Older cubers discussions

[JanW]

Did another Ao100 today. First 12 solves were horrible, average >55 seconds with 3 sup 60 solves. Couldn't find anything on the cube during F2L. Then I decided to turn a lot slower, did F2L at my slow practice speed, and times immediately dropped. Apparently slow is always the answer. Ao100 ended up at 48.6, nice improvement from last time. Best single 34.6 with one skip in last layer. Hopefully it won't be long before I break the 45s average barrier, and I suppose even a sub30 single could show up soon given a lucky scramble.

For now I'll still focus on getting F2L more fluent, then at some point I must think about how to proceed with last layer. I have a little problem there. A few years ago when I first learned to solve the cube, I somewhere picked up a 4lll method that went orient edges->position corners->orient corners->position edges. When I picked up the cube again I did some experimenting and realized I could do the first 2 in one step and found algs for orienting edges and positioning corners. Later I found this thread, mostly with the same algs I already found, but with some improved algs. So currently I'm doing 3lll which goes CPEOLL->OCLL->EPLL. All algs are faily short and simple. Problem is that this doesn't transition into CFOP very well. Even if I learn full PLL, I'm still stuck with 3lll, with more and longer algs. Not really looking forward to taking that step...

 

[Chree]

https://www.youtube.com/watch?v=g62O8chdwq0

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.

Hoya fo life... Marcel uses Hoya, right?

PS - This is my first time posting here. I wasn't sure if I was old enough to join 31?

 

[muchacho]

Welcome! I saw your video about the yuhu, very useful. But I ended buying the stickerless even if from the video I knew the colors were going to be a problem, I have more trouble with the greens (and the yellow) than with the blues, I've tried but probably won't get used to them.

 

[Logiqx]

Very cool solves Mike. My little boy (6 years) asks if that is you in your avatar with the surfboard hanging upside down.

Thanks. Yes... that photo was when I was learning to loop on holiday a few years ago.

A few years ago when I first learned to solve the cube, I somewhere picked up a 4lll method that went orient edges->position corners->orient corners->position edges.
...
So currently I'm doing 3lll which goes CPEOLL->OCLL->EPLL

EO CP CO EP is what I teach beginners (easiest algs imo) but once they start to time themselves, I switch them to EO CO CP EP which is the path towards full CFOP.

If you already know CPEOLL you could consider combining the last steps as the algorithms are readily available (2GLL). I don't know if that approach has the same potential as OLL + PLL or CLL + ELL and you may well be the only person using it as standard for their LL. I think 2-look OLL + 2-look PLL is probably the safest option for you to take unless you're feeling adventurous!

You may find this diagram interesting - https://www.speedsolving.com/forum/...ast-Layer-Sub-Steps-Subsets-of-OLLCP-and-ZBLL

PS - This is my first time posting here. I wasn't sure if I was old enough to join 31?

Hehe. I think it was discussed a while back and some of us thought 25 was a reasonable measure of "old".

 

[RicardoRix]

I got my 1st 3x3 BLD success yesterday. hoorah!
Today I captured one on video, had 3 failed attempts, all over 12 minutes, and then a success of 11:20 just as my laptop battery died.
https://www.youtube.com/watch?v=EqCj4oDk_ig

My memo takes forever, especially for Edges, corners are not too bad, but the edges are killing me. Even tried reconstructing the solve sighted and got the edges execution wrong twice.
So just over a week till the UK comp, and with any luck it maybe 25% chance of success, it's gonna be a long 12 minutes for the judge waiting to see if I'll ever start before the timer explodes from displaying so many digits.
Possibly a total anti-climax and the feeling of never getting precious minutes of your life back when the cube shall be equally scrambled afterwards as before.

 

[mafergut]

Very nice practice... another Lingpo that doesn't suck... now I'm sure the problem is not my Lingpo but my not very precise movements.

It was a nice OH and a nice 2x2, Mike. Really. I still have much to improve myself to get there, at least in OH, in 2x2 I think I'm a bit closer. Regarding what you say, David, about the LingPo... I hated how my LingPo chatched so frequently so I decided to get a Dayan Zhanchi 2x2 50mm and a Cyclone Boys Stickerless. The CB is great but it also catches a bit. The Dayan is terrible and I'm getting tired of trying to break it in so it becomes the superb 2x2 everyone says it should be. So I came back to the LingPo and either it's catching less and less or I'm adapting my turning style with 2x2 so that it happens less to me. I think the trick is to not grasp the cube too firmly and turn it softly. With that in mind, right now my LingPo is probably the best of the three, closely followed by the CB.

