Older cubers discussions

[mark49152]

Guys, I added a TL;DR to the top of that OLL guide - let me know what you think!

 

[earth2dan]

including some lefty mirrors that I'm sure are slower than righty alternatives but they seemed so easy to learn at the time... mistake!!!

I don't know if I'd call that a mistake per say. If you're proficient with the mirror and the righty alternative has a different starting position you'll be ready to attack that OLL from multiple angles, speeding up recognition and eliminating U set ups. I have a couple PLL's where I learned new algs from a different starting position, but I'll often use my original alg to avoid set ups or cube rotations if the case appears in the right starting position for it. I really need to do this with my U perms, so many wasted moves because I can't execute them from the back.

Guys, I added a TL;DR to the top of that OLL guide - let me know what you think!

Beautiful

 

[mafergut]

I don't know if I'd call that a mistake per say. If you're proficient with the mirror and the righty alternative has a different starting position you'll be ready to attack that OLL from multiple angles, speeding up recognition and eliminating U set ups. I have a couple PLL's where I learned new algs from a different starting position, but I'll often use my original alg to avoid set ups or cube rotations if the case appears in the right starting position for it. I really need to do this with my U perms, so many wasted moves because I can't execute them from the back.

To be honest that doesn't work for me. I end up not using the other alg and forgetting it in a few months. With PLLs I just know two PLLs for U perms but I learned the <M,U> ones (my main ones are the <R,U> ones) and I'm much slower with those that I think it pays for me to do pre and post AUF, maybe not with both but for sure with the one that has the U's. As I do M2 with my left ring and middle I have to do the U' with my right index in OH fashion (push on FUR with the back of the finger) and I'm terrible at that. Other PLLs that I changed the alg for I have already forgotten the old algs and the only other PLL I know for different orientations is A perms, but it's just lefty mirror again. I'm a bit slower here with left but if the orientation is already correct for left I am proficient enough to save some time avoiding cube rotations or AUFs, also because A perms have a rotation and the post AUF ends up being on B instead of on U, with takes a bit more time to regrip/rotate back.

 

[megagoune]

Anyone here know full OLL? I know less than half and I two look most cases. Looking for some tips on how to learn the rest efficiently. If I can stay focused, and avoid distraction or falling into bad habits I can consistently squeak out a sub-20 average, but globally I'm still around 20.3. I know I get a lot of sup-20 solves where my bad OLL is pushing me over 20.

Do you recommend learning them in groups of similar cases, or just powering through drilling one at a time until it's down? I usually go to http://algdb.net/ for algs.

The Winter months are when I seem to make all my speedcubing progress, then I just kind of coast through Summer. I've set some Winter goals for myself and I'm sure I can hit my big cube goals with just practice and improving look-ahead, but I think I'm due to learn some new algs to improve 3x3.

I remember the good old times of learning full OLL. Learning full sets, at least for me, drops your times suddenly, and it feels super nice. I learned OLL to drop under 30s.
I learned by groups or U sticker shapes (P, T, W, knight, awkward...). I find useful to associate a name to an alg (or at least a group of algs).
The main problem I had, and I have no solution for it, is that it's difficult to setup a case. When learning PLL, most of them are the inverse of themselves, but not for OLL. So you have to repeat the same OLL many times to come back to solve state. It meant longer time to learn how to recognize cases for me.

 

[newtonbase]

I'm probably less than halfway through OLL but I know 3 of each U perm!

 

[Shaky Hands]

The main problem I had, and I have no solution for it, is that it's difficult to setup a case. When learning PLL, most of them are the inverse of themselves, but not for OLL. So you have to repeat the same OLL many times to come back to solve state. It meant longer time to learn how to recognize cases for me.

You can setup fairly easily by reversing the alg.

e.g.: to setup the easy T-case solved by F [R U R' U'] F' you can setup with F [U R U' R'] F' or to setup the other T-case solved by [R U R' U'] [R' F R F'] you can setup with [F R' F' R] [U R U' R'].

