Welcome to the Speedsolving.com, home of the web's largest puzzle community! You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

What do you guys think about a collection of cubing resources for intuitive solvers? I've found that there is a dearth of resources for intuitive solving. Here's what I was thinking of:

Links to helpful intuitive resources that already exist, e.g. Heise method, commutator tutorials, and intuitive guides for other puzzles.

Video tutorial of Heise for beginners (this was my first method). There is no good Heise tutorial video.

The intuitive and semi-intuitive CMLL methods that I use. One uses only commutators and one uses 2 commutators and 2 algorithms (both of which are technically commutators).

Explanation of odd and even parity, and why it matters when using commutators. Suggestion for learning how to convert odd to even parity.

My completely intuitive 4x4 and NxN method which uses no algorithms.

My intuitive and semi-intuitive 3BLD parity methods, which either use no algorithms at all, or just Niklas and Sune.

Explanation of why I don't learn algorithms (because I want my skills to last for the rest of my life, and I will forget algorithms when I take a few years off from cubing).

It took me some time to find, understand, or create all of these things, since the vast majority of cubing resources are based on memorized algorithms. I think it would be helpful to have these in one place, which would also help intuitive solvers to exchange information with each other (assuming there are other intuitive solvers besides me).

What do you guys think about a collection of cubing resources for intuitive solvers? I've found that there is a dearth of resources for intuitive solving. Here's what I was thinking of:

Links to helpful intuitive resources that already exist, e.g. Heise method, commutator tutorials, and intuitive guides for other puzzles.

Video tutorial of Heise for beginners (this was my first method). There is no good Heise tutorial video.

The intuitive and semi-intuitive CMLL methods that I use. One uses only commutators and one using 2 commutators and 2 algorithms (both of which are technically commutators).

Explanation of odd and even parity, and why it matters when using commutators. Suggestion for learning how to convert odd to even parity.

My completely intuitive 4x4 and NxN method which uses no algorithms.

My intuitive and semi-intuitive 3BLD parity methods, which either use no algorithms at all, or just Niklas and Sune.

Explanation of why I don't learn algorithms (because I want my skills to last for the rest of my life, and I will forget algorithms when I take a few years off from cubing).

It took me some time to find, understand, or create all of these things, since the vast majority of cubing resources and based on memorized algorithms. I think it would be helpful to have these in one place, which would also help intuitive solvers to exchange information with each other (assuming there are other intuitive solvers besides me).

That sounds great! I'm not really an intuitive solver, though I do enjoy doing it, and it would be great to have a bunch of resources in one place. it could also help new cubers with more advanced cubing topics if they are all in one place, so they have an easy time finding info, and don't shy away from intuitive methods.

I also consider myself an intuitive solver, and for similar reasons as you mentioned (don't want to lose my skills after a long break). and yes, I had a hard time finding resources too.

I have a ZZ-Spike variation for petrus solvers, @GenTheThief i need your permission to make this a thing, but here’s the idea:
For uses of demonstration, build first layer in white.
1. Do Cross and F2L or make a Petrus block and expand all through the bottom layer.
2. Make a ballint block on the lime green face.
3. Solve the pink side and the orange side next to it (but make sure your not purposefully solving light blue and cream pieces)
4. Solve the grey green, grey pink, and grey orange pieces
5. Do EO like normal ZZ-Spike (which can be found Here)
6. CO
7. EP
8. CP

I often wonder why ZZ users don't incorporate non-matching blocks into their solves. Unlike Roux, there isn't much of a penalty for doing so. In Roux, CMLL recognition is different or you use the NMCLL recognition system. EO is also difficult to recognize. But in ZZ, it's not as difficult. There's NMLL, a last layer method designed specifically for this. Or users can learn tricks for recognizing OLL/PLL. I'm going to show a complete polar system for ZZ. This includes the method itself, the free blockbuilding, and the LL method. The blockbuilding can be taken to various depths, from normal blocks to non-matching blocks to blocks with varying pairs and pieces. As the solver learns these, they can add additional algorithms to the LL method.

