New(?) LS + LL Approach

[JackJ]

Hi all,
Before I start, I've developed this alot more than most that post in the new subset thread, that's why I've decided on a new thread. Just a clarification.

This is something I've been working on for a few weeks now and I'm finally able to announce it! It's a bit like MGLS & L2L4.

Essentially, on your last slot you solve EO + F2L corner then you solve F2L edge + CO. 65 cases for the first step and 34 for the second step.

I've only published about half of the first steps algs, due to a few reasons. It can be done intuitively, the cube images are kind of confusing, and I'm back in school now. No time.

I've settled on the name JJLS, for now. I need to give a huge thank you to qqwref for generating the algorithms for me and helping me along the way. You're the best!

Anyway, here's an example solve and the link to the website.

Spoiler

U2 B F2 R L' D' R2 D2 L' R' D' U R B2 R' F2 L' F' U2 R' L2 F L2 F' B'
y x2
R D R' D2 U' F2
U2 L' U L U R' U R
L' U' L U' L' U L
y U2 R U2 R2 U' R2 U' R'
L' U' L y' R U R'
R U R' U2 R U2 R' U2 R U' R'
y' R U' R U R U R U' R' U' R2

http://johnstoncubing.webs.com/jjls.htm

 

[mark49152]

Why? Is there any particular advantage in this approach?

 

[Rubiks560]

Why? Is there any particular advantage in this approach?

IMO this is like asking if there is an advantage to using WV. Of course there is. You get an OLL skip.

 

[StachuK1992]

I like it, got a 22-move LS+LL.
I'll be sure to play around with this some more!

Edit:
Another example:

Scramble: B' F' L2 B R2 D2 B' D2 L2 D2 R2 D R2 B F' L U' B2 L2 R' D'

Cross: (x2) B' R2 D R' L F2 (6/6)
F2L 1: F' U2 F R' U' R (6/12)
[I want to do R U R' U F L F' L' to fix EO but don't want to skip steps!]
F2L 2: L U' L' U L U' L' (7/19)
F2L 3: F' U F2 U F' (5/24)
JJLS1: U2 (y') R U2 R2 U' R2 U' R' (8/32)
JJLS2: U2 (y) R U' R' U' R U' R' U R U R' (12/44)
PLL: (y2) Gperm (12/56), not bad.

Spoiler

I was originally going to do this with ZZ but accidentally skipped the EP!

EOLine: z2 R' D' F2 L2 D B2 (6/6)
Right block: L' U2 L R U R U' L U2 R (10/16)
Left block: (y2) R' U2 R' U' R' U R' U' R (9/25)
OLL (with influence to PLL): R' U L U' R U L' (7/32)
PLL: R2 F2 R' B' R F2 R' B R' (9/41)

 

[mark49152]

IMO this is like asking if there is an advantage to using WV. Of course there is. You get an OLL skip.

Not sure if you're being sarcastic but I'll reword it. Why is this better than just solving the pair and doing OLL? Skipping OLL doesn't come for free.

 

[Rubiks560]

Not sure if you're being sarcastic but I'll reword it. Why is this better than just solving the pair and doing OLL? Skipping OLL doesn't come for free.

This is less moves than doing insert last pair + OLL.

 

[JackJ]

Not sure if you're being sarcastic but I'll reword it. Why is this better than just solving the pair and doing OLL? Skipping OLL doesn't come for free.

In certain situations it's fast. Doing a 3 move EO + corner is definitely better than a normal ~7 move F2L pair. While I haven't crunched any numbers, I'd say CO + edge is about the same movecount as a normal OLL.

I never claimed an advantage. It's just a different approach much like MGLS in that respect. I'd say they're about even.

EDIT: ninja'd

 

[mark49152]

Ok thanks. I assumed you had a good reason for swapping the edge and corner in MGLS but your site doesn't say what it is, unless it's to make the second step 2gen? Is move count better than MGLS?

 

[JackJ]

Actually both CLS & this can be 2gen. If you're wondering about movecount or anything similar, I'd refer you to qqwref. I'd assume they're very close in movecount, though.

 

[blokpoi]

I really like this, once all the algs are generated and I play around with it for a bit I may just end up learning it all. It sounds very promising to me!

 

[TheNextFeliks]

KLMGLSBNR

Spoiler

Kimda like MGLS but not really

 

[StachuK1992]

I see this as a trick for ZZ more than anything else.
I think knowing bits of CLS+WV+RSL+this+others are pretty useful for the easiest-to-recognize cases of each set.

I don't think it'd be worth doing all on its own, but it's worth further investigation!

 

[mark49152]

Actually both CLS & this can be 2gen. If you're wondering about movecount or anything similar, I'd refer you to qqwref. I'd assume they're very close in movecount, though.

Personally I don't think move count is a big deal. An extra look could be. As you said above, if the corner+EO is short and flows out of F2L with no pause, maybe it's fast.

Is MGLS widely used?

I see this as a trick for ZZ more than anything else.
I think knowing bits of CLS+WV+RSL+this+others are pretty useful for the easiest-to-recognize cases of each set.

Agreed. With ZZ at least the first step will always be short and fast. Once the first step is done though, you're committed to the second step so I don't see how it's worth learning just a few of the 34 second step cases like it is with WV, VH/ZBLS, COLL etc.

 

[uberCuber]

Is MGLS widely used?

I'm sure a decent number of people know/use some of the easiest CLS cases just because they are so nice, but the MGLS method in full? No.

 

[bobthegiraffemonkey]

I'd like to have a look at the second step algs, some could be useful for my weird mega L2F system. I don't care about images, I'm happy to work out for myself what each alg does.

 

[BillyRain]

I'm going to learn like... everything.

 

[JackJ]

Glad to hear some are very interested in it!

 

[antoineccantin]

I thought this was MGLS?

 

[vcuber13]

No for mgls you do orient the edges while placing it, and orient the corners while placing it. This variation is doing corner/edges then edge/corners.

 

[Escher]

Imo for Fridrich users you'll get most utility by just learning the latter step (LE + CO) - it's easy to simply leave a slot that has a corner already done in the middle of your solve, and then make sure by your final pair you've solved EO intuitively too.