square-1 speed solving

[AlanAlanine]

how does one speed solve this puzzle -- do you have to memorize all of the algs on cubezone?

I orient corners then edges, then permute the corners and edges using a commutator equivalent to the J algorithm, but i'm pretty slow.

is there a stepping stone between a beginner method and memorizing tons of algs?

 

[James Kobel]

This is the method Thrawst teaches in his video, am I right? Well most begginers methods are usually aimed towards Lars Vandenbergs method.Generaly, you just jump from your first method to that, start by learning all orientation algs for edges(most efficient you can find), then corner permutation(very easy algs), then learn a bunch of EP.When you know around 50 learn all of the cube shape algs, then finish them up. Then just practice, practice, practice.
However, this is certainly not the only method. I would like to compare Lars' method to the Fridrich method(very popular, already proven to be fast) and Jason Baum's method with the Roux method(relatively new and growing in popularity).Most people use Lars method because(1)it is proven to be fast and (2)it is easier to find than other methods.But don't think this is the only method. I use Jason Baums method and I average 35-40, and I have only been solving it for a few months(3-4) and I think it has a lot of potential. Also, another reason I compared it to Roux is because Partially intuitive. But when you average around 1:00-1:10 is when i recomend choosing what method. Sorry I'm so bias

 

[fanwuq]

What if I compare Lars's to Guimond and Jason's to Fridrich?
That's closer to how the actually solving feels.
I use Jason's because it's intuitive F2L with PLL to finish.

 

[smskill12]

I just learned Lars's EPs and I already knew fast ways to put it into cube shape.

 

[blade740]

Start by learning full EO (I need to finish that >_<), CO, and CP, as well as the adj/adj, opp/opp, and parity EP's. Then just learn new EP's as you see fit. Whichever ones take you a long time (or too many algs) to do, learn it optimal. Learn ones that come up a lot. Learn adj/opp. That one is annoying to do in parts.

 

[AvGalen]

There are beginner methods as well for Square-1. I use one that requires only two really short "algs" and I added a parity fix for increased speed. That is enough to reach times of about 1 minute on average. (I learned it from Nerds Paradise)

 

[Swordsman Kirby]

Aww, no one mentioned my LBL method.

I use Jason's because it's intuitive F2L with PLL to finish.

You can do the entire first layer without ANY algorithms? I should learn this.

 

[blade740]

Aww, no one mentioned my LBL method.

That's because you don't have a site with it up.

Or if you do have a site, nobody can find it <_<

 

[James Kobel]

What if I compare Lars's to Guimond and Jason's to Fridrich?
That's closer to how the actually solving feels.
I use Jason's because it's intuitive F2L with PLL to finish.

I see what you're saying, but I made these comparisons because Lar's method has been proven fast and uses mainly algs(like Fridrich) and Jason's method has not been shown it's full potential and uses more intuition, not by how they feel. But you are right if you were comparing by solving style.

 

[AlanAlanine]

so learning the optimal cube shape algs is just memorizing that big diagram on Jaap's page?

 

[AlanAlanine]

BUMP

how do you learn to make the square-1 into a cube with the fewest turns?

 

[qqwref]

You don't need to bump something that's just a day old, this forum isn't THAT busy...

Anyway, yeah, memorizing the table of possible positions (or something similar) is pretty much the way you would learn optimal cubeshape. You can start with the closest positions (one or two turns away from solved), and then when you want to learn the next longer set of positions all you have to do is learn how to set it up to a position you already know the optimal solution for. Or, you could just learn an optimal way to solve each one separately, but that might be more memorization work.

 

[not_kevin]

You don't need to bump something that's just a day old, this forum isn't THAT busy...

Anyway, yeah, memorizing the table of possible positions (or something similar) is pretty much the way you would learn optimal cubeshape. You can start with the closest positions (one or two turns away from solved), and then when you want to learn the next longer set of positions all you have to do is learn how to set it up to a position you already know the optimal solution for. Or, you could just learn an optimal way to solve each one separately, but that might be more memorization work.

I'm thinking of a kind of pseudo-optimal style: memo all of the 1,2, and 3-turn positions (not hard; it's something like 11 cases), and then find the cases that always come up. I've found that adjacent edges, 5 corners with 2-4 edges, 3 corners comes up quite often (all cases with paired edges [aka, all possible (2x,2x) turns work] go to it unless you know faster, and it's almost always optimal, since it's a 4-turn), and adjacent edges, 5 corners with 1-5 edges, 3 corners comes up for about half of the 6-move cases (that particular case is a 4-turn). Those two are pretty big; I can use those to degenerate most of my cases (maybe 10 left?) and when I do, I can average sub-30 without difficulty.