Older cubers discussions

[h2f]

@mark49152 @newtonbase

I was feeling it because I've counted once average number of moves for both methods for any targets. I wasnt very detailed counting Turbo targets but my careful conclusion was that M2 has a little bit lower movecount. Although our sample is small it is clear that in 2 of 3 the movecount in M2 solution was much lower than in Turbo. It's not a big surprise. In Poland Maskow often suggests that M2 is a **** method but in my opinion he's wrong. He claims it due to UF buffer and transition to comms but the results show that 3 best times in 3bld are made with DF buffer. M2 has also big potential in easy transition to 3style. In fact 3style algs I'm using are mostly advanced M2 tricks plus some usefull comms and algs. Nothing more.

 

[One Wheel]

Today I've made big progress. Still using the beginner's method but my goal was to average less than 1:30.

Today's times:

Average of 5 1:24.82
Average of 12 1:27.81
New Personal Best 1:02.79

I have yet to get a sub 1 minute solve. But one of the things Youtuber "Spotty" suggested was not to try switching to CFOP until your beginner's method was consistently below 1:30. I can see why. There's a lot to learn just on handling the cube smoothly, not mistaking clockwise vs counter-clockwise algs. Getting faster at the cross, etc.

I've been practicing the 3 F2L moves separately. I think I'll wait until I can consistently (over 25 solves or more) get below 1:30 before I start changing methods. Then the real work starts with memorizing OLL and PLL.

What a blast it's been learning this cube though.

I don't remember when I switched, and if you're enjoying it that's certainly the main thing, but generally I'm not a fan of waiting until a certain speed or anything else to learn something new. Just go ahead and learn it, and especially with f2l, in the process of learning you will get a better feel for what the cube is doing and get faster. Just my 2 cents.

 

[newtonbase]

Here we go. Scramble in solving orientation.

1. D2 L2 U2 B F2 U2 F' U2 R2 D2 F2 U R' B D' F' D B2 R2 F2

[L R: U M' U2 M U] // BL FR 9/9
x' [U L U': M2] // LD 7/16
[U' R' U: M2] x // RD 7/23
M2 // UB 1/24
Uw2 M' Uw2 M' // UF DB 4/28
x' [U L' U': M2] x // LU 7/35
[U' L' U: M2] // FL 7/42
x' [U' R U: M2] x // RU 7/49
[R': ((M' U)3 M U)2] // flip RB 18/68

2. U2 L' D2 L' D2 R' D2 F2 U2 L2 U L2 U' F' L' D' L' R D2 U

[U' L' U: M2] // FL 7/7
x' [U L' U': M2] x // LU 7/14
U R2 U' M2 U R' U' M2 U R' U' // DR FR 11/25
M' U2 M U2 // UB UF 4/29
[B': [U R' U', M2]] // BU LB 10/39
[U R' U': M2] // BR 7/46
[L2: U M' U2 M U] // DL UR 7/53
U2 M' U2 M' // DB 4/57
[U' L U: M2] // BL 7/64
[D' L2 D: M2] // parity for OP 7/71

3. R2 U B2 U' L2 F2 D' F2 L2 D L2 F' U2 B' D2 R F'U2 F' D F2

[Uw: [U R' U', M2]] // LB UL 10/10
x' [U L U': M2] x // LD 7/17
[L' R: U' M' U2 M U'] // FR FL 9/26
[Bw: U M' U2 M U] // RU RD 7/33
(U M')3 U M (U M')4 // BU 16/49
M Uw2 M Uw2 // UF DB 4/53
[U R' U': M2] // BR 7/60
U2 M' U2 M' // DB 4/64

I would have done most of those in a very similar way but one thing I notice is that I only use comms/tricks with a letter pair but you will use the 2nd letter from 1 pair and the 1st from the next such a when you did UB on its own then a UF DB comm trick.
A couple other differences - for RU RD you use the Bw setup where I use [U M' U': L2].
Do you know this for LD RD - M' D' M D2 M' D' M?

