[Help Thread] 4x4 Discussion

[EddyEddy]

Hey Cubers, I have been stuck on the professor cube for a few weeks now. I have two edge pieces that need to be swapped. I can't find any solution. I know that I can solve it like the revenge and 3x3 If I could get past this part. If anyone can help I would greatly appreciate it.

 

[BenChristman1]

Can you post a picture?

 

[abunickabhi]

There are a few cases that you can get in 5x5 and 5x5+ cubes.

You can try to intuitively understand, how slice flip undo slice works, and then last 2 edges becomes intuitive. Otherwise you can lookup the optimal algs from kirjava meep or any other 5x5 expert's pdf.

 

[EddyEddy]

I need to swap the red/green edge with the orange/yellow edge piece. I figured out how to do that but then when re-orienting the centers, some of the edge pieces get messed up. Do I solve the edges first or the centers?

 

[Tabe]

Sarah's Cubing Site — 5x5x5 — Last 2 Edges

Speedcuber Sarah Strong's collection of Rubik's Cube algorithms.

sarah.cubing.net

 

[Kaiju_cube]

with the last 2 edges I usually end up flipping the edges and doing 4x4 pairing until I have 2 edge that have center pieces that need to be swapped (on odd number cubes), then I swap them,.. then usually I have to do the edge flip-parity algorithm on 1 edge to flip the wings.

Some people have every edge algorithm memorized but I found this works for me. I can always get the two edges into a solvable state this way.

Edge Flip (R, U, R', F, R', F', R)

Free-slice

4x4 pairing (Dd, R, F', U, R', F, Dd')

Center swap (R2', F2, U2, R2', U2, F2, R2) (*how deep the R move is varies with the cube size, sometimes it's 2 or 3 layers deep that you need to go to get all the the R pieces that need to be involved. Like on a 5x5 I'd go 2 layers deep in an R move because the 3rd layer from either side is the center. On a 7x7 I'd go 3 layers deep because the 4th layer is the center.)

Edge flip/Parity alg to flip 2 wings but not an entire edge piece (R', U2, L, F2, L', F2, F2, U2, F, U2, F', U2, F2, F2, F2) (*again how deep the R is varies with cube size)

maybe not the most elegant solutions but I'm not speedsolving I just like solving big cubes. I know a few more edge pairing algorithms but honestly these are the ones that work pretty much every time. Sometimes I have to play with the last 2 edges for a minute to get them into a solvable state, but it always works for me.

Good luck. Big cubes are a ton of fun and (at least in my experience) very addicting.

 

[Neural cuber 22]

watch jperms video

 

[EddyEddy]

Well, I can solve all the centers and all the edges. but which should I complete first?

 

[OreKehStrah]

Well, I can solve all the centers and all the edges. but which should I complete first?

Solve the centers first. Then build 8 edges using free slice, and put them on the U and D layers. Then restore the centers and do last 4 edges.

 

[EddyEddy]

Solve the centers first. Then build 8 edges using free slice, and put them on the U and D layers. Then restore the centers and do last 4 edges.

So, do the white edge pieces and the yellow edge pieces first?

 

[Tabe]

Do whatever you want for the first 8 - any colors.

 

[Waffles]

I have a competition in 3 weeks, and the cutoff for 4x4 is 1:10. As of right now, I average 1:15 and occasionally get sub 1:10.

If there’s any tips you guys have that can make me improve ~5 seconds on 4x4 quickly, could you tell me thanks

Also here is a multi-phase solve that I did for an example:

First 2 Centers: 4.01
3 Cross edges: 12.72
Last 4 centers; 26.73
Edge Pairing: 41.87
F2L: 48.22
OLL (parity): 56.41
PLL (parity): 1:04.96

 

[CodingCuber]

I have a competition in 3 weeks, and the cutoff for 4x4 is 1:10. As of right now, I average 1:15 and occasionally get sub 1:10.