This should be quicker than cycling the cube's state by repeating the alg multiple times.

 

[pglewis]

Guys, I added a TL;DR to the top of that OLL guide - let me know what you think!

Nice touch, thumbs up. I love this guide, I started down the same road but not spending the time to gather real metrics, just glancing at the algs and going by "feel". Made my day when you linked it a few weeks ago.

The main problem I had, and I have no solution for it, is that it's difficult to setup a case. When learning PLL, most of them are the inverse of themselves, but not for OLL.

You probably know you can just invert the alg, reading it backwards and reversing all clockwise/anticlockwise moves. Granted, I probably have a 60% flub rate when I do it that way. alg.cubing.net has a button in options to invert whatever you paste into the moves.

Do you recommend learning them in groups of similar cases, or just powering through drilling one at a time until it's down?

The advice I always see and can verify it holds true for me as well: learn all the algs for a shape. That way, you see the shape and you know you can 1-look it, rather than spend time figuring out if you know that one or not.

 

[h2f]

I have a very busy week. I got nice 58.xx 3bld solve. After few hours I did another few solves and the first was this:

There's a lot of room for improvment but I was tired. Execution was 35 and it took 82stm 82/35=2,34 - in fact it was a little bit more due to fact I did one comm twice because I knew I did it inverserly.

 

[Selkie]

@h2f Grzegorz, great solve mate with a wow memo!, so, so motivating.

Mixed week for me, trying to get a sub 6m 7x7 mean and so close but with a 5:44.xy single hopefully get it on on film tomorrow.

Life a bit stressy last week or so, have a PB for Risk of Redundancy which Should be an official average by Christmas!!

 

[mark49152]

That's crappy news Chris, I hope it works out.

 

[Jason Green]

I learned OLL by printing at bad mephisto's cheat sheet and picking out whatever looked easy progressing to hard. This is not how I would suggest. I should probably get some better ones, but I just know lookahead is my biggest pitfall still that's why I still mostly just practice.

Also @MarcelP has an OLL trainer software that's pretty cool! I didn't use it much though.

 

[Shaky Hands]

Good luck Chris. Hope everything works out as best as possible.

 

[kbrune]

@mark9152

That 4 BLD solve.. wow!

Learning 4 and 5 blind will be coming soon for me. I'm really enjoying BLD much more these days. I managed my first sub 3 min 3BLD recently and I've made a first attempt at 5 MBLD. Got 4/5 in 48.xx. I need to organize a better memo system though if I'm going to improve more. My memo isn't very efficient. Do you use Roman rooms for longer memo? What's your memo strategy? Anything you wish you knew in your earlier days?

@h2f
Very nice sub 1 solves. Same questions to you.

 

[JohnnyReggae]

Anyone here know full OLL? I know less than half and I two look most cases. Looking for some tips on how to learn the rest efficiently. If I can stay focused, and avoid distraction or falling into bad habits I can consistently squeak out a sub-20 average, but globally I'm still around 20.3. I know I get a lot of sup-20 solves where my bad OLL is pushing me over 20.

Do you recommend learning them in groups of similar cases, or just powering through drilling one at a time until it's down? I usually go to http://algdb.net/ for algs.

The Winter months are when I seem to make all my speedcubing progress, then I just kind of coast through Summer. I've set some Winter goals for myself and I'm sure I can hit my big cube goals with just practice and improving look-ahead, but I think I'm due to learn some new algs to improve 3x3.

Knowing full OLL will definitely help shave off a few seconds. It took me around 6 months in total to learn full OLL. it can obviously be done much faster. I worked through the shapes, grouping similar shapes together. I used this as a reference initially and then built up a spreadsheet will all the algorithms that I use ... http://www.cubewhiz.com/oll.php

I don't use all the algs in the link above, but rather tried to find algs that suited me using http://algdb.net/ . I have even changed some algs that I used for quite a while because I didn't like the way they flowed. I still have some OLL's that I really don't like executing but haven't found alternatives that I like.

