Roux F2L to OLL and PLL

I just thought of something interesting. If you can make the 2x2x1 blocks on both sides with the Roux method and then just match the 2 remaining cross pieces to their centers, it would minimize times to get to the OLL and PLL, unless you already have a really fast Fridrich F2L. Since after you make the 2x2x1 blocks on both sides, usually there is at least one of the remaining white(or whatever color you use for the base) already matched with its corresponding center. It's also really not a hassle to match them. I have to do a report right now, so I don't really have time to further elaborate, if you don't understand.

Ah, thats what I used to do when experimenting with Roux.

Unfortunately I'm too slow at block building.

I'm pretty slow too. It takes about a minute just to build on both sides most of the time. >.> I'll keep practicing though. Perhaps I could perfect this into a good method, lol. It's not like it exactly involves many steps, either.

Remember, Roux himself avgs around 15. If you get good at block building, anything can work for making you fast.

if I force myself to turn slowly and look ahead, i can do the 2 blocks in about 15 to 17 seconds. i think I'm going to further investigate this method.
i just noticed, orienting the edges before you complete the remaining 'cross' on bottom gives you already oriented last layer edges also.

i think it's a quite interesting idea. i like it. but it needs people to practice it and post some reports right now i have some other priorities..
good luck anyway, and i'll check this thread

Ahh, before I actually learned the Roux method completely, I used this.
But a little different of what you said.

Right after you build the 2 1x2x3 blocks quickly orient the corners using one of the 7 algorithms, after that, finish the 2 edges, FD and BD. While doing that though, looking ahead to see which of 3 algorithms you need to orient the whole OLL then finish off the PLL.

You can use this with fridrich 3 look also,(the OLL part)

I wish there were more Roux Method videos/tutorials. Block building is harder than Fridrich F2L for me, and probably for the majority of ppl. I still need to actually learn the OLL and PLL algs, heh.

masterofthebass said:

Remember, Roux himself avgs around 15.

Click to expand...

Gilles averages around 13.

And often under, with good lights and without stress. He did it again yesterday, after having drunk a coffee.

Dyste said:

I wish there were more Roux Method videos/tutorials. Block building is harder than Fridrich F2L

Click to expand...

It's also something you need to learn mostly by yourself.

Piercy said:

masterofthebass said:

Remember, Roux himself avgs around 15.

Click to expand...

Gilles averages around 13.

And often under, with good lights and without stress. He did it again yesterday, after having drunk a coffee.

Click to expand...

Hmm... I was just going by his Official results. That is impressive.

I don't see the point in doing this:

1. Still more sequences to memorize if you don't orient edges first
2. Seems like it would have a slightly longer move count
3. If you go with the orient edges first then insert DF and DB method, why not just, after making the blocks, do the corners and continue with the more efficient LSE method.

I don't know what you're talking about with the LSE method. What do you mean by more sequences? I don't know how you'd orient the edges before inserting the DF and DB. You would orient the corners, if anything. The thing to note is that instead of like Fridrich, where you are inserting the corner pairs, after you've completed the two 2x2x1 blocks, they'll already be in place, which gives you more freedom. Then, after building those, all you have to do is insert the DF and DB, which is no hassle at all. Or else, you will be able to orient the LL corners before you do this, and it will not mess them up. You will still get different OLL cases, but notably, they will all be ones where all the corners are oriented, so you can just do OLL and PLL easily.

Piercy said:

Gilles averages around 13.

And often under, with good lights and without stress. He did it again yesterday, after having drunk a coffee.

Click to expand...

Next time, Piercy, you should drink coffee, not beer.
You know, beer makes you solve with blue center on top instead of white, difficult when attempting a blindfold solve. ;-)

No one said you needed to or that this was a good method for BLD.. I don't think you'd want to use the same methods to solve it with 2H and 2H BLD.

Dyste said:

No one said you needed to or that this was a good method for BLD.. I don't think you'd want to use the same methods to solve it with 2H and 2H BLD.

Click to expand...

I think it was more an internal joke than an actual reply.

Oh, I see. My bad. Heh.

Dyste said:

I don't know what you're talking about with the LSE method. What do you mean by more sequences? I don't know how you'd orient the edges before inserting the DF and DB. You would orient the corners, if anything. The thing to note is that instead of like Fridrich, where you are inserting the corner pairs, after you've completed the two 2x2x1 blocks, they'll already be in place, which gives you more freedom. Then, after building those, all you have to do is insert the DF and DB, which is no hassle at all. Or else, you will be able to orient the LL corners before you do this, and it will not mess them up. You will still get different OLL cases, but notably, they will all be ones where all the corners are oriented, so you can just do OLL and PLL easily.

Click to expand...

I'll explain it a different way: Doing this doesn't make sense to me. Why solve the blocks, have a really short step, then have to perform two sequences (I guess I should have called it algs in my last post) for the LL when you can perform one after the two blocks then have a very easy last step (the last six edges [LSE]) and not have anything to memorize.

they'll already be in place, which gives you more freedom.

Click to expand...

But after placing DF and DB you are removing that freedom that M provides (which allows for less moves also).

Edge orienting:

I was talking about the edge orienting idea h3ndrik had:

"i just noticed, orienting the edges before you complete the remaining 'cross' on bottom gives you already oriented last layer edges also."

Also, the blocks are 3x2x1. I'm not saying this way of solving can't be fast, but I think the normal way would be faster.

after further analyzing i realized that its not a very good method. I think normal roux should be faster.
this 'method' could function like a stepping-stone for fridrich users looking into roux (knowing oll&pll but none of the coll algs). although it should be faster to do the corners in 2 steps (orienting them and then permuting them with T (J) or Y perm if you don't yet know the coll cases).
conclusion: even if you knew full fridrich and want to adapt roux-style-block-building in your solves, you should avoid this technique and just use a fast pll alg to permute the corners instead of doing oll & pll.