Spoiler: Polar ZZ

Non-Matching Blocks

Scramble: F' D2 F2 D2 R F2 L D2 F2 D2 B2 L R' D' L' F' R F' L B D'
EOLine: L B' D2 L' R' F L D
1x2x2: L U2 R U' R U2
1x2x2: L U2 R' U2
Pair: R' U R' U' R U2 R'
Pair: U2 L U L' U' L U2 L'
Separation: U2 R' U' R U' R' U2 R
Permutation: U2 F R2 U' L' U R2 U' L U F' (U r')

Any Pair

Scramble: R2 F' R2 D F2 U F2 D2 B2 U R2 U2 B L F2 D' U' L2 R' D2
EOLine: F U2 L R F L' D'
1x2x3: L' R' U' L' U' L R' U2 R U L
1x2x3: R' U R' U R'
Separation: U2 R U2 R2 U' R2 U' R2 U2 R
Permutation: L' U2 L R U2 R' U2 R2

It can be taken further with pairs from the opposite side in the 1x2x3s, just 1x1x3s, no pairs at all and just pieces, and so on. Of course the deeper you go, the more algorithms for the LL permutation step. This provides a lot of freedom for how you want to blockbuild and gives choices for how many algorithms you want to learn.

Spoiler: Polar Roux

Scramble: L B' R' F L U2 F' U' L2 B2 R2 U' L2 D' L2 D2 B2 F'
FB: y2 x' F U' L' U x L' U M' F'
1x2x2: U r U' R
Last Pair + Polar Variation: U R U2 R' U2 r U' R'
EO+ Edge Separation: U2 M U' M2
Permutation: F' U' L' U2 L U F U L' U2 L U2 L'

This can also be applied to Petrus, even starting with a non-matching 2x2x2 and progressing to a non-matching Step 4 1x2x3. It would also work in Heise for even more insanity.

Maybe it's one of those things that takes time. 8-10 years ago I developed what's now called EOLR and was heavily promoting it because it seemed greatly beneficial. But most Roux users back then weren't very interested and only used a few cases. Now every Roux user sees it as a necessity. Maybe the future of ZZ is this freedom of blockbuilding.

I've just had an idea. It's to do with NM blocks and LL. Firstly, EOCross+1 suddenly becomes a lot easier. Sceondly, F2L becomes better. There are only 2 problems: the extra moves at the end (but this is made up by being easier to do cool stuff), secondly ZBLL and F2L recog. The solution for F2L is to practice, the solution to ZBLL is quite easy. Learn to recog NM PLL, then use the twist+PLL method of ZBLL recog. This could be amazing, but equally, too over complicated. I'll have a play around.

I've just had an idea. It's to do with NM blocks and LL. Firstly, EOCross+1 suddenly becomes a lot easier. Sceondly, F2L becomes better. There are only 2 problems: the extra moves at the end (but this is made up by being easier to do cool stuff), secondly ZBLL and F2L recog. The solution for F2L is to practice, the solution to ZBLL is quite easy. Learn to recog NM PLL, then use the twist+PLL method of ZBLL recog. This could be amazing, but equally, too over complicated. I'll have a play around.

Exactly. More options for the first edges, pair, 1x2x2, or anything. At the end it is only < .75 moves. If the final algorithm ends in an R or L move, there is a chance of cancellation. Cancellations can also be caused at the beginning or the middle of an algorithm.

The general idea is that you solve one piece at a time using 3 cycle algorithm.This is an advanced version of Old Pochmann(as it uses the BUFFER of OP)..It uses the concept of Orozco method and BH method.The memo should be done in pairs.There are three important terms used in this method.

Buffer (Corner: ULB Edge: UR)

Helper (Corner: UBR Edge: UL)

Target

The piece in Buffer/Helper is needed to shoot to the target.All the pieces need to move into the target by using setup move to solve.

Corners: ULB/UBR

Edges: UR/UL

Target: Target can be its original location(like BH Method).But for beginner Target for Corner is UFL or URF and Target for Edge is UF or UB.

The most important concept is CYCLE.There are two types of cycles

ODD CYCLE:

In odd cycle we shoot BUFFER to the TARGET Making sure that the HELPER ends in the BUFFER.The piece in the the TARGET will move to the HELPER.

