F2L question

[jms]

When practising F2L is there some kind of program that will generate scrambles with particular F2L cases to be solved?

For instance if I am trying to practise a particular algorithm, I am finding it hard work to get the cube into that state every time so I can keep practising that case.

I am finding it hard to learn any algs as I keep getting different F2L pair cases each time I scramble the cube.

 

[PenguinsDontFly]

When practising F2L is there some kind of program that will generate scrambles with particular F2L cases to be solved?

For instance if I am trying to practise a particular algorithm, I am finding it hard work to get the cube into that state every time so I can keep practising that case.

I am finding it hard to learn any algs as I keep getting different F2L pair cases each time I scramble the cube.

Do f2l intuitively. If you really want to learn algs, just do them over and over until you are 1 repetition away from solving the cube, look at the case, and do it.

 

[jms]

Do f2l intuitively. If you really want to learn algs, just do them over and over until you are 1 repetition away from solving the cube, look at the case, and do it.

I am learning it intuitively in the main, but for some of the awkward cases I am learning algs. The same issue applies though.

When I am trying to learn how to solve particular F2L pair cases, whether using intuition or an alg, I am looking for something which will generate scrambles of the case I am trying to learn.

Does that make sense? I'm conscious I am probably not explaining it very well.

 

[brian724080]

I am learning it intuitively in the main, but for some of the awkward cases I am learning algs. The same issue applies though.

When I am trying to learn how to solve particular F2L pair cases, whether using intuition or an alg, I am looking for something which will generate scrambles of the case I am trying to learn.

Does that make sense? I'm conscious I am probably not explaining it very well.

I see what you're saying, but in terms of memorizing the algorithm, it would be much more efficient for you to just execute it over and over again. Also, you should probably try and understand where the algorithms came from, since you don't want to memorize separate algorithms for every single cube orientation.

 

[PenguinsDontFly]

I am learning it intuitively in the main, but for some of the awkward cases I am learning algs. The same issue applies though.

When I am trying to learn how to solve particular F2L pair cases, whether using intuition or an alg, I am looking for something which will generate scrambles of the case I am trying to learn.

Does that make sense? I'm conscious I am probably not explaining it very well.

I do not know of any scramble generator that gibes you a specific f2l case, so just repeat the alg to learn it. However, to practise different cases, I think qq timer has f2l -1 pair scrambles. (last slot and ll are left)

 

[jms]

I see what you're saying, but in terms of memorizing the algorithm, it would be much more efficient for you to just execute it over and over again. Also, you should probably try and understand where the algorithms came from, since you don't want to memorize separate algorithms for every single cube orientation.

It's not learning the algs that is my main issue.

It's learning how to solve particular F2L pairs.

However once I have solved a particular f2l pair, I am not then able to generate that same case again, so I can't get to practise it repeatedly.

This is especially the case when I doing it intuitively (which is most of the time) because if I have intuitively solved the pair I have no idea how to get it back to how it was.

(If I am using an alg to solve an f2l pair, I can then reverse it to get it back to how it was - but I can't do this intuitively).

 

[brian724080]

It's not learning the algs that is my main issue.

It's learning how to solve particular F2L pairs.

However once I have solved a particular f2l pair, I am not then able to generate that same case again, so I can't get to practise it repeatedly.

This is especially the case when I doing it intuitively (which is most of the time) because if I have intuitively solved the pair I have no idea how to get it back to how it was.

(If I am using an alg to solve an f2l pair, I can then reverse it to get it back to how it was - but I can't do this intuitively).

Sorry - I guess I didn't recognize your problem. When I learned awkward cases, I just solved the cube into that position. It also helps you learn how the cubies interact. However, I think that if your goal is to learn to solve particular F2L pairs, you should just do untimed solves as scrambling to a certain position doesn't give the same effect as if you were actually solving the cube.

 

[Gordon]

I think Marcel's ALG Trainer can do it:
http://www.speedsolving.com/forum/showthread.php?39661-MarcelP-Last-Layer-trainer

Maybe you have to add the F2L cases on your own, but then it should work.

 

[jms]

I think Marcel's ALG Trainer can do it:
http://www.speedsolving.com/forum/showthread.php?39661-MarcelP-Last-Layer-trainer

Maybe you have to add the F2L cases on your own, but then it should work.

Thanks mate. I appreciate that. It's difficult trying to get across what I mean, so I'm glad someone understands what I'm trying to say.

Thanks for the other replies too.

