Random Blindfold Cubing Discussion

[Hssandwich]

Can you send a picture of it Roman?

 

[unsolved]

I meant the picture of the position Roman has that he is looking for the solution to at present.

 

[Roman]

I meant the picture of the position Roman has that he is looking for the solution to at present.

First of all, I was looking for a case when just two edges are flipped, but I guess it's impossible. Then I need an algorithm that flips one edge and changes minimum other elements, as long as it's not just a single 180deg turn...

 

[unsolved]

First of all, I was looking for a case when just two edges are flipped, but I guess it's impossible. Then I need an algorithm that flips one edge and changes minimum other elements, as long as it's not just a single 180deg turn...

For the time being, I am not sure how to code it. If it is something you are seriously interested in, give me some more details, and I'll give it a try.

 

[mark49152]

Can ksolve cope with curvy copter? Just curious, I don't intend to try it myself

 

[Daniel Lin]

First of all, I was looking for a case when just two edges are flipped, but I guess it's impossible. Then I need an algorithm that flips one edge and changes minimum other elements, as long as it's not just a single 180deg turn...

Left jumble right jumble F
flips one edge and swaps 2 corners

there must be an alg that flips two edges, but I cant find a good one

 

[Daniel Lin]

Can anyone figure out how it can be done?

This is not the most efficient way, but here you go
So do the setup move, flip the edge you want to flip, undo the setup move
when two or four edges are flipped, do an alg. I know it is possible to flip 2 edges

Piece notation
(speffz) Setup moves
UF(C) L R
DF(U) x' R L x' R L
DB(W) B x' R L
UB(A) x' R L B R L
UR(B) z' B R F z B
BR(T) z R F L x' L
DR(V) z' L F x' B R
FR(J) B z' B R z L F
UL(D) mirror of UR(B)
BL(R) mirror of BR(T)
DL(X) mirror of DR(V)
FL(L) mirror of FR(J)

 

[Daniel Lin]

seriously had the worst scramble of my life
D2 B' D2 B2 R2 F R2 B' L2 R2 B' D B F L' U F' L' D2 R' B2
everybody should try this scramble

I got 1:26.05 but it was a DNF by 2 edges

 

[CyanSandwich]

Any good algs for double parity in 4BLD?

By which I mean swapping UBL-UBR, and DF-UB

Edit: Actually, I might be switching to UFr buffer for wings.
So an alg for UBL-UBR and UFr->(whatever produces a good alg)?

 

[ottozing]

Any good algs for double parity in 4BLD?

By which I mean swapping UBL-UBR, and DF-UB

Edit: Actually, I might be switching to UFr buffer for wings.
So an alg for UBL-UBR and UFr->(whatever produces a good alg)?

Could always try L Rw2 B2 Rw' U2 Rw' U2 x' U2 Rw' U2 Rw U2 Rw' U2 Rw2 U2 x U2 L', although it does slightly different pieces

 

[CyanSandwich]

Could always try L Rw2 B2 Rw' U2 Rw' U2 x' U2 Rw' U2 Rw U2 Rw' U2 Rw2 U2 x U2 L', although it does slightly different pieces

Ah, OLL parity with a setup. My OLL parity alg is different, but I can swap UBL-BRU and UFr-UBl with an r setup.

 

[pinser]

I'm curious how it is for dem faster BLD solvers.
Do they DNF because of memo or execution error?
And what's their success rate?

 

[TheCoolMinxer]

I'm curious how it is for dem faster BLD solvers.
Do they DNF because of memo or execution error?
And what's their success rate?

I am not fast but I mostly DNF due to my memo. Also I tend to forget twisted corners/flipped edges because I memorise them visually...
My success rate is somewhere around 55-60%, and 1:15 avg

 

[Shiv3r]

I was bored, and generated some algs for a 3x3 BLD method, similar to Old pochmann but with simpler memorization(I think)--- Similar to memoing 3OP permutation, I think.(just 2 set of phone numbers, like 3OP)
so this method idea has two parts that can be taken seperately: Corners, and edges.
So for the edges, I had the Idea of using an H-permutation for swapping (instead of a T permutation like in OP). to memorize the pieces you remember piece and orientation(solved=x, flipped = xa) and you use either H perm or modified H perm that also flips 2 opposite edges to place the pieces. This method was originally based on an idea to swap 2 pairs of edges simultaneously using H permutation, so you get edges done twice as fast. But instead you can pick any of the 4 pieces on the U layer as a Buffer piece, and breaking into cycles are a little more simple. Think of it as a boomerang method but you solve orientation simultaneously as permutation.

and for the corners: instead of just the Y-perm, I came up with 2 more algorithms, one that solves the corners when they are twisted like a bowtie OCLL, and the other that does the same but the OCLL pattern is U2. So you just memorize the corner(1, 2, etc.) and the orientation(1a, 2b, 3, etc.)

any thoughts?

 

[the super cuber]

I'm curious how it is for dem faster BLD solvers.
Do they DNF because of memo or execution error?
And what's their success rate?

not sure if I come under 'fast' or not, but for me mostly when I dnf when I execute a wrong comm, overshoot, forget to memo a flipped edge, and make a memo error (not too common)

seriously had the worst scramble of my life
D2 B' D2 B2 R2 F R2 B' L2 R2 B' D B F L' U F' L' D2 R' B2
everybody should try this scramble

I got 1:26.05 but it was a DNF by 2 edges

51.43 success, memorised most corners visually

 

[Roman]

Any good algs for double parity in 4BLD?

By which I mean swapping UBL-UBR, and DF-UB

Edit: Actually, I might be switching to UFr buffer for wings.
So an alg for UBL-UBR and UFr->(whatever produces a good alg)?

You might wanna ask @unsolved for such things

 

[Cale S]

Has anyone tried doing something for 4BLD centers where you solve centers onto an incorrect face and then swap centers at the end? Could be useful if a side already is mostly a wrong color but you can't solve those pieces because you already oriented to solve other pieces. And what are some algs to swap two centers? (like all 4 on each face with all 4 on another face)

I also think things like r2 u2 r2 u2 could be useful

 

[obelisk477]

Has anyone tried doing something for 4BLD centers where you solve centers onto an incorrect face and then swap centers at the end? Could be useful if a side already is mostly a wrong color but you can't solve those pieces because you already oriented to solve other pieces. And what are some algs to swap two centers? (like all 4 on each face with all 4 on another face)

I also think things like r2 u2 r2 u2 could be useful

r2 U2 D2 r2 x y U2 m2 U2 m2 ?

 

[Cale S]

r2 U2 D2 r2 x y U2 m2 U2 m2 ?

Thanks, what about adjacent centers? I'm guessing there isn't a really great alg, which is probably the main factor in whether or not it would be worth it

 

[obelisk477]

Thanks, what about adjacent centers? I'm guessing there isn't a really great alg, which is probably the main factor in whether or not it would be worth it

You could do a setup to the opposite: y r U2 r2 F2 r y' (opp center swap) r' F2 r2 U2 r' . Obviously make sure you rotate back after the opp swap. It works, but its too late at night to figure out why undoing the setup doesnt need rotations