By the way, even though my times are not comparable to those of Mike, I have some new PBs from this week in these two events precisely:

2x2: 4.84 Ao5, 5.71 Ao12, 6.51 Ao100 so, clearly sub-7 with Ortega

3x3 OH: 32.70 single (not sub-30 yet but it was full step), 39.62 Ao5 (first sub-40 Ao5), 42.42 Ao12, 48.19 Ao100 (bad but with a 46.61. Ao50 so I'm rolling the Ao100 this weekend)

In other news
I have ordered a Tanglong, a Cubicle Premium Meiying and a SS Megaminx. Will tell you how I do with them.

EDIT: I just fell 0.02 short of full step PB at 3x3 with a 13.43

 

[MarcelP]

Hoya fo life... Marcel uses Hoya, right?

Hey man. Yeah you are old enough And Yes I still use Hoya.

Thanks. Yes... that photo was when I was learning to loop on holiday a few years ago.

In my childhood at age 12 - 13 or so I remember Stefan van de berg surfing at the lake where I lived. He became Olymic champion or world champion and did these loops as one of the first people to do that. Surfboards where a lot heavier then. Did they call it something like barrel loop? I surfed a lot back then too. I doubt I could today stand on a surfboard and hold my balance.

 

[Chree]

Welcome! I saw your video about the yuhu, very useful. But I ended buying the stickerless even if from the video I knew the colors were going to be a problem, I have more trouble with the greens (and the yellow) than with the blues, I've tried but probably won't get used to them.

Yeah, I know I really liked it at the time I was making that video.. and the turning only ever got better. But the recognition was killing me. I've been back on my SS ever since.

I get corner twists on my SS though. I might have it a bit too loose.

Hehe. I think it was discussed a while back and some of us thought 25 was a reasonable measure of "old".

Haha... I guess I'm in, then.

In other news
I have ordered a Tanglong, a Cubicle Premium Meiying and a SS Megaminx. Will tell you how I do with them.

Tanglong is ridiculous. I love it! I think I'm gonna wind up ordering some of the special color ones. At first I wanted 1 of each, but it's hard to justify spending >$100 on a single model of cube.

 

[mafergut]

Tanglong is ridiculous. I love it! I think I'm gonna wind up ordering some of the special color ones. At first I wanted 1 of each, but it's hard to justify spending >$100 on a single model of cube.

Hi Chris and welcome to the club
I really hope I'll love both the Tanglong and Meiying as my current mains are the Hualong for OH and the Yueying for 2H

 

[RicardoRix]

Did another Ao100 today. First 12 solves were horrible, average >55 seconds with 3 sup 60 solves. Couldn't find anything on the cube during F2L. Then I decided to turn a lot slower, did F2L at my slow practice speed, and times immediately dropped. Apparently slow is always the answer. Ao100 ended up at 48.6, nice improvement from last time. Best single 34.6 with one skip in last layer. Hopefully it won't be long before I break the 45s average barrier, and I suppose even a sub30 single could show up soon given a lucky scramble.

For now I'll still focus on getting F2L more fluent, then at some point I must think about how to proceed with last layer. I have a little problem there. A few years ago when I first learned to solve the cube, I somewhere picked up a 4lll method that went orient edges->position corners->orient corners->position edges. When I picked up the cube again I did some experimenting and realized I could do the first 2 in one step and found algs for orienting edges and positioning corners. Later I found this thread, mostly with the same algs I already found, but with some improved algs. So currently I'm doing 3lll which goes CPEOLL->OCLL->EPLL. All algs are faily short and simple. Problem is that this doesn't transition into CFOP very well. Even if I learn full PLL, I'm still stuck with 3lll, with more and longer algs. Not really looking forward to taking that step...

