[Help Thread] F2L Discussion and Help

[Llewelys]

For the case above I want my alg to be B’RUR’B

It's best to rotate and insert than to regrip and use awkward B moves.
It's not that good of an idea to aim for rotationless solves with CFOP, if you really don't want to rotate use ZZ or Roux!

"Solving F2L cases from all angles" means:
- not rotating when you don't have to (oriented edge), even if it means using your left hand
- with above point: never rotating twice (y2, y then later in the alg y or y')
- when you do have to rotate, rotating so that you solve a back slot rather than a front slot

Edit: wait I realized you're talking about solving FR slot here, not BL.
Forget what I said, I don't have a good alg for this either (I use f' L' f to pair the edge and corner, then insert with R U2 R')

 

[TreyH]

I’ve been working on my 3x3x3 F2L, and the “cross solved” preset in ChaoTimer is helpful to me for that—sure, solving the cross is easy, but I’m focusing on F2L, so it saves a step.

The weird thing is—while most sequences that it gives me work on either a fully-solved cube or an F2L-solved cube with the top layer still unsolved, sometimes I hit a sequence that only works (i.e., gives me a scramble with cross already solved) on a fully-solved cube, but, if only F2L is solved, does not give me cross solved.

I’ve been trying to work out how this is even possible, and I can’t figure it out. In my mind, none of the top level pieces should matter, since they aren’t involved in cros solved, and all the cross solved pieces are in the F2L solution.

So can someone tell me how I can start with a solved cube, do a set of moves and end up with a cross-solved scramble, but start with an arbitrary F2L-solved cube, do those same moves, and not end up with cross solved?

 

[Owen Morrison]

That doesn't make sense to me either, but I would recommend that you don't skip doing cross, you can skip OLL and PLL for F2L practice, but one of the hardest parts of F2L is the Cross to F2L transition. If you aren't solving the cross before F2L you are not getting the practice you need to have a good cross to F2L transition.

 

[TreyH]

That doesn't make sense to me either, but I would recommend that you don't skip doing cross, you can skip OLL and PLL for F2L practice, but one of the hardest parts of F2L is the Cross to F2L transition. If you aren't solving the cross before F2L you are not getting the practice you need to have a good cross to F2L transition.

Apologies—I was posting to the “Puzzle Theory” forum because it’s a question I’ll wrestle with even if I stop using cross-solved scrambles. I’d just like to understand how this is possible, whether or not it’s useful. Sorry I wasn’t more clear about that.

 

[Mike Hughey]

It should not be possible. Odds are that you are simply misapplying the scramble when doing it with the F2L-only solved cube.

If you have an example where this actually happens, please post it here - if your example genuinely and repeatably gives a solved cross with a solved cube, but an unsolved cross with just F2L solved, it would definitively prove me wrong.

 

[TreyH]

It should not be possible. Odds are that you are simply misapplying the scramble when doing it with the F2L-only solved cube.

If you have an example where this actually happens, please post it here - if your example genuinely and repeatably gives a solved cross with a solved cube, but an unsolved cross with just F2L solved, it would definitively prove me wrong.

Hmm... I use a couple smart cubes that catch me if I make a scrambling error—and I almost never do. When I first started noticing this, I stopped, solved a cube completely and re-tried the scramble—cross solved—then solved F2L and retried the scramble—no cross solved. Then solved the cube and did it again, with the exact same scramble after F2L not giving me a cross again. And then a third time. So I’d convinced myself that human (my) error couldn’t be the issue.

But now I’m wondering if perhaps my error is this: the smart cubes all want you to start a scramble white-side-up. ChaoTimer, OTOH, wants you to start yellow-side-up—which I do, otherwise the cross wouldn’t appear on the white face as I’m used to. But perhaps I’ve been consistently forgetting to reorient “upside-down” when trying to rescramble an F2L solve? (Or rather, consistently reorienting for a solved cube, but every now and again forgetting with an F2L solve—then consistently doing it wrong when I’ve been trying to focus on this issue?)

It had been happening only about 1 out of 10 scrambles, so let me try a bunch with this new hypothesis in hand (so, making sure yellow’s up even when F2L is solved) and see if that was the issue all along.... I hope so, it was truly unsettling to think somehow the top layer could matter in generating a solved cross.

 

[Owen Morrison]

Hmm... I use a couple smart cubes that catch me if I make a scrambling error—and I almost never do. When I first started noticing this, I stopped, solved a cube completely and re-tried the scramble—cross solved—then solved F2L and retried the scramble—no cross solved. Then solved the cube and did it again, with the exact same scramble after F2L not giving me a cross again. And then a third time. So I’d convinced myself that human (my) error couldn’t be the issue.

But now I’m wondering if perhaps my error is this: the smart cubes all want you to start a scramble white-side-up. ChaoTimer, OTOH, wants you to start yellow-side-up—which I do, otherwise the cross wouldn’t appear on the white face as I’m used to. But perhaps I’ve been consistently forgetting to reorient “upside-down” when trying to rescramble an F2L solve? (Or rather, consistently reorienting for a solved cube, but every now and again forgetting with an F2L solve—then consistently doing it wrong when I’ve been trying to focus on this issue?)

It had been happening only about 1 out of 10 scrambles, so let me try a bunch with this new hypothesis in hand (so, making sure yellow’s up even when F2L is solved) and see if that was the issue all along.... I hope so, it was truly unsettling to think somehow the top layer could matter in generating a solved cross.

Could you send a scramble where this happens?

 

[TreyH]

Could you send a scramble where this happens?

If I find one—but I suspect, having thought of how a systematic error might have been creeping in, I’m going to find out that’s all it was. When I get home and can try it I’ll report back.

 

[jdh3000]

In F2L of CFOP, when you have a pair inserted, but you pull it out to pair up and insert another pair, then quickly reinsert it. Sometimes the reinsertion can come before the other insertion of the second pair.

I'm searching for a tutorial on this and didn't know what it was called. I don't know if anyone has a tutorial specifically on it, but anything that goes in it pretty deep.

I've been experimenting with it and just want to see some variation on it.

If anyone knows of a tutorial I'd appreciate it.

Thanks!

 

[GenTheThief]

So, something like this case? As far as I know, that style of F2L alg case where you take out a solved pair to solve another doesn't have any specific name, it just comes up whenever. In general, solving F2L without taking out whole pairs is more intuitive, so cases where you do take out full pairs are typically optimized or genned algs.

 

[jdh3000]

So, something like this case? As far as I know, that style of F2L alg case where you take out a solved pair to solve another doesn't have any specific name, it just comes up whenever. In general, solving F2L without taking out whole pairs is more intuitive, so cases where you do take out full pairs are typically optimized or genned algs.

Yes, I've got a couple that work well that completely reinsert the f2l pair simultaneously. I've seen Brody the cuber and jperm do an example but nothing extensive.

I was was wanting to see what all was available.

Thank you!