BUFFER->TARGET->HELPER

EVEN CYCLE:

In even cycle we shoot HELPER to the TARGET Making sure that the BUFFER ends in the HELPER.The piece in the the TARGET will move to the BUFFER.

HELPER->TARGET->BUFFER

Note:The original piece in the HELPER is always switches with BUFFER.

Algorithms:

Note: Previously I had used UF as Helper for edges.This had slows down the setup moves very much.For this reason I had changed my Helper for edges to UL.For most of the pieces can setup to UB by doing B Layer move and UF by doing F Layer move.For Special Targets like DL/LD,DR/RD you can use special algorithms or use two move setups.

Edges Cycle(Normal – Odd Cycle)

UR->UF->UL: (M2 U') (M U2 M') (U' M2)

UR->UB->UL: (M2 U) (M' U2 M) (U M2)

UR->FU->UL: (M U M') U2 (M U M')

UR->BU->UL: (M' U' M) U2 (M' U' M)

UR->DL->UL: y U2 M U2 M' y'

UR->DR->UL : y M' U2 M U2 y'

UR->LD->UL: y U (M' U' M') U2 (M U' M) U' y'

UR->RD->UL : y U (M U M) U2 (M' U M') U' y'

Edges Cycle(Inverse – Even Cycle)

UL->UF->UR: (M2 U) (M U2 M') (U M2)

UL->UB->UR: (M2 U') (M' U2 M) (U' M2)

UL->FU->UR: (M U' M') U2 (M U' M')

UL->BU->UR: (M' U M) U2 (M' U M)

UR->DL->UL: y M U2 M' U2 y'

UR->DR->UL: y U2 M' U2 M y'

UR->LD->UL : y U (M' U M') U2 (M U M) U' y'

UR->RD->UL: y U (M U' M) U2 (M' U' M') U' y'

Corner Cycle(Normal – Odd Cycle)

ULB->URF->UBR: x R2 D2 (R U R') D2 (R U' R) x'

ULB->RFU->UBR: (F R F' L F) R2 (F' L' F R F')

ULB->FUR->UBR: (F' L F) R' (F' L' F) R

ULB->UFL->UBR: x (L U' L) D2 (L' U L) D2 L2 x'

ULB->FLU->UBR: L (F R' F') L' (F R F')

ULB->LUF->UBR: (F' L F R' F') L2 (F R F' L F)

Corner Cycle(Inverse – Even Cycle)

UBR->URF->ULB: x (R' U R') D2 (R U' R') D2 R2 x'

UBR->RFU->ULB: (F R' F' L F) R2 (F' L' F R' F')

UBR->FUR->ULB: R' (F' L F) R (F' L' F)

UBR->UFL->ULB: x L2 D2 (L' U' L) D2 (L' U L') x'

UBR->FLU->ULB: (F R' F') L (F R F') L'

UBR->LUF->ULB: (F' L' F R' F') L2 (F R F' L' F)

Parity:

Since For solving corner ULB and UBR are affected.And For Solving Edges UR and UL are affected.When you have a parity you must do F Perm in order to swap the corners(ULB and UBR) and edges(UR and UL).

F Prem: y R' U' F' R U R' U' R' F R2 U' R' U' R U R' U R y'

Alg for Flipping edges(UR and UL): M U M U M U2 M' U M' U M' U2

Alg for Twistwing corners(ULB+ and UBR-)

Notation: LBU->ULB and RUB->UBR

y (R U2 R' U' R U' R') (L' U2 L U L' U L) y'

y (R' D' R D R' D' R) U (R' D R D' R' D R) U' y'

Alg for Twisting corners(ULB- and UBR+)

Notation: BUL->ULB and BRU->UBR

y (L' U' L U' L' U2 L) (R U R' U R U2 R') y'

y (R' D R D' R' D R) U (R' D' R D R' D' R) U' y'

Notice:

Shoot the buffer to any unsolved location

When you see the helper piece during memo eg: UL add the piece UL and the piece that is in UL not LU.During Execution just skip the UL piece and continue to solve normally.

For advanced cuber you can shoot to the original location using BH commutators.

PS:If my english is bad,sorry for that.Also I request to contribute examples using 3 Cycle method..Inform me if some algs are incorrect..Try to contribute this 3C method to wiki...
I am making documentation I will release it when it is finnished.

