Older cubers discussions

[mafergut]

Okay, I'll be aiming at that. I suppose the reason I make more mistakes that way is that I do it faster. Must slow down.


Yes, noticed this also. It's much easier to build it under correct or opposite colors. If I build it at 90 degree rotation, I get the opposites correct, but then the other pair of opposites is easily the wrong way around. I suppose this comes with experience.

Yeah, I assume that's a technique that helps a lot for being CN. I am just dual color neutral (white / yellow) and in this case it is very easy to know by heart the order of the colors (blue-red-green-orange for white and the other way around for yellow), but don't ask me right now the order of colors for, e.g. blue cross, I just know they are white, red/orange, yellow, red/orange but don't know which one, so, more or less what you just said. When I try cross on other colors I always use some seconds during inspection to review the color scheme for that cross color but the solve is nowhere near as fluid as with white/yellow.

 

[muchacho]

The thing about how you solve the pairs is that inserting that one edge sucks compared to normal pairing. I understand why you would do this though since you are a Roux solver and do not do F2L but I really think that sticking to the "conventional" method (inserting that edge in 1 first) is definitely better.

Thanks, I'll use the conventional way then.

 

[mafergut]

Btw not many fast people use Yu Da-Hyun's S2L order because there are so few resources on it. However, I believe I read somewhere that Rob Yau uses it. I also use it (switched from Westlund), and I find that transition from solving a 3x3 is slightly easier because instead of thinking of it as S2L, a whole new step, you are thinking of it as more of just cross+F2L. I do white star, then rotate so that purple is on bottom, and my last layer is on pink (I chose these colours only because they were the easiest to do the first time I tried this order). Also, the pieces you know you need to find are more specific, so lookahead is somewhat easier.

It is funny but I just realized I use Westlund in the exact same way as described in the 1st video with just one difference: I go around clockwise and I always start with red. Not sure anymore if I came up with that method intuitively or after seeing a solve by Spanish cuber Alberto Pérez de Rada (APdRF1) but it feels comfortable to me. Actually, watching the vague video of Yu Da-Hyun I didn't really get anything, it looked like the same method as Westlund to me.

 

[Logiqx]

Yeah, I assume that's a technique that helps a lot for being CN. I am just dual color neutral (white / yellow) and in this case it is very easy to know by heart the order of the colors (blue-red-green-orange for white and the other way around for yellow), but don't ask me right now the order of colors for, e.g. blue cross, I just know they are white, red/orange, yellow, red/orange but don't know which one, so, more or less what you just said. When I try cross on other colors I always use some seconds during inspection to review the color scheme for that cross color but the solve is nowhere near as fluid as with white/yellow.

Oddly, I still don't know the colour scheme for green/blue/red/orange off the top of my head. I just glance at 2 centres to confirm the order whilst planning the cross. Whenever I get two opposite cross pieces mixed up it's always on a yellow or white cross which are the colours where I actually know the order!

Nice session this lunch time... Ao12 (17.63) and Ao50 (18.59).

 

[muchacho]

A 30.42 seconds Ao5 (I average 32-33),

Should I focus in something in particular?

I did all the solves with the same color, I'm still usually slower with others.

I know about half of CMLLs, but with half of them I'm probably slower than 2-look.

Spoiler: times and scrambles

29.727 L2 F2 D R2 D2 B2 U L2 D2 R2 F2 R F' D B' F2 D' L2 B L' U
35.645 R2 L2 U' B2 R2 L2 U' R2 U L2 B R' D2 R2 B' L' U' B' D' L' F2 U'
27.814 R2 F2 L2 U L2 U' B2 D' L2 U2 R2 B' R2 F' L' B F' D R2 F' U
28.224 D F2 U R2 D2 R2 D' R2 D' L2 B2 R L' U' R2 D L2 F' L' B' U2 R'
33.310 L2 B2 F2 D' B2 U R2 B2 F2 D' F2 L R2 B2 R' D' R2 F' L' D2 B R

 

[mafergut]

A 30.42 seconds Ao5 (I average 32-33),

Should I focus in something in particular?

I did all the solves with the same color, I'm still usually slower with others.

I know about half of CMLLs, but with half of them I'm probably slower than 2-look.

I don't know a lot about Roux (I learned the basics some time ago and I was around 50 sec average) but I'd say you have a decent FB and LSE but your second block has significant pauses to search for pieces. I also wonder (or maybe I am interpreting your solves all wrong) why you seem to do EO before CMLL as I think most people do CMLL before. In fact I assumed that the most efficient (short) CMLL algs don't even preserve EO so I'm just a bit puzzled here. Not completely sure, as I never learned any CMLL, I just used 2x2 OLLs to orient the corners and T/Y perm to permute them afterwards. And yeah, your CMLL could be a bit faster too, but I'd focus mainly on SB first and try to learn full 1L CMLL as a bonus.

 

[h2f]

I keep improving which is very surprising me, cause it came out of nowhere. I was going to practice crosses but had no time to do it. Today's session gave pbs in ao 5 (17.43), ao 50 (20.48) and ao100 (20.91). I'm only a second to sub-20 in ao100.