 

[Fyzzna]

I don't remember when I switched, and if you're enjoying it that's certainly the main thing, but generally I'm not a fan of waiting until a certain speed or anything else to learn something new. Just go ahead and learn it, and especially with f2l, in the process of learning you will get a better feel for what the cube is doing and get faster. Just my 2 cents.

I very much agree, never been a fan of waiting until you are below an arbitrary time before learning something new.
Although CFOP is the most commonly used method, I would suggest you look through this thread and play around with some other methods before you decide which one to go with.

 

[h2f]

I've counted avg number of moves per alg + setup it goes like Turbo is around 6,8 per target and M2 is around 5,8 per target.

 

[mark49152]

I was feeling it because I've counted once average number of moves for both methods for any targets. I wasnt very detailed counting Turbo targets but my careful conclusion was that M2 has a little bit lower movecount. Although our sample is small it is clear that in 2 of 3 the movecount in M2 solution was much lower than in Turbo. It's not a big surprise. In Poland Maskow often suggests that M2 is a **** method but in my opinion he's wrong. He claims it due to UF buffer and transition to comms but the results show that 3 best times in 3bld are made with DF buffer. M2 has also big potential in easy transition to 3style. In fact 3style algs I'm using are mostly advanced M2 tricks plus some usefull comms and algs. Nothing more.

Thanks for doing this, it was interesting and I think I understand Turbo a bit better now. From previous discussions on the forum it seems there's not much of a difference between UF and DR buffers. It's just preference. Move count is not hugely different (it might be less than vanilla M2) and the setups aren't as bad as I expected. The algs also seem quite nice, just OLLs. It forces you to solve a pair at a time, but I don't see how it could be easy to transition to comms, at least not as easy as M2, and I'm not sure what potential there is for move-saving tricks, etc.

I would have done most of those in a very similar way but one thing I notice is that I only use comms/tricks with a letter pair but you will use the 2nd letter from 1 pair and the 1st from the next such a when you did UB on its own then a UF DB comm trick.

I do that a lot because I got in the habit of checking every adjacent pair for opportunities to combine BD or FU, since it makes such a massive difference to move count. I don't tend to think in pairs for edges, just individual targets plus opportunities to combine them. I know from corners that will change when I switch to 3-style. For example, I always execute my last three corner targets (with parity) as comm then Y perm. It's often occurred to me afterwards that it would have given nicer cases to do the Y perm first and finish with a comm, but I cannot mentally "unglue" those pairs any more during corners .

A couple other differences - for RU RD you use the Bw setup where I use [U M' U': L2].
Do you know this for LD RD - M' D' M D2 M' D' M?

I use exclusively M slice interchanges with very few exceptions. [U M' U': R2] doesn't qualify as advanced M2 . There are a couple of E slice comms I have started using in slow solves simply because they are way more efficient than anything I can set up on the M slice.

I have seen M' D' M D2 M' D' M but couldn't figure out how to finger it quickly.

 

[h2f]

It's just preference.

That's how I think about it.

 

[newtonbase]

Thanks for doing this, it was interesting and I think I understand Turbo a bit better now. From previous discussions on the forum it seems there's not much of a difference between UF and DR buffers. It's just preference. Move count is not hugely different (it might be less than vanilla M2) and the setups aren't as bad as I expected. The algs also seem quite nice, just OLLs. It forces you to solve a pair at a time, but I don't see how it could be easy to transition to comms, at least not as easy as M2, and I'm not sure what potential there is for move-saving tricks, etc.


I do that a lot because I got in the habit of checking every adjacent pair for opportunities to combine BD or FU, since it makes such a massive difference to move count. I don't tend to think in pairs for edges, just individual targets plus opportunities to combine them. I know from corners that will change when I switch to 3-style. For example, I always execute my last three corner targets (with parity) as comm then Y perm. It's often occurred to me afterwards that it would have given nicer cases to do the Y perm first and finish with a comm, but I cannot mentally "unglue" those pairs any more during corners .