It can be used to beginner BLD method solving..

This method is invented by Me(Md Shahil Ahmed) during Lockdown days of Corona Virus..

Credits:
1.Erno Rubiks
2.Inventor of Pochmann Method
3.Inventor of BH Method
4.Inventor of Orozco method
5.You

Since I completed the A2 method, I've thought a lot about what A3 would be. I'm sure there's a precise, technical answer, but I have a pretty good idea of the system. There are already many 3x3 methods and people are averaging 5-6 seconds with those. And we don't really need any more 3x3 methods. But I think this is useful in that the system can be applied to existing methods as I outlined a few posts before this one. Anyway, here is A3, or the extension of A2 to 3x3.

Passive Blockbuilding: Using normal, non-matching, and misoriented pieces, freely blockbuild what is necessary to reach an algorithm step in the method. What is built may be F2L, two 1x2x3s, or any other option. This step means being method neutral or applying this blockbuilding style to the currently used method. See the ZZ and Roux examples a few posts before this one.

In F2L for example, this could be pairs in various slots, cross pieces from a different layer, and other possibilities. In ZZ, the left and right side could consist of pairs put together in any way from either side or even not have complete pairs at all. The same for Roux.

Resolve: Use an LL, CLL, or any final algorithm and simultaneously correctly order the pieces from step 1. In Roux this would be CMLL and in CFOP this would be PLL. This step has more freedom than it may seem.

There are many chances for cancellations during performance of the algorithm.

The number of cases in large algorithm sets can be greatly reduced by taking advantage of the LL pieces and the misplaced pieces, and other techniques.

There is a freedom of progression. Users can start by learning simple types of misplaced pairs and the associated algorithms. Then they can continue to progress, learning more situations and the algorithms that fix them.

Remainder: If the method has remaining steps, as in Roux LSE, those can be completed now, or sometimes completed intuitively during the algorithm performed in step 2.

Spoiler: ZZ Example Images

Then after PLL, EPLL, NMLL, or ZBLL the F2L would be solved.

Spoiler: Roux Example Images

->CMLL-> ->CMLL-> ->CMLL->

People have made use of purposefully having a misoriented corner in F2L, there is the PEG method which places pairs in any F2L slot, and non-matching blocks exist in Heise and Roux. There is also freedom of blockbuilding in the FreeFOP method and freestyle. I think this can be taken further with non-matching and misorientation and implemented into speedsolves in a way that provides improvement. Of course a solver wouldn't use everything and freely build whatever they see. That's just too much effort for little or no gain. At first, only simple mismatched pairs and non-matching blocks would be great for speedsolves.

What do you think about this system? Obviously there are positives and negatives like any other. The difficulty would be in learning additional tricks for case recognition in the more advanced block types and no one is really method neutral. So for now the system is probably best applied to someone's currently used method. And obviously it requires learning more algorithms for each case, which means algorithm sets with few cases will be easiest. For standard Roux, it may not be worth it to go any deeper than simple non-matching blocks. ZBLL would be extremely difficult because that would mean learning a large number of cases for each type of mismatched pair. For other methods, it makes for a good fit because the second step in two-look LL methods often has very few cases. The second step of NMLL is only 15 algorithms and the recognition for both steps is a perfect match for this. COLL and EPLL may be good too because EPLL is only four algorithms.

I have a ZZ-Spike variation for petrus solvers, @GenTheThief i need your permission to make this a thing, but here’s the idea:
For uses of demonstration, build first layer in white.
1. Do Cross and F2L or make a Petrus block and expand all through the bottom layer.
2. Make a ballint block on the lime green face.
3. Solve the pink side and the orange side next to it (but make sure your not purposefully solving light blue and cream pieces)
4. Solve the grey green, grey pink, and grey orange pieces
5. Do EO like normal ZZ-Spike (which can be found Here)
6. CO
7. EP
8. CP

This sounds like the first iteration of spike that I proposed, and also how Oscar Roth Anderson solves, getting to the last two faces, doing EO and then solving them. I think that overall it's not a good idea to try and have a method halfway between two--I initially had EO on only 3 faces but shadowslice pointed out that it wasn't really worth it, so I expanded the EO step so that it would include another face and more pieces and be more worthwhile. Also, from a direct solving view point, I don't think that that's the best order either.