If there’s any tips you guys have that can make me improve ~5 seconds on 4x4 quickly, could you tell me thanks

Also here is a multi-phase solve that I did for an example:

First 2 Centers: 4.01
3 Cross edges: 12.72
Last 4 centers; 26.73
Edge Pairing: 41.87
F2L: 48.22
OLL (parity): 56.41
PLL (parity): 1:04.96

I average sub 1:10 and what helped me get there was mostly just good edge pairing. You should be able to go into edge pairing straight away with little pause. Look for your next set of edges while doing one, etc. That's all I can really offer though. Good luck!

 

[xyzzy]

I have a competition in 3 weeks, and the cutoff for 4x4 is 1:10. As of right now, I average 1:15 and occasionally get sub 1:10.

If there’s any tips you guys have that can make me improve ~5 seconds on 4x4 quickly, could you tell me thanks

Also here is a multi-phase solve that I did for an example:

First 2 Centers: 4.01
3 Cross edges: 12.72
Last 4 centers; 26.73
Edge Pairing: 41.87
F2L: 48.22
OLL (parity): 56.41
PLL (parity): 1:04.96

If your OLL and PLL are taking 8 seconds each, you can surely cut a lot from there. 3 seconds OLL parity + 2 seconds OLL + 2 seconds PLL parity + 3 seconds PLL = 10 seconds.

Besides that, the other big time losses are from 3cross and L4C. Make a conscious effort to focus harder during those steps; that might help you look ahead better, cutting down pauses. (Cross is a really annoying step to get good at because the pieces can be almost anywhere, and the unsolved colourful mess in the visible part of the cube makes it harder to find the pieces.)

 

[Waffles]

I do want to point out that that solve had terrible edge pairing, I found a video of it and saw I had like an 8 second pause in the middle. Also my first centers+cross are normally a bit longer cause I have a pause after doing the first one (as far as I can plan with around 20 seconds of inspection).

My OLL Parity takes about 4 seconds to execute and I had a really bad awkward shape case in this solve. The pll was weird cause I started doing an A perm and realised it was J Perm Parity so had to retrace my steps. With all that I’m kind of confused how I actually managed to get sub 1:10 on that solve.

My f4c is usually okay, it seemed around normal in that solve and same thing with cross.

Generally I’m getting a sub 1:10 every 3 solves.

I thought I’d include another one just for references:

F2C: 4.35
Cross-1: 9.06
L4C: 15.30
LCP (last cross piece): 17.92
EP: 29.78
F2L: 37.72
OLL (parity): 44.41
PLL (parity): 51.00

What a solve lol

 

[Christopher Mowla]

My OLL Parity takes about 4 seconds to execute [. . .]

Maybe you need to use Waffle's parity alg instead of the one you're using? (Although off-topic, I show how that algorithm (and Lucas parity and the Redbull) algorithms work in 2010.)

 

[Waffles]

I use the Rw U2 x algorithm and only just learnt it after using the R2 B2 U2 L alg so that’s probably the reason.

 

[Waffles]

Also I can show you a solve that wasn’t super ridiculously lucky now:

F2C: 3.52
C-1: 11.05
L4C: 9.55
LCE: 4.35
EP: 19.97
F2L: 11.41
OLL: 8.19
PLL: 7.71

Total: 1:15.773

 

[GodCubing]

Hey, does anyone use OBLBL? I recently started learning it and was wondering what faster solvers do. I think I can blockbuild ok, but I think I need help on edge pairing and 3x3 stage

 

[xyzzy]

Hey, does anyone use OBLBL? I recently started learning it and was wondering what faster solvers do. I think I can blockbuild ok, but I think I need help on edge pairing and 3x3 stage

I tried it a few years ago. It's okay but doesn't seem ideal; you need to be able to distinguish the two wings (which ones belong in the second layer versus in the third layer), which might be especially difficult if you're at least dual CN.

Edge pairing should be pretty straightforward: slice to make pair, replace, slice back to make another pair. Basically the "2-3" in Yau's 3-2-3 pairing, repeated until all the edges are paired.