I use exclusively M slice interchanges with very few exceptions. [U M' U': R2] doesn't qualify as advanced M2 . There are a couple of E slice comms I have started using in slow solves simply because they are way more efficient than anything I can set up on the M slice.

I have seen M' D' M D2 M' D' M but couldn't figure out how to finger it quickly.

I'm using some E and S slices in practice solves too. There's a bit too much thinking required for them at the moment.

I don't worry too much about fingertricks when I can halve the move count

 

[pglewis]

But one of the things Youtuber "Spotty" suggested was not to try switching to CFOP until your beginner's method was consistently below 1:30. I can see why. There's a lot to learn just on handling the cube smoothly, not mistaking clockwise vs counter-clockwise algs. Getting faster at the cross, etc.

Great progress, you should be shopping around for a comp nearby soon!

I'm another that doesn't subscribe to "don't switch from beginners' until sub X". Moving to F2L is initially going to slow you down and the faster you get with the Beginners' method the more reluctant you'll be to leave your comfort zone where you've already put effort into developing some lookahead. I also agree with @One Wheel that you're having fun so it's guaranteed you're not doing it "wrong", in any event.

As far as algs: I would get the 7 OCLL algs down first, if you haven't already, I think most people learn those first with 2-look OLL systems that orient edges as the first step. Next up would be PLL, IMO. I think they're the next "bang for your buck" alg-set to learn and except for a couple they're more frequent than any particular OLL so you'll run across them more often in actual solves. IME, recognition is the biggest challenge so it can be highly beneficial to start learning to recognize the PLLs you come across, even if you're still 2-looking and haven't learned the algs for them yet. I think recognition is more than half the battle.

General advice: if you're not comfortable with M-slice moves yet give them extra practice. Some things like H and Z perm can be executed very quickly with the M-slice algs, once you have the finger tricks down. I learned alternate versions of Z first because my M slices were awkward at the time and I wouldn't take that route if I had it to do over. Slice moves are super handy for working with edges.

And of course always have fun. I continue to waffle around and experiment with all sorts of things I don't always follow up on (color neutral, ZZ), and sometimes silly things that don't even progress my speedsolving: solving straight to checkerboard or solving 3x3 corners with a 2x2 method and the edges with U-Perms because... why not?

 

[mafergut]

@pglewis: those RU algs for H, Z and U perm will be handy anyway for OH and big cubes so it was not wasted time to learn them.

 

[moralsh]

That's how I think about it.

I'd also say it's mostly preference, and as far as getting a good single goes it's clear the buffer is not that important. However some experienced BLDers argue that having a buffer in the U layer helps setting up to an OLL or PLL parity fix. Not that any of us is at that level.. yet

 

[mark49152]

some experienced BLDers argue that having a buffer in the U layer helps setting up to an OLL or PLL parity fix

How does that work?

 

[NewbieCuber]

I don't remember when I switched, and if you're enjoying it that's certainly the main thing, but generally I'm not a fan of waiting until a certain speed or anything else to learn something new. Just go ahead and learn it, and especially with f2l, in the process of learning you will get a better feel for what the cube is doing and get faster. Just my 2 cents.

I'm finding that by waiting until I can average 1:30 or faster I'm focusing on handling the cube better, remembering the algs properly, and building my cross faster. I can see the value in waiting a bit before moving on so I can concentrate on the fundamentals.

However, I also have my "F2L" cube which is a cube with all the top layer stickers removed (except for the yellow center piece). I'm practicing the F2L moves on that cube because it simplifies the problem and helps build pattern recognition. In fact, I have a similar "cross cube" with all the stickers removed except for the centers and the white cross. It really helps me focus on a single aspect of the solve.

