The New Method / Substep / Concept Idea Thread

[supercavitation]

How are you calculating case count?

I get 29 ell * 6 cp + 21 pll = 195

I've generated algs for every case before, and I had 156 algs. Your math ignores symmetry.

 

[gyroninja]

I've generated algs for every case before, and I had 156 algs. Your math ignores symmetry.

I was counting on ell to handle the symmetry problems but I guess it wouldn't eliminate the symmetry on like dot cases.

 

[Aiminer357]

OLL skip method

Again, I don't know if anyone came up with this method yet. Tell me if somebody came up with this. Thanks

This method is recomended for those who use the CFOP method. People who pre-orient the Last Layer edges are also welcome.

I have been doing some research on different ways to solve the cube. One of the method is the Petrus Method. It creates a 2x2x3 block, orients the edges, finishes the F2L and then the Last Layer.

For someone who has been using CFOP for a long time, block building is too hard for me. So I came up with a method that will orient last layer while inserting the Last F2L pair and you don't need to learn hundreds of algorithms. You only need to learn a minimum of 27 algorithms and a bit of intuitive moving. I say minimum because the 27 cases are the most likely to happen. You could learn more.

Step 1: Build your cross and your F2L minus one pair. Make sure the last slot is on your FR or FL, depending on your turning style

Step 2: Find all the misoriented edges. You can only have an even amount of "bad edges"

Step 3: Orient the bad edges. I orient the edges the same method as the Petrus Method. There will be a link. http://lar5.com/cube/fas3.html

Step 4: Use Summer variation to orient the corners and insert the last F2L. Don't learn the mirrors unless you want to go advance. You can also learn Winter variation.

Step 5: PLL.

Pros: OLL skip! Can't think of anything else
Cons: Extra look during edge orientation and Summer.

I don't know if this is an effective way to solve the cube. I'm still sub 40. But tell me if it is effective. Or it's just useless and we all wasted our time.

 

[Sajwo]

So basically you found this method on the internet and then said that you invented it?

 

[shadowslice e]

Again, I don't know if anyone came up with this method yet. Tell me if somebody came up with this. Thanks

This method is recomended for those who use the CFOP method. People who pre-orient the Last Layer edges are also welcome.

I have been doing some research on different ways to solve the cube. One of the method is the Petrus Method. It creates a 2x2x3 block, orients the edges, finishes the F2L and then the Last Layer.

For someone who has been using CFOP for a long time, block building is too hard for me. So I came up with a method that will orient last layer while inserting the Last F2L pair and you don't need to learn hundreds of algorithms. You only need to learn a minimum of 27 algorithms and a bit of intuitive moving. I say minimum because the 27 cases are the most likely to happen. You could learn more.

Step 1: Build your cross and your F2L minus one pair. Make sure the last slot is on your FR or FL, depending on your turning style

Step 2: Find all the misoriented edges. You can only have an even amount of "bad edges"

Step 3: Orient the bad edges. I orient the edges the same method as the Petrus Method. There will be a link. http://lar5.com/cube/fas3.html

Step 4: Use Summer variation to orient the corners and insert the last F2L. Don't learn the mirrors unless you want to go advance. You can also learn Winter variation.

Step 5: PLL.

Pros: OLL skip! Can't think of anything else
Cons: Extra look during edge orientation and Summer.

I don't know if this is an effective way to solve the cube. I'm still sub 40. But tell me if it is effective. Or it's just useless and we all wasted our time.

Err... What's the difference between this and normal WV/SV?

 

[2180161]

Err... What's the difference between this and normal WV/SV?

There isn't one.

 

[Berd]

There isn't one.

I swear SV doesn't preserve EO?

 

[supercavitation]

I swear SV doesn't preserve EO?

SV is RUR' inserts, WV is RU'R' inserts.

 

[crafto22]

Hey guys, I thought of this a while back and realized it could work.

Steps:
1. Solve F2L-1 with any method.
Moves: 24
2. Solve CP with the last slot. I prefer solve the pair and then use one of only six algs to solve CP.
Moves: ~8
3. Solve OLL using a pseudo 2-gen alg which will preserve CP.
Moves: ~12?
4. Solve EPLL.
Moves: 6

Total moves: ~50
Still better than CFOP for moves.

So basically you just need to re-learn OLL with these pseudo 2-gen algs I will be generating.

 

[Jbacboy]

Hey guys, I thought of this a while back and realized it could work.

I am a nub, but I feel like this has been done before. (I just cant remember by who)

 

[2180161]

Hey guys, I thought of this a while back and realized it could work.

Steps:
1. Solve F2L-1 with any method.
Moves: 24
2. Solve CP with the last slot. I prefer solve the pair and then use one of only six algs to solve CP.
Moves: ~8
3. Solve OLL using a pseudo 2-gen alg which will preserve CP.
Moves: ~12?
4. Solve EPLL.
Moves: 6

Total moves: ~50
Still better than CFOP for moves.

So basically you just need to re-learn OLL with these pseudo 2-gen algs I will be generating.

Where are you getting 6 moves for EPLL? 10 for H, 9 for U, 12 for Z

 

[crafto22]

Get your math right:
7 for U and H, 8 for Z and 0 for solved averages out to 6.75

 

[2180161]

Get your math right:
7 for U and H, 8 for Z and 0 for solved averages out to 6.75

My apologies. I forgot about 0

 

[Torch]

Get your math right:
7 for U and H, 8 for Z and 0 for solved averages out to 6.75

Z is only 8 if you use the one with E moves, you have to take into account pre and post alg AUFs, and 6.75 moves simply cannot be rounded down to 6 moves no matter how much you try to convince yourself.