Spoiler

solves/total: 100/100

single
best: 15.94
worst: 31.59

mean of 3
current: 19.17 (σ = 2.99)
best: 17.89 (σ = 1.90)

avg of 5
current: 19.89 (σ = 1.87)
best: 17.43 (σ = 0.86)

avg of 12
current: 20.79 (σ = 1.74)
best: 19.12 (σ = 1.35)

avg of 50
current: 20.52 (σ = 2.13)
best: 20.48 (σ = 2.00)

avg of 100
current: 20.91 (σ = 2.17)
best: 20.91 (σ = 2.17)

Average: 20.91 (σ = 2.17)
Mean: 21.08

Time List:
19.36, 20.17, 22.99, 19.68, 23.87, 20.71, 23.20, 25.34, 20.94, 22.20, 25.28, 20.40, 20.14, 22.04, 22.96, 18.30, 19.07, 22.12, 19.48, 24.03, 24.80, 22.43, 25.51, 22.73, 18.70, 20.23, 17.45, 20.56, 26.82, 17.50, 20.18, 18.98, 19.64, 24.74, 17.68, 20.04, 18.92, 20.13, 28.66, 18.80, 18.93, 22.67, 23.02, 18.59, 21.86, 20.13, 17.62, 23.99, 23.28, 20.90, 20.09, 26.20, 21.02, 19.45, 22.04, 18.31, 16.60, 19.99, 17.39, 16.30, 20.38, 19.54, 20.36, 18.46, 31.59, 20.01, 20.14, 21.02, 21.46, 27.05, 29.87, 20.96, 23.10, 22.27, 25.04, 21.30, 18.92, 20.10, 17.15, 16.71, 20.59, 19.44, 17.80, 18.59,19.94, 22.02, 19.02, 20.23, 19.08, 20.64, 22.54, 18.98, 21.68, 26.14, 23.23, 22.08, 18.10, 19.76, 21.82, 15.94

 

[MarcelP]

A 30.42 seconds Ao5 (I average 32-33),

Should I focus in something in particular?

Very nice! Your L6E is very good compared to the total time. I saw two weaknesses. F2B are not very efficient. Well, without reconstructions it is hard to tell, but I think I saw too many moves. And during F2B your pause a lot to find your next pair.. This could be fixed by training lookahead. Unfortunatly I think lookahead on Roux is harder than on CFOP since edges can be on the bottom.

 

[muchacho]

I don't know a lot about Roux (I learned the basics some time ago and I was around 50 sec average) but I'd say you have a decent FB and LSE but your second block has significant pauses to search for pieces. I also wonder (or maybe I am interpreting your solves all wrong) why you seem to do EO before CMLL as I think most people do CMLL before. In fact I assumed that the most efficient (short) CMLL algs don't even preserve EO so I'm just a bit puzzled here. Not completely sure, as I never learned any CMLL, I just used 2x2 OLLs to orient the corners and T/Y perm to permute them afterwards. And yeah, your CMLL could be a bit faster too, but I'd focus mainly on SB first and try to learn full 1L CMLL as a bonus.

I think I did CMLL right after F2B in those solves ***edit: yes, you are correct, in some of the solves I'm removing bad edgest between the first and second step of CMLL *** (although in the past sometimes I removed bad edges before the first step of CMLL, or after the first step or after the second).

Maybe I should be learning how each CMLL affect EO but that's too much information for now.

 

[muchacho]

Very nice! Your L6E is very good compared to the total time. I saw two weaknesses. F2B are not very efficient. Well, without reconstructions it is hard to tell, but I think I saw too many moves. And during F2B your pause a lot to find your next pair.. This could be fixed by training lookahead. Unfortunatly I think lookahead on Roux is harder than on CFOP since edges can be on the bottom.

My L6E is improving in speed, but it has some pauses and it's not more efficient, and the same happens to the F2B. Lookahead may have improved a little, but when I'm rushing I get lost easily.

I'll try to reconstruct some of the solves. I hope some of them have less than 70 moves.

 

[muchacho]

Impressive, you are on fire!

I keep improving which is very surprising me, cause it came out of nowhere. I was going to practice crosses but had no time to do it. Today's session gave pbs in ao 5 (17.43), ao 50 (20.48) and ao100 (20.91). I'm only a second to sub-20 in ao100.

Spoiler

solves/total: 100/100

single
best: 15.94
worst: 31.59

mean of 3
current: 19.17 (σ = 2.99)
best: 17.89 (σ = 1.90)

avg of 5
current: 19.89 (σ = 1.87)
best: 17.43 (σ = 0.86)

avg of 12
current: 20.79 (σ = 1.74)
best: 19.12 (σ = 1.35)

avg of 50
current: 20.52 (σ = 2.13)
best: 20.48 (σ = 2.00)

avg of 100
current: 20.91 (σ = 2.17)
best: 20.91 (σ = 2.17)