There are also more alternative OLL algs here ... https://www.speedsolving.com/wiki/index.php/OLL

Definitely persevere it will pay off. I must say with OH I do mostly 2Look for OLL and it frustrates me having to do 2Look, but I don't have the energy ATM to learn almost another full set of OLL algs.

 

[mafergut]

Sorry to hear that, Chris. I hope it all ends well.

Also, Grzegorz, you make it look easy! Go for that official sub 1!

 

[mark49152]

Do you use Roman rooms for longer memo? What's your memo strategy? Anything you wish you knew in your earlier days?

Thanks Ken. Yes I use rooms for BigBLD and MBLD. On 4BLD I use audio for centres and memo those last, execute first.

The only major change I made since the early days was to compile a letter pair image list. At first I convinced myself that it was better to go with whatever came into my head naturally, but it's hard to get faster that way when there are weird letter combinations. So I created a spreadsheet and chose an image for every pair. I went for images that are visually strong and all distinct from each other, then used Anki to drill them until my recall is instant. Now my memo forms much more quickly when I read the cube, and I don't stumble over difficult pairs.

I also started using more locations, and I cycle them so I don't confuse memos between adjacent solves. I have several buildings, each of which can be used for four MBLD cubes or 4BLDs, or two 5BLDs, on different floors or in different groups of rooms. So if I'm doing 4BLD practice I have more than enough locations that I don't have to reuse any in the same day, and they all get plenty of practice so they are clear when I come to use them for a big MBLD.

 

[muchacho]

A fast Square-1 solve for me, I'm averaging around 1:20.

Spoiler: Screw method

- Cubeshape

- DL Block (CEC group) - Intuitive
M2: 1,0 / -1,-1 / 0,1

- DR Block (CEC group) - Intuitive
M2: 1,0 / -1,-1 / 0,1

- U Layer Corner Permutation
J: / 3,-3 / 3,0 / -3,0 / 0,3 / -3,0 /
N: / 3,3 / 3,0 / 3,3 / 3,0 / 3,3 /

- UR / UL Edges
M2: 1,0 / -1,-1 / 0,1
M'U2MU2: / 3,0 / 0,3 / 1,0 / 0,-1 / 0,1 / -1,-3 / -3,0 /

- M Slice
M2: 1,0 / -1,-1 / 0,1
M'U2MU2: / 3,0 / 0,3 / 1,0 / 0,-1 / 0,1 / -1,-3 / -3,0 /

- Parity
UOpp: / 3,3 / -1,0 / 2,-4 / 4,-2 / 0,-2 / -4,2 / 1,-5 / 3,0 / 3,3 / 3,0

 

[h2f]

Same questions to you.

Follow Mark's advice. There's nothing more. I've noticed a little progress in last few weeks. What I did? First, a I've rebuilt my memo scheme I mean word/images and I've found a better ones (vivid) for a letter pairs I havent got good word. Second, I've practiced algs and I've stopped to care about efficiency. There's a lot dnf series, but who cares.

@h2f Grzegorz, great solve mate with a wow memo!, so, so motivating.

Thanks.

 

[mark49152]

Your memo is looking good, Grzegorz. 24 seconds is getting fast. Mine has gone backwards. I did a 1:13 mo7 on video after breakfast and the memo averaged 36, a few second slower than two months ago. My execution has got faster though.

An example for Ken @kbrune: the letter pair OK is obvious... except that "okay" is hard to visualise and thus easy to forget when you come back to it during execution. So I replaced it with okra.

 

[Shaky Hands]

So apparently Marty has had a delivery of Qiyi WuHua 6x6's. I'll be hovering about the store waiting this evening. Really want to put a lot of effort into 4x4 and 6x6 for Birmingham Open and yes and of course now 3BLD too

OK, so my Qiyi 6x6 has arrived. First turns are very stiff and I may have underestimated the number of solves I did on the Yuxin 6x6. I've started to work some lube in. No pops or lockups after 30 minutes of solving.

Anyone else have the Qiyi 6x6 yet?