 

[newtonbase]

I'm finding that by waiting until I can average 1:30 or faster I'm focusing on handling the cube better, remembering the algs properly, and building my cross faster. I can see the value in waiting a bit before moving on so I can concentrate on the fundamentals.

However, I also have my "F2L" cube which is a cube with all the top layer stickers removed (except for the yellow center piece). I'm practicing the F2L moves on that cube because it simplifies the problem and helps build pattern recognition. In fact, I have a similar "cross cube" with all the stickers removed except for the centers and the white cross. It really helps me focus on a single aspect of the solve.

That's proper practice. You'll reap the benefits.

 

[pglewis]

@pglewis: those RU algs for H, Z and U perm will be handy anyway for OH and big cubes so it was not wasted time to learn them.

lol, at this point I'd have to relearn those... luckily OH comes just slightly before feet as far as me getting involved . Bigger cubes maybe, but I actually still just use my normal 3x3 algs for both 4x4 and 5x5.

 

[muchacho]

I will post solutions that I would use in a real solve, so maybe a few tricks, but fundamentally M2. Same should go for Turbo. It's not so interesting to see "vanilla" solutions, but also not interesting to see unrealistic tricks that turn it into a different method. I'm interested to learn more about how Turbo would be used for real by an experienced Turbo user, too. Maybe others will post better M2 solutions too.

Here we go. Scramble in solving orientation.

1. D2 L2 U2 B F2 U2 F' U2 R2 D2 F2 U R' B D' F' D B2 R2 F2

[L R: U M' U2 M U] // BL FR 9/9
x' [U L U': M2] // LD 7/16
[U' R' U: M2] x // RD 7/23
M2 // UB 1/24
Uw2 M' Uw2 M' // UF DB 4/28
x' [U L' U': M2] x // LU 7/35
[U' L' U: M2] // FL 7/42
x' [U' R U: M2] x // RU 7/49
[R': ((M' U)3 M U)2] // flip RB 18/68

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

Not an experienced Turbo user, but here I go (after like 5 tries):
L R U M' U2 M U R' L'
R2 L2 M U' M U2 M' U' M' L2 R2
U' U M' U2 M U U
B R' M U M U2 M' U M' R B'
L' U' M' U2 M U' M U M U2 M' U M' L
B F2 U2 M U M U M U2 M' U M' U M' F2 B'

68 STM // looks like a lucky solve for Turbo (DF buffer)

I'll try to do the other two.

 

[mark49152]

Not an experienced Turbo user, but here I go (after like 5 tries):

Interesting to see Turbo done with a DF buffer.

 

[pipkiksass]

SO glad I never got sucked in to BLD! I barely have time for 3x3 and 4x4...

Anyway, did about a dozen 3x3 solves this morning, averaged 16.8 and had a sub-16 Ao5, which was nice. Also included PB NL single (11.180)... does it count as non-lucky if there's a fairly straightforward x-cross?

Apologies if my reconstruction is wrong, I NEVER use cube rotation notation, so I'm never sure how to use it, and I can't get Alg.garron.us to work for me (I'm on a mac, bleugh).

Give the scramble a go, it's loverly! And I missed an even easier 2nd pair - there may well have been easier F2Ls.

D2 R2 D B2 F2 U B2 U’ L2 F2 D’ F U’ L2 D2 L’ F’ U’ L D’

Inspection (X2 Y)

D’U R’ F D2 F (x-cross)
R’ U2 R d’ R U R’ (2nd pair)
Y L’ U’ L U2 R U’ R’ (3rd pair)
U L’ U L (4th pair)
f R U R’ U’ R U R’ U’ (OLL
U M2 U M U2 M’ U M2 (PLL)
U’ (AUF)

50 STM
11.180
4.472 TPS

 

[mark49152]

SO glad I never got sucked in to BLD! I barely have time for 3x3 and 4x4...

Dude you're missing out .

 

[newtonbase]

SO glad I never got sucked in to BLD! I barely have time for 3x3 and 4x4...

The obvious answer is to stop doing those trivial events.