFly]

so can anyone generate some cmll + oll parity algs? im a lazy ***.

 

[tukra]

Found 2 pair edge flip works between UF and UB w/o messing up FB and 2/3SB.
x' z' Uw' R U R' F R' F' R u

 

[shadowslice e]

I'm playing around with shadowslice Meyer. I like that more of the solve is direct. I'm wondering how people like to pair edges for this. I'm doing 1-3-3 but is there something better? Also what's a good way to deal with 2 edges left? What I'm doing is pretty slow (r l D2 l' U2 l D2 l' U2 r' and variants). I don't see any way to do the flipping thing reduction first solvers do without disturbing the solved blocks.

well I pretty much do 1-3-3 as well to pair edges. For 2 edges left I do l U' F R' U' F' l' U' F' or the mirror

 

[dboeren]

I'm new to 4x4s, what is 1-3-3? I tried looking on the wiki but could't find this notation. I'm using Roux for 3x3 (and the 3x3 stage of my 4x4) so Meyer seems a natural thing to check out.

 

[PenguinsDontFly]

I'm new to 4x4s, what is 1-3-3? I tried looking on the wiki but could't find this notation. I'm using Roux for 3x3 (and the 3x3 stage of my 4x4) so Meyer seems a natural thing to check out.

if you want to use shadowslice meyer, you must use 1-3-3. You pair 1 edge, then 3, and 3 again. if you want to use normal meyer, you can use 3-2-3, which is much faster. also, there are many 3-2-3 tutorials and stuff.

 

[dboeren]

Thanks, I'll check out those tutorials. I expect it's probably going to be easier to start with vanilla Meyer (which will be more documented) before trying the shadowslice variant.

 

[tukra]

I'm new to 4x4s, what is 1-3-3?

I learned pairing from Yau videos since the process is basically the same. Cyoubx has a good one. In Meyer you're pairing on the M slice with your FB on the left vs Yau where you pair on the E slice with the cross on the bottom. So it's just a different orientation.

For regular Meyer you have 3 edges paired from the FB. Make one more pair naively and put it at UL. Now there are 8 left and you pair just like Yau. You can use 3-2-3 or 6-2 pairing. See videos.

For shadowslice you have 5 edges complete leaving 7 unpaired. I pair one edge, put it at UL. Then 3-3 just like the first and last sets of Yau 3-2-3.

 

[tukra]

For 2 edges left I do l U' F R' U' F' l' U' F' or the mirror

I can't figure out how to get that to work. alg.cubing

 

[shadowslice e]

I can't figure out how to get that to work. alg.cubing

it pairs edges in UF and UB but they must be in the same orientation (eg, both white up)

Or you could just look up the algs from Stadler which are probably more efficient for this case.

Also, more documented? I proposed the variant a whole month ago! Meyer only proposed his in 2008. ;-P

 

[tukra]

it pairs edges in UF and UB but they must be in the same orientation

Sorry if I'm being daft but when I put that alg into alg.cubing.net it doesn't look like the right thing. It changes 3 pairs, leaves some centers split and doesn't preserve the roux blocks. Maybe I made a mistake in transferring the notation but I can't tell what's wrong. Thanks for any help.

See: https://alg.cubing.net/