Sometimes you just need a break from practising one thing like F2L. Muscle memory a few algs, because it's only over time they get set to memory. There are a few really nice PLL's like T-Perm and Y-Perm which are very similar.
Y-Perm is also 2 OLLS in a run, and repeating one of those OLLs gives you a different OLL case, so leaning the Y-Perm algorithm can be used in multiple ways. I struggle, and still do in finding pairs during F2L, the only way to get smoother is to start off slower. Certain F2L cases, where the 2 cubies are adjacent but wrongly orientated, if you practice it over and over then your fluency improves a lot and you need to think about what's happening less so you can concentrate on looking for the next pair. Ultimately you need to do each F2L case without even thinking about and instead 'looking ahead' to find your next pair.

 

[JanW]

EO CP CO EP is what I teach beginners (easiest algs imo) but once they start to time themselves, I switch them to EO CO CP EP which is the path towards full CFOP.

If you already know CPEOLL you could consider combining the last steps as the algorithms are readily available (2GLL). I don't know if that approach has the same potential as OLL + PLL or CLL + ELL and you may well be the only person using it as standard for their LL. I think 2-look OLL + 2-look PLL is probably the safest option for you to take unless you're feeling adventurous!

I'm always feeling adventurous!

I actually did already look into combining the last 2 steps. I noticed that if I made some small mistake in the OCLL alg, I ended up with another OCLL case, then I realized that the mistakes lead to an alternative algorithm for orienting the corners for some other case, with different effect on edge positioning. So a couple of weeks ago I made an excel spreadsheet where I documented all the different variations of the corner orienting algorithms I could come up with and linked them to whatever case they solved. I managed to find algs for 68 of the 80 different 2GLL cases that require corner orientation.

For example, the standard T case I knew (R U2 R' U' R U' R2 U2 R U R' U R) is basically an anti sune, then an inverse anti sune. So I tried out all possible combinations of sunes, anti sunes and their inverses, also by adding U, U' or U2 moves in between them. Turns out 39 different cases (including the standard single S/AS cases) can be solved this way. Then there are other variations, like the alg I used for H case cancelling into the Pi case alg (R U R' U R U' R' U R' U' R2 U' R2 U2 R), combinations of H algs and sunes, and so on.

The good thing with this is that I pretty much know most of the algs already. If I solve with these, I look at the case on the cube, look it up on my list and see something like y' S U' AS, and then I can execute it quickly without having to recall any complex algorithm. The bad news is that they are all so similar that learning to combine each variation with the correct case would take quite a while. Recognition isn't that bad, I look at the color of the F and L edge pieces in relation to the corner colors and that gives away the case.

I am considering learning this, but haven't made any serious efforts yet. Only a couple of the easiest to recognize cases have made it into my normal solving routines. The other option would be to completely change course and head for standard CFOP, haven't decided yet which way to go. If I learn this, I think it would be enough to learn the 48 algs for T, U, L and Pi cases at first. The CPEOLL algs performed on the different OLL cases give an uneven distribution of CO cases, those 4 seem to be a lot more common than the S, AS and H cases. And when a sune comes up, it's still very fast to do that+EP in 2 looks.

 

[Logiqx]

I'm always feeling adventurous!

I actually did already look into combining the last 2 steps ... I managed to find algs for 68 of the 80 different 2GLL cases that require corner orientation.

I am considering learning this, but haven't made any serious efforts yet. ... The CPEOLL algs performed on the different OLL cases give an uneven distribution of CO cases

That's pretty cool. It's certainly easier than figuring out good PLLs without Cube Explorer.

You can always refer to known algs for the remaining cases. Antoine Cantin has a nice list: https://sites.google.com/site/antoineccantin/2gll

I don't think your CPEOLL algs will change the distribution of corner orientation cases. In theory most of them will be equal (4/27) except for H (2/27) and oriented (1/27).

If you do go down this approach, you can always start off by learning the first alg for each case and with practice (thousands of solves) you'll automatically predict the second alg.

 

[LL Cool Skip]

I thought I'd better do a bit of 2x2 practice what with a competition coming up. Here's a decent Ao12...

https://youtu.be/o1eOllvEMb4

You do pretty good with Ortega. I didn't get that fast before I decided to switch to CLL. I recommend sticking with it until you can predict OLL during inspection most of the time. I've been thinking about doing some Ortega practice for this reason. I recorded a spontaneous Ao5 the other night with My crappy LingPo and did pretty well. Hopefully I can do this in comp. It looks to me like case recognition is my weakest point, and inspection. Naturally my video ends in technical difficulties.

I just got my stickerless Dayan 2x2 and I love it!