 

[AlphaSheep]

Another way of doing it is to write the inverse of the alg (write it backwards, and then invert each turn or type it into the alg translator). Then you can apply the inverse to a solved cube to get to a state that can be solved by applying the alg. Keep doing it over and over until you're comfortable that you know the alg.

This is a great way to learn any alg, not just F2L algs.

 

[jms]

Another way of doing it is to write the inverse of the alg (write it backwards, and then invert each turn or type it into the alg translator). Then you can apply the inverse to a solved cube to get to a state that can be solved by applying the alg. Keep doing it over and over until you're comfortable that you know the alg.

This is a great way to learn any alg, not just F2L algs.

That's also a good idea. Thanks.

2look OLL and PLL are easier to learn as the algs 'cycle' if you do them a few times.

ie. they go back to their starting point if you just keep repeating them.

 

[Alphalpha]

Another way of doing it is to write the inverse of the alg (write it backwards, and then invert each turn or type it into the alg translator). Then you can apply the inverse to a solved cube to get to a state that can be solved by applying the alg. Keep doing it over and over until you're comfortable that you know the alg.

This is a great way to learn any alg, not just F2L algs.

This is the key.

 

[tseitsei]

That's also a good idea. Thanks.

2look OLL and PLL are easier to learn as the algs 'cycle' if you do them a few times.

ie. they go back to their starting point if you just keep repeating them.

Literally EVERY SINGLE move sequence has this same property...

You can try it yourself. Just take some random moves and repeat them long enough on a solved cube and you will end up with a solved cube

Sometimes you just need very many repetitions...

But for example RUR'U' x6 or RU x 64(iirc) or R2U2 x6

 

[jms]

Literally EVERY SINGLE move sequence has this same property...

You can try it yourself. Just take some random moves and repeat them long enough on a solved cube and you will end up with a solved cube

Sometimes you just need very many repetitions...

But for example RUR'U' x6 or RU x 64(iirc) or R2U2 x6

It's the 'very many repetitions' that worry me.

Also with OLL (for instance) you are only affecting the last layer so it's easy to see if you've messed up the alg when you are practising it.

With F2L I wouldn't know where to begin tracking things to see if I was doing it right every time.

 

[obelisk477]

Literally EVERY SINGLE move sequence has this same property...

You can try it yourself. Just take some random moves and repeat them long enough on a solved cube and you will end up with a solved cube

Sometimes you just need very many repetitions...

But for example RUR'U' x6 or RU x 64(iirc) or R2U2 x6

RU is 105

 

[TDM]

It's the 'very many repetitions' that worry me.

You can check how many times you need to repeat the alg by typing (or copy+pasting) it into this calculator.

 

[jms]

You can check how many times you need to repeat the alg by typing (or copy+pasting) it into this calculator.

Thanks.

I think I might be going about this the wrong way though....

All last week I was getting around 1m20 average. I've spent about a week working on F2L now my averages have dropped by about 20 seconds. I also keep getting more and more cases where I completely mess up and get scores way of 2 minutes.

I also keep going wrong with basic F2L moves where both endge and corner are already on the top layer. These sort of moves I used to do routinely.

I *think* it's because as I am expanding my knowledge of F2L I am trying to look out for more things, instead of just looking for 'opposie colours, same colours or white on top'. Also I am trying to insert using F, R', F', R more and I think trying to use this extra info is actually slowing me down.

Does this make any sense? Is it normal for F2L to get worse before it gets better?

 

[Ranzha]

Fun fact: any sequence requires n repetitions to reach an identity, where n is some factor of 1260. The simplest case I know of that requires 1260 repetitions is R z.

Can anyone prove why this is the case?

 

[TDM]

Fun fact: any sequence requires n repetitions to reach an identity, where n is some factor of 1260. The simplest case I know of that requires 1260 repetitions is R z.

Can anyone prove why this is the case?

R r2 u' needs 840 repetitions... 840 isn't a factor of 1260

 

[EvilGnome6]

Does this make any sense? Is it normal for F2L to get worse before it gets better?

It's totally normal for F2L to get worse before it gets better.

The way I think about it, there are really only two basic ways of setting up the pair to be inserted. They can be split apart with the top colors being different or they can be connected with the top and side colors matching.

The goal is to set up the pair for insertion. If they are both already in the top layer, you either have to split them up or connect them depending on which way they are facing. If one of the pieces is stuck in an edge, you line up the top piece when you bring it up so you can insert it afterwards.

That's an oversimplified view, but it's a better way to look at it than a large collection of different cases.