[Help Thread] Existing Method?

[Loiloiloi]

I am quite new to the cubing community, and just learned 8355. I am hitting times which are below my global CFOP average occasionally and I wanted to ask, is there a method which uses 8355 as a base and creates an advanced version of this? This might be Heise but I have trouble understanding commutators so I don't really know. What I'm referring to is a step which algorithmizes the last 2 steps of 8355. You can build cross, 2 pairs, and then theoretically have a pair insertion which solves ELL, and lastly have an algorithm for the last 5 corners. I'm doing the last 5 corners in anywhere from 3 to 10 seconds without any algorithms so I'm wondering why I can't find any variation of an advanced 8355. If could sub-5 cross+2 pairs, and then sub-4 the F2L/ELL algorithm, it wouldn't be hard to get sub-10 with this, my only concern is that the F2L/ELL might have way too many algorithms. There's 41 F2L algos, and 29 ELL algos, so does that mean it would have 1189 algorithms for this step? I almost always find some way to solve L5C with only 6 sexy moves, so if you could sub-1 sexy*6 couldn't this be a good method?

 

[shadowslice e]

So are you talking about
1) cross
2) 2 pairs
3) ELLLS
4) L5C?
(though I seem to be missing where you solve a final edge)
It does somewhat seem like a CFOP variant of ZZ-CT.
It does also seem pretty similar to some of the variants of M-CELL though because it never became particularly popular I never genned all the algs.

 

[mDiPalma]

you would have to do cross -> 2 pairs - > insert edge - > do LPELL on the final pair - > rotate+sexymoves

you have to do 4 corners, in general, and if you use RUR'U' and URU'R', each corner takes an average of 2 executions. plus 5 ADFs, meaning L5C would only take around 37 moves [RUD], plus the occasional cycle breaking + 5.

that's not that bad, tbh

 

[Loiloiloi]

You don't need to insert the edge. It's impossible to have ELL solved and the edge be flipped, as soon as you complete ELL the last edge will always solve.

One more thing, if you try to solve L5C and fail, you will always get one of 2 easy cases. I'll put the way I solve each of them.

Spoiler

U R U' R' z2 y' D2 (sexy) D' (sexy) D (sexy)*3 D2 (sexy). This is 6 sexy moves, 7 if you count the inverse at the start plus only 4 D moves and can be executed pretty quick.

(Sexy) z2 y' (sexy) D2 (sexy) D' (sexy) D (sexy)*3 D2. This is again 7 sexy moves and 4 D moves.

 

[shadowslice e]

You don't need to insert the edge. It's impossible to have ELL solved and the edge be flipped, as soon as you complete ELL the last edge will always solve.

Also you guys are speaking in a language I don't understand lol, but I'm glad this is interesting.

Yeah but if you solve that at the same time as LS, you will actually have an even higher alg count (around 5!*(2^3)/4= 240 which is a pretty large alg set in addition to L5C which would be 5!*(3^4)/8=1215 leading to a total of 1455 which is pretty ridiculous (although I might be misunderstanding what you're trying to say)

 

[Loiloiloi]

Yeah but if you solve that at the same time as LS, you will actually have an even higher alg count (around 5!*(2^3)/4= 240 which is a pretty large alg set in addition to L5C which would be 5!*(3^4)/8=1215 leading to a total of 1455 which is pretty ridiculous.

So basically insert edge before ELL, and 2-look L5C and count will be reasonable?

 

[shadowslice e]

So basically insert edge before ELL, and 2-look L5C and count will be reasonable?

Well, you would need 4!*(2^3)/4=48 and TTLL+CLS (from MGLS) which is much more reasonable.

 

[Loiloiloi]

Well, you would need 4!*(2^3)/4=48 and TTLL+CLS (from MGLS) which is much more reasonable.

I don't really understand the TTLL+CLS role in this. The way @mDiPalma
Explained how it could work using LPELL made sense to me, wouldn't that be a way to do it?

 

[mDiPalma]

LPELL assumes you have all edges oriented, because i keep forgetting that people actually have misoriented edges at the last slot.

you would otherwise need VHLS

 

[shadowslice e]

I don't really understand the TTLL+CLS role in this. The way @mDiPalma
Explained how it could work using LPELL made sense to me, wouldn't that be a way to do it?

I assumed orient and permute edges with ls, orient corners (CLS), Permute corners (TTLL)

 

[Loiloiloi]

I assumed orient and permute edges with ls, orient corners (CLS), Permute corners (TTLL)

I'm gonna learn some last slot techniques, I think I can hit sub-20 with 8355 using F2L and LS