EPLL:
Pre alg AUF: 3/4 chance of AUF, 0.75 moves
Alg: 2/3 chance of U perm (7 moves), 1/6 chance of Z perm (9 moves), 1/12 chance of H perm (7 moves), 1/12 chance of solved (0 moves), total 6.75 moves
Post alg AUF: 3/4 chance of AUF, 0.75 moves

8.25 moves total
When EPLL skip, you can combine the AUFs into one move meaning you can lose 0.75 moves 1/12 of the time

8.25 moves-0.0625 moves=8.1875 moves

(Someone please correct me if I'm wrong)

 

[xyzzy]

EPLL:
Pre alg AUF: 3/4 chance of AUF, 0.75 moves
Alg: 2/3 chance of U perm (7 moves), 1/6 chance of Z perm (9 moves), 1/12 chance of H perm (7 moves), 1/12 chance of solved (0 moves), total 6.75 moves
Post alg AUF: 3/4 chance of AUF, 0.75 moves

8.25 moves total
When EPLL skip, you can combine the AUFs into one move meaning you can lose 0.75 moves 1/12 of the time

8.25 moves-0.0625 moves=8.1875 moves

(Someone please correct me if I'm wrong)

H perm can also have AUFs combined, and Z perm can have AUFs combined half the time. Being able to execute an EPLL from multiple angles is probably common (e.g. M2 U M' U2 M U M2 and M2 U M U2 M' U M2), which would reduce the probability of having to do a pre-AUF, so the exact average move count for EPLL depends on the number of algs you use for the different AUF cases. This differs from person to person so you can't really make a blanket statement like "EPLL is 8.xxx moves on average for everyone".

So basically you just need to re-learn OLL with these pseudo 2-gen algs I will be generating.

You can also use SuneOLL for this.

 

[Torch]

H perm can also have AUFs combined, and Z perm can have AUFs combined half the time. Being able to execute an EPLL from multiple angles is probably common (e.g. M2 U M' U2 M U M2 and M2 U M U2 M' U M2), which would reduce the probability of having to do a pre-AUF, so the exact average move count for EPLL depends on the number of algs you use for the different AUF cases. This differs from person to person so you can't really make a blanket statement like "EPLL is 8.xxx moves on average for everyone".



You can also use SuneOLL for this.

Yeah, I knew I was probably forgetting some things. Either way, we can agree that 6 moves is an incredibly ridiculous estimate, right?

For some data, I did 50 EPLLs (I used the ZBLL scrambler and then did COLL and counted the moves from then until solved) and got an average of 7.76 moves.

 

[crafto22]

Yeah, I knew I was probably forgetting some things. Either way, we can agree that 6 moves is an incredibly ridiculous estimate, right?

For some data, I did 50 EPLLs (I used the ZBLL scrambler and then did COLL and counted the moves from then until solved) and got an average of 7.76 moves.

An experimental average doesn't really matter. Either way I'm basing this off of the quite extensive math done by someone else. I don't know where it is in this thread, but someone managed to prove that the average movecount for EPLL is in fact 6.75 moves. One way or another, the method I proposed would still have a movecount of 50 moves with this added move.

 

[shadowslice e]

An experimental average doesn't really matter. Either way I'm basing this off of the quite extensive math done by someone else. I don't know where it is in this thread, but someone managed to prove that the average movecount for EPLL is in fact 6.75 moves. One way or another, the method I proposed would still have a movecount of 50 moves with this added move.

No, because the movecount you included does not have AUFS so the movecount is more like 52-53. Still passable but definitely not 50.

The way they showed that it was a 6.75 (which does not round down to 6 in any case) is by only taking the length of the algs and averaging them with their chance of occuring and not the AUF.

In addition, I don't particularly like including skip cases because they happen so rarely but can move the average movecount down significantly for a small alg set (of which EPLL is one) to a number which will almost never happen in a comp but rather only when doing big averages at home.

 

[supercavitation]

An experimental average doesn't really matter. Either way I'm basing this off of the quite extensive math done by someone else. I don't know where it is in this thread, but someone managed to prove that the average movecount for EPLL is in fact 6.75 moves. One way or another, the method I proposed would still have a movecount of 50 moves with this added move.

That was me, and the 6.75 moves was without accounting for AUF. Those were accounted for elsewhere in the analysis of that method. His move count is correct if you're AUFing before and after performing the EPLL rather than rotating.

 

[PenguinsDontFly]

Again, I don't know if anyone came up with this method yet. Tell me if somebody came up with this. Thanks

This method is recomended for those who use the CFOP method. People who pre-orient the Last Layer edges are also welcome.

I have been doing some research on different ways to solve the cube. One of the method is the Petrus Method. It creates a 2x2x3 block, orients the edges, finishes the F2L and then the Last Layer.

For someone who has been using CFOP for a long time, block building is too hard for me. So I came up with a method that will orient last layer while inserting the Last F2L pair and you don't need to learn hundreds of algorithms. You only need to learn a minimum of 27 algorithms and a bit of intuitive moving. I say minimum because the 27 cases are the most likely to happen. You could learn more.

Step 1: Build your cross and your F2L minus one pair. Make sure the last slot is on your FR or FL, depending on your turning style

Step 2: Find all the misoriented edges. You can only have an even amount of "bad edges"

Step 3: Orient the bad edges. I orient the edges the same method as the Petrus Method. There will be a link. http://lar5.com/cube/fas3.html

Step 4: Use Summer variation to orient the corners and insert the last F2L. Don't learn the mirrors unless you want to go advance. You can also learn Winter variation.

Step 5: PLL.

Pros: OLL skip! Can't think of anything else
Cons: Extra look during edge orientation and Summer.

I don't know if this is an effective way to solve the cube. I'm still sub 40. But tell me if it is effective. Or it's just useless and we all wasted our time.

or you could just learn VLS