The general idea is that you solve one piece at a time using 3 cycle algorithm.This is an advanced version of Old Pochmann(as it uses the BUFFER of OP)..It uses the concept of Orozco method and BH method.The memo should be done in pairs.There are three important terms used in this method.

Buffer (Corner: ULB Edge: UR)

Helper (Corner: UBR Edge: UL)

Target

The piece in Buffer/Helper is needed to shoot to the target.All the pieces need to move into the target by using setup move to solve.

Corners: ULB/UBR

Edges: UR/UL

Target: Target can be its original location(like BH Method).But for beginner Target for Corner is UFL or URF and Target for Edge is UF or UB.

The most important concept is CYCLE.There are two types of cycles

ODD CYCLE:

In odd cycle we shoot BUFFER to the TARGET Making sure that the HELPER ends in the BUFFER.The piece in the the TARGET will move to the HELPER.

BUFFER->TARGET->HELPER

EVEN CYCLE:

In even cycle we shoot HELPER to the TARGET Making sure that the BUFFER ends in the HELPER.The piece in the the TARGET will move to the BUFFER.

HELPER->TARGET->BUFFER

Note:The original piece in the HELPER is always switches with BUFFER.

Algorithms:

Note: Previously I had used UF as Helper for edges.This had slows down the setup moves very much.For this reason I had changed my Helper for edges to UL.For most of the pieces can setup to UB by doing B Layer move and UF by doing F Layer move.For Special Targets like DL/LD,DR/RD you can use special algorithms or use two move setups.

Edges Cycle(Normal – Odd Cycle)

UR->UF->UL: (M2 U') (M U2 M') (U' M2)

UR->UB->UL: (M2 U) (M' U2 M) (U M2)

UR->FU->UL: (M U M') U2 (M U M')

UR->BU->UL: (M' U' M) U2 (M' U' M)

UR->DL->UL: y U2 M U2 M' y'

UR->DR->UL : y M' U2 M U2 y'

UR->LD->UL: y U (M' U' M') U2 (M U' M) U' y'

UR->RD->UL : y U (M U M) U2 (M' U M') U' y'

Edges Cycle(Inverse – Even Cycle)

UL->UF->UR: (M2 U) (M U2 M') (U M2)

UL->UB->UR: (M2 U') (M' U2 M) (U' M2)

UL->FU->UR: (M U' M') U2 (M U' M')

UL->BU->UR: (M' U M) U2 (M' U M)

UR->DL->UL: y M U2 M' U2 y'

UR->DR->UL: y U2 M' U2 M y'

UR->LD->UL : y U (M' U M') U2 (M U M) U' y'

UR->RD->UL: y U (M U' M) U2 (M' U' M') U' y'

Corner Cycle(Normal – Odd Cycle)

ULB->URF->UBR: x R2 D2 (R U R') D2 (R U' R) x'

ULB->RFU->UBR: (F R F' L F) R2 (F' L' F R F')

ULB->FUR->UBR: (F' L F) R' (F' L' F) R

ULB->UFL->UBR: x (L U' L) D2 (L' U L) D2 L2 x'

ULB->FLU->UBR: L (F R' F') L' (F R F')

ULB->LUF->UBR: (F' L F R' F') L2 (F R F' L F)

Corner Cycle(Inverse – Even Cycle)

UBR->URF->ULB: x (R' U R') D2 (R U' R') D2 R2 x'

UBR->RFU->ULB: (F R' F' L F) R2 (F' L' F R' F')

UBR->FUR->ULB: R' (F' L F) R (F' L' F)

UBR->UFL->ULB: x L2 D2 (L' U' L) D2 (L' U L') x'

UBR->FLU->ULB: (F R' F') L (F R F') L'

UBR->LUF->ULB: (F' L' F R' F') L2 (F R F' L' F)

Parity:

Since For solving corner ULB and UBR are affected.And For Solving Edges UR and UL are affected.When you have a parity you must do F Perm in order to swap the corners(ULB and UBR) and edges(UR and UL).