Average: 20.91 (σ = 2.17)
Mean: 21.08

Time List:
19.36, 20.17, 22.99, 19.68, 23.87, 20.71, 23.20, 25.34, 20.94, 22.20, 25.28, 20.40, 20.14, 22.04, 22.96, 18.30, 19.07, 22.12, 19.48, 24.03, 24.80, 22.43, 25.51, 22.73, 18.70, 20.23, 17.45, 20.56, 26.82, 17.50, 20.18, 18.98, 19.64, 24.74, 17.68, 20.04, 18.92, 20.13, 28.66, 18.80, 18.93, 22.67, 23.02, 18.59, 21.86, 20.13, 17.62, 23.99, 23.28, 20.90, 20.09, 26.20, 21.02, 19.45, 22.04, 18.31, 16.60, 19.99, 17.39, 16.30, 20.38, 19.54, 20.36, 18.46, 31.59, 20.01, 20.14, 21.02, 21.46, 27.05, 29.87, 20.96, 23.10, 22.27, 25.04, 21.30, 18.92, 20.10, 17.15, 16.71, 20.59, 19.44, 17.80, 18.59,19.94, 22.02, 19.02, 20.23, 19.08, 20.64, 22.54, 18.98, 21.68, 26.14, 23.23, 22.08, 18.10, 19.76, 21.82, 15.94

 

[h2f]

A 30.42 seconds Ao5 (I average 32-33),

https://www.youtube.com/watch?v=13XHK83MNn0

Should I focus in something in particular?

I did all the solves with the same color, I'm still usually slower with others.

I know about half of CMLLs, but with half of them I'm probably slower than 2-look.

Spoiler: times and scrambles

29.727 L2 F2 D R2 D2 B2 U L2 D2 R2 F2 R F' D B' F2 D' L2 B L' U
35.645 R2 L2 U' B2 R2 L2 U' R2 U L2 B R' D2 R2 B' L' U' B' D' L' F2 U'
27.814 R2 F2 L2 U L2 U' B2 D' L2 U2 R2 B' R2 F' L' B F' D R2 F' U
28.224 D F2 U R2 D2 R2 D' R2 D' L2 B2 R L' U' R2 D L2 F' L' B' U2 R'
33.310 L2 B2 F2 D' B2 U R2 B2 F2 D' F2 L R2 B2 R' D' R2 F' L' D2 B R

You should focus on your FB and SB. In the beginning it's hard to plan whole FB, so you should plan 2x2x1 in few moves. FB you can made in no more than 9 moves. Try focus to do the square in 5-6 moves and try to plan it in preinspection. Your 2-look cmll is very good in this stage. Soon it will come time to learn all cmlls. Also LSE is ok. At this stage if you do better FB/SB you will be easily sub-30.

And try to improve your efficency. For exapmle during first solve when you got a 2x2x1 and a edge in the back you could do D' U' R2 D to finish your FB instead of taking an edge adding a corner to make pair with M moves and inserting it. There few tricks like this but you can find it only making untimed solves.

For example first solve I would do like this:
y' x' U' R' M2 D' // square
B U' M2 B // FB (8 moves)
r U2 R U' r2 U R U' R' U' M' U R' U R U' R' U r // long sb, not best but nice fingertricks
R U R' U R U2 R' // sune
U R U R' F' R U R' U' R' F R2 U' R' // perm J
M U2 M U2 M U' M // EO
U'M2 U M' U2 M U2

alg.cubing.net

 

[muchacho]

In untimed solves for the first square I try sometimes inserting it that way, but I'm not doing it yet on timed solves (and anyway I usually just don't see that move in time). Yeah, probably it's in F2B where I can improve more time, I'll work on being more efficient on that, but it's also the hardest part

I think I always try to avoid using B/B' moves, are they comfortable to you?


Oh, I think you are the first subscriber to my youtube channel, thanks!


That D' U' R2 D have never occurred to me, amazing :tu

 

[mafergut]

Oh, I think you are the first subscriber to my youtube channel, thanks!

Now I'm subscribed too

 

[muchacho]

Great, once I reach 10k I'll do a subscribers giveaway

 

[Jason Green]

Great, once I reach 10k I'll do a subscribers giveaway [emoji14]

I'll probably be subscribing tonight when I'm home.

 

[h2f]

Yeah, probably it's in F2B where I can improve more time, I'll work on being more efficient on that, but it's also the hardest part

I think I always try to avoid using B/B' moves, are they comfortable to you?

Yeah, f2b give a lot space to improve. And it's very hard. It was one of the reasons I stopped Roux solves - I could only sometimes plan entire FB. But there's some tricks I use sometimes. B/B' moves are not the best, but I've noticed a lot of Roux solves do it. You can do it like x' U/U'.

Oh, I think you are the first subscriber to my youtube channel, thanks!


Great, once I reach 10k I'll do a subscribers giveaway

 

[Jason Green]

Great, once I reach 10k I'll do a subscribers giveaway [emoji14]

Subscriber number 3. When we are rich and famous cubers we can go on late night TV together and tell how I was your third subscriber when we both averaged 30 seconds! [emoji12] These are the silly thoughts that go through my head sometimes.

 

[Jason Green]

Ok I definitely just had a LL skip! Only got 23 seconds though.

 

[LL Cool Skip]

Just a couple of 4x4 solves. I would've went on but did the same scramble twice and rage quit. Nothing special but everyone should learn Hoya.