Squan parity help

[bldcuber314]

Currently for Squan parity I use the alg you do halfway through the solve, right after orienting both layers. You have to tell whether or not each PLL is solvable or not, is there a good way to do this? I currently have to guess some of the time and sometimes get it wrong. It thought I had found a way(check if the edges are solvable then the corners, there is parity if only the edges or only the corners have parity) but this didn't work. Thanks in advance!

 

[abunickabhi]

I did not understand your way of detecting parity.

How do you find it exactly halfway through?

 

[BVCuber13]

I did not understand your way of detecting parity.

How do you find it exactly halfway through?

I think @bldcuber314 means that right after solving CO and EO, he checks both PLLs on each side to see if they are solvable. @bldcuber314 my suggestion is: solve CP (corner permutation) and then check to see if the PLLs are solvable. If only one is solvable, you have parity. Do the parity alg and then solve EP. If none or both of the PLLs are solvable, you don’t have parity, and you can just solve EP.

 

[Bingus cuber]

I would say do it after CP, because those cases are much easier to recognize parity.

 

[bldcuber314]

Is there a different way to detect it though? I use Beginners with even less algs so sometimes I get a really bad CP. Also after doing the parity alg where CP is already done, I have a diagonal swap on the bottom meaning I have to switch the layers and do another alg.

 

[BVCuber13]

Is there a different way to detect it though? I use Beginners with even less algs so sometimes I get a really bad CP. Also after doing the parity alg where CP is already done, I have a diagonal swap on the bottom meaning I have to switch the layers and do another alg.

I also use beginners. I don’t even do CP/EP even though I mentioned it in my previous post. What I do is I solve one layer of the PBL step, then I do a rotation (x2 or z2) and solve the other layer. If you have parity on the first layer, there is an alg that I use to switch the parity to the other layer (or get rid of parity on all layers if both had parity). The alg is just the line/line EO twice, or 1,0 / -1,-1 / 6,0 / 1,1 / -1,0. If you then have parity on the second layer, just do the parity alg and solve it normally.

 

[bldcuber314]

I also use beginners. I don’t even do CP/EP even though I mentioned it in my previous post. What I do is I solve one layer of the PBL step, then I do a rotation (x2 or z2) and solve the other layer. If you have parity on the first layer, there is an alg that I use to switch the parity to the other layer (or get rid of parity on all layers if both had parity). The alg is just the line/line EO twice, or 1,0 / -1,-1 / 6,0 / 1,1 / -1,0. If you then have parity on the second layer, just do the parity alg and solve it normally.

I don't know the standard parity alg, only the one that messes up CP and EP.

 

[Fred]

My beginner approach is to do CO, EO, and EP, and then check for parity before doing CP, whereas usually people do parity then CP then EP. My parity alg only effects CP, so I thought why not do CP last to make the parity checking easier. For parity, I go to scallop/scallop and swap 3 corners in a way that preserves CO, then reverse my steps to go back to cube shape. I had to write it down or else I'd take a different way back to cube shape and mess up some more stuff.

My EP method is various conjugations of M2: (1,0)/(-1,-1)/(0,1), as described in Cube Master's beginner tutorial. I realized I could adapt that to CP too with a corner version of M2: (-2,0)/(2,2)/(0,-2). (You can also swap around blocks of larger sizes.) This way of doing CP preserves EP.

I think ideally you'd check the parity of each layer (by counting cycles, or comparison to 3x3 PLLs) and eventually get familiar enough to recognize it quickly. I've (rarely) had to do a parity alg twice, even with my simpler method, so don't worry if you mess up some of the time.

 

[Fred]

That sounds the same as mine. If you do a U2, the edges are now correct and it's the corners that need to swap. It ends up being two diagonal corner swaps on top and one on the bottom.

Edit- Ninja'd.

 

[bldcuber314]

My beginner approach is to do CO, EO, and EP, and then check for parity before doing CP, whereas usually people do parity then CP then EP. My parity alg only effects CP, so I thought why not do CP last to make the parity checking easier. For parity, I go to scallop/scallop and swap 3 corners in a way that preserves CO, then reverse my steps to go back to cube shape. I had to write it down or else I'd take a different way back to cube shape and mess up some more stuff.

My EP method is various conjugations of M2: (1,0)/(-1,-1)/(0,1), as described in Cube Master's beginner tutorial. I realized I could adapt that to CP too with a corner version of M2: (-2,0)/(2,2)/(0,-2). (You can also swap around blocks of larger sizes.) This way of doing CP preserves EP.

I think ideally you'd check the parity of each layer (by counting cycles, or comparison to 3x3 PLLs) and eventually get familiar enough to recognize it quickly. I've (rarely) had to do a parity alg twice, even with my simpler method, so don't worry if you mess up some of the time.

How do you fix parity when doing EP first? Do you just leave it and deal with it later?

 

[bldcuber314]

First of all: what do you average? Learning a good parity algorithm and learning to recognize it easily isn't necessary until about sub 1.

I think I'm barely sub 1 lol. I'm wondering as it ruins a bunch of my solves.

 

[Fred]

How do you fix parity when doing EP first? Do you just leave it and deal with it later?

If you're thinking what I think you're thinking, a ninety-degree turn fixes that. For example an adjacent 2-cycle turns into a 3-cycle. So the edges can always be solved.

 

[CaptainK]

I'd recommend learning all CP algorithms if you don't know them already. Only 8 algs are needed (5 algs if you flip [x2]) and they're surprisingly easy to learn. All the moves are +3/-3 which you can think of as U/U' and D/D' moves. And there's like a nice symmetry or easy pattern to the CP algs, if that makes any sense.

Not sure if the parity alg you use will mess up CP. But as others have said, PLL cases will be much easier to recognize if corners are solved.

 

[bldcuber314]

If you're thinking what I think you're thinking, a ninety-degree turn fixes that. For example an adjacent 2-cycle turns into a 3-cycle. So the edges can always be solved.

That's true, but it's a little inconvenient.

 

[DGCubes]

Currently for Squan parity I use the alg you do halfway through the solve, right after orienting both layers. You have to tell whether or not each PLL is solvable or not, is there a good way to do this? I currently have to guess some of the time and sometimes get it wrong. It thought I had found a way(check if the edges are solvable then the corners, there is parity if only the edges or only the corners have parity) but this didn't work. Thanks in advance!

Do you know full PLL on 3x3? Personally I just recognize each layer as if it was a 3x3 PLL, and if only one of them is a valid 3x3 PLL I know I have parity. I think this is the easiest way but unfortunately it's not that feasible if you don't know how to recognize 3x3 PLLs.

 

[Jorian Meeuse]

How I do it is I just don't do squan so I'm 100% certain I don't have parity

 

[bldcuber314]

Do you know full PLL on 3x3? Personally I just recognize each layer as if it was a 3x3 PLL, and if only one of them is a valid 3x3 PLL I know I have parity. I think this is the easiest way but unfortunately it's not that feasible if you don't know how to recognize 3x3 PLLs.

I know full PLL apart from 3 G perms and 1 N perm, I do recognise them but don't know the alg. The problem is that I primarily don't know all the features of each case, only enough to recognise them relatively efficiently.

 

[bldcuber314]

How I do it is I just don't do squan so I'm 100% certain I don't have parity

Or you didn't realise that there is an unsolved squan with parity lying forgotten under your bed. In which case you have 100% parity.