F Prem: y R' U' F' R U R' U' R' F R2 U' R' U' R U R' U R y'

Alg for Flipping edges(UR and UL): M U M U M U2 M' U M' U M' U2

Alg for Twistwing corners(ULB+ and UBR-)

Notation: LBU->ULB and RUB->UBR

y (R U2 R' U' R U' R') (L' U2 L U L' U L) y'

y (R' D' R D R' D' R) U (R' D R D' R' D R) U' y'

Alg for Twisting corners(ULB- and UBR+)

Notation: BUL->ULB and BRU->UBR

y (L' U' L U' L' U2 L) (R U R' U R U2 R') y'

y (R' D R D' R' D R) U (R' D' R D R' D' R) U' y'

Notice:

Shoot the buffer to any unsolved location

When you see the helper piece during memo eg: UL add the piece UL and the piece that is in UL not LU.During Execution just skip the UL piece and continue to solve normally.

For advanced cuber you can shoot to the original location using BH commutators.

PS:If my english is bad,sorry for that.Also I request to contribute examples using 3 Cycle method..Inform me if some algs are incorrect..Try to contribute this 3C method to wiki...
I am making documentation I will release it when it is finnished.

It can be used to beginner BLD method solving..

This method is invented by Me(Md Shahil Ahmed) during Lockdown days of Corona Virus..

Credits:
1.Erno Rubiks
2.Inventor of Pochmann Method
3.Inventor of BH Method
4.Inventor of Orozco method
5.You

Ok so there are 27 vls cases for winter variation. There are 3 2x2 pll cases. Y, T, and Solved. 27x3=81
if my math is correct the subset would have 81 cases

Ok so there are 27 vls cases for winter variation. There are 3 2x2 pll cases. Y, T, and Solved. 27x3=81
if my math is correct the subset would have 81 cases

So I designed a new Method called EOCF.
Edge Orientation Corners First. The hardest part is indeed step one and after it its just CF with some other algs.
Step 1 Edge Orientation
Step 2 Corners of the First Layer with R U L and CLL in the Style of R U L moves to not destroy EO (maybe i have to gen new algs for that)
Step 3. Edges of the first layer done in pairs
Step 4. Edges of the last layer (gen new algs or use U-Perm)
Step 5. Insert last redge or last ledge by U M U2 M U or U' M' U2 M' U' and midge permuation. It can be really fast to execute and has in the End not much algs maybe 42 (Step 2) + ~10 (Step 4) + 2 (Step 5) ~ 54 Algs. And the movecount is fairly good as EO takes around 6 STM, maybe 15 STM max for whole step 2 and step 3 and 4 can be combined together max 15 STM maybe Step 5 takes no more than 8 STM whole. So you get your cube in around 44 STM which is fairly good. And Fingertricks and no real need for rotations are also really noticeable.
So what do you think? Should I reasearch this method more?

So I designed a new Method called EOCF.
Edge Orientation Corners First. The hardest part is indeed step one and after it its just CF with some other algs.
Step 1 Edge Orientation
Step 2 Corners of the First Layer with R U L and CLL in the Style of R U L moves to not destroy EO (maybe i have to gen new algs for that)
Step 3. Edges of the first layer done in pairs
Step 4. Edges of the last layer (gen new algs or use U-Perm)
Step 5. Insert last redge or last ledge by U M U2 M U or U' M' U2 M' U' and midge permuation. It can be really fast to execute and has in the End not much algs maybe 42 (Step 2) + ~10 (Step 4) + 2 (Step 5) ~ 54 Algs. And the movecount is fairly good as EO takes around 6 STM, maybe 15 STM max for whole step 2 and step 3 and 4 can be combined together max 15 STM maybe Step 5 takes no more than 8 STM whole. So you get your cube in around 44 STM which is fairly good. And Fingertricks and no real need for rotations are also really noticeable.
So what do you think? Should I reasearch this method more?

Step 2 would not need any new algs genned, because you can use coll instead of totally new algs. Also, you can’t us u perms for step 4 because you still have to orient those edges before permuting them. What you could do is use ELL for this step, but ell isn’t the best subset. This method is a pretty interesting method, I would recommend you learn ell and coll and do some solves with it and see what you think.