 

[mark49152]

I got my 1st 3x3 BLD success yesterday. hoorah!
... So just over a week till the UK comp, and with any luck it maybe 25% chance of success, it's gonna be a long 12 minutes for the judge waiting to see if I'll ever start before the timer explodes from displaying so many digits.

Congrats, and good luck at the comp!

A couple of tips: there's a 12 minute total time limit across all 3 attempts, so even if you do only one attempt aim to finish inside 12 min (doubt they will be strict on a single attempt, but just in case). If you know how long your exec is likely to take, set yourself a time limit for your memo and keep an eye on the timer.

Also, remember the stackmat timers cut out at 10 min, so if you think you might go over that, tell the judge and make sure they time you with the stopwatch.

 

[christmasx2]

Look here: (it includes usernames)
https://www.speedsolving.com/forum/showthread.php?54128-How-fast-are-the-over-40-s-in-competitions

Are you old too?

I am 47 and have done at least 10 competitions in 5 years. Always the oldest by a lot!!!! I plan to keep competing. Northeastern 2015 in November is possible.

 

[JanW]

You can always refer to known algs for the remaining cases. Antoine Cantin has a nice list: https://sites.google.com/site/antoineccantin/2gll

Yes, already completed the list with algs from algdb. I like complete lists.

I don't think your CPEOLL algs will change the distribution of corner orientation cases. In theory most of them will be equal (4/27) except for H (2/27) and oriented (1/27).

You are very right. I had a hunch sunes and anti sunes came up very rarely and then kept track of the cases for hundred solves or so. Obviously way too small sample size, but as it aligned with my earlier observations I thought it to be true. But now when I looked closer at the OLL cases and what they really are, it became clear that the distribution is even. If I wanted, I suppose I could slightly affect the distribution by picking the right direction for executing algs on the cases with all edges already oriented. Those algs can be done from 2 or 4 different directions with different results.

If you do go down this approach, you can always start off by learning the first alg for each case and with practice (thousands of solves) you'll automatically predict the second alg.

Yes, I'm sure this is how it would eventually go. If the alg is S+U?+?, then the last part can be pretty well predicted by knowing what effect the first sune will have on the cube. Only need to know if the second part is an inverse or not. Maybe a first step would be to learn to recognize the four cases for S, AS and their inverses based on EP, ideally from all 4 directions. Then I could at least always tell what the latter part of the alg is with a quick look, if uncertain.

 

[Logiqx]

You do pretty good with Ortega. I didn't get that fast before I decided to switch to CLL. I recommend sticking with it until you can predict OLL during inspection most of the time.

That's my plan since there is a lot of scope for me to improve. I'm still very lazy and only try to predict the OLL on easy cases; perhaps 1 solve in 10. I'm not planning on learning CLL or EG any time soon as I don't practice 2x2 regularly and just think of it as a side event.

I am 47 and have done at least 10 competitions in 5 years. Always the oldest by a lot!!!! I plan to keep competing. Northeastern 2015 in November is possible.

I'm guessing you were born in 1968? I'll add you to the next update...

If the alg is S+U?+?, then the last part can be pretty well predicted by knowing what effect the first sune will have on the cube. Only need to know if the second part is an inverse or not. Maybe a first step would be to learn to recognize the four cases for S, AS and their inverses based on EP, ideally from all 4 directions. Then I could at least always tell what the latter part of the alg is with a quick look, if uncertain.

That's kind of what I was thinking.

I think CPEOLL + 2GLL is probably good for OH should you ever get into that event. This is purely based on ease of execution... I have no idea how easy / hard it will be to get used to the recognition.

 

[JanW]

Ao50 today 46.16. I suppose this pace of dropping 2 seconds every 50-100 solves won't go on for long, but I'm happy as long as it does.

New PB 33.89 with a weird solve. I didn't do any Xcross, at least not on purpose, but after placing the first F2L pairs the whole F2L was magically solved. Probably spent 2-3 seconds looking for the remaining corner pieces until I realized what had happened.

 

[Jason Green]

I got a Gans 356 today!! I've been thinking about ordering one since I read Marcel's comment about it being the best. I really like it as is with no tension or lube changes yet. I'm having to adjust a little too the colors since I've been using my stickerless mainly, but I think I'll be using it for now and see how it goes.

 

[christmasx2]

Yes, born in 1968. What's an update?