Older cubers discussions

[mark49152]

If in understand this correctly this must be done at the end, when the buffer already holds the correct corner (but maybe twisted or not, depending on how many other twisted corners and directions remain), right?

Yes it's best done at the end. I once made the mistake of trying to twist corners first. But that also (usually) twists the buffer so then the first OP corner solved is wrongly oriented.

Also I now try to memo twisted corners as two targets and do this trick if I remember during exec. If I forget and execute as two targets, at least it works. For a long time, I used visual memo for twists and my main reason for DNFs was forgetting twists or getting them wrong.

 

[Selkie]

After all this Sq1 talk I did an avg of 50.

Very impressive times, can I ask what method you use. Well the method is obvious, what sub steps and how many algs you know?

Wow, like 40 more pages have been added since I last posted. I don't want to lower the level of conversation too much, so suffice it to say my collection has grown a bit after deliveries from Cubicle and Cubezz. The Cubicle delivery contained a Fisher cube

I was inactive most of yesterday cooking a roast dinner for mothers day and came on yesterday evening to three pages. A very active community which I think is a great thing. When I started speed cubing 6 years ago I was left wondering how many older solvers there were. Marcel started an invaluable resource and place of collaboration here :tu

 

[mafergut]

Yes, I do it in the end. And almost always I twist coner and a buffer. But the main thing is that every rotation inverses the twist. I mean if you need to twist DFL and and the at the F (I mean in my color scheme white is at FDL) I need to do setup F. This means that instead doing U R etc. I start with RU (inverse of previous). This rule is also applied to x/y/z rotations. I dont know if my explanation is clear?

For UFL twist you dont need setup if the buffer is UBL.

Yeah, for UFL you don't need setup with UBL buffer, that's why I said that the "setup" couldn't be easier I think after a couple practice examples I got how this works and it's really nice but what I'm not sure is what you mean by every rotation inverses the twist. If you think of the twist to be either clockwise or counter-clockwise it doesn't matter how you move the corner around the cube, it will still need to be twisted cc or ccw as originally. Also it's funny that the inverse can be done as you suggest as I had just inverted literaly, giving
L' (R U R' U')x2 L (U R U' R')x2
instead of (U R U' R')x2 L' (R U R' U')x2 L
both work equally well

Yes it's best done at the end. I once made the mistake of trying to twist corners first. But that also (usually) twists the buffer so then the first OP corner solved is wrongly oriented.

Also I now try to memo twisted corners as two targets and do this trick if I remember during exec. If I forget and execute as two targets, at least it works. For a long time, I used visual memo for twists and my main reason for DNFs was forgetting twists or getting them wrong.

For now I also memo twists and flips visually but I assume for multi that's not an option. Problem with two-target memo is that then you have to think during execution whether that was a cw or ccw twist in order to do the trick right.

Also, if you have two twisted corners I assume you can do a setup to put them both at UBL and UFL and just fix both of them with one alg. That would be a great improvement over 4 Y-perms As this alg does preserve the rest of the cube (in contrast with Y-perm) I assume you don't have to be careful to preserve the buffer corner and adjacent edges to use this. I will try to setup an example case with 2 corners this afternoon.

Thanks a lot!

 

[Shaky Hands]

Wow, like 40 more pages have been added since I last posted.

Indeed, I've had another weekend away and it's been good to catch up. Congratulations to everyone on their recent achievements!

It took me 800 solves to get to 1:30, then I was at 1:20 by 1700 solves, then hit 1:13 at about 2700 and have been plateaued there since. So projecting that, it will probably take 5000+ solves to reach 1:00. Then again, I applied a similar prediction to 5x5 and it was wrong.

I've only got a decent number of timed solved for 3x3 but thought I'd share my progression of PB's on it to date:

Spoiler

1. 4m 24s
11. 3m 6s
21. 2m 29s
39. 1m 59s
51. 1m 53s
70. 1m 37s
126. 1m 34s
129. 1m 26s
131. 1m 14s
175. 1m 6s
430. 1m 4s
431. 52s
577. 48s
754. 43s
778. 42s
1030. 41s
1057. 35s
1098. 33s
1297. 29s
1431. 27s
2251. 25s
2379. 23s
2940. 21s
5425. 20s
7450. 19s

Despite getting that 19s PB during my 5x5 binge, my times ended up going up (averaging high 31.xx instead of low 29.xx,) so have been trying to re-drill 3x3 recently in hope of graduating the Race to Sub-30 this week. Re-tensioned the cube last night and it dropped back down a bit so hopefully that will help. Deadline for the Race is today, so will let you know how it goes.

Cheers.

 

[h2f]

Yeah, for UFL you don't need setup with UBL buffer, that's why I said that the "setup" couldn't be easier I think after a couple practice examples I got how this works and it's really nice but what I'm not sure is what you mean by every rotation inverses the twist. If you think of the twist to be either clockwise or counter-clockwise it doesn't matter how you move the corner around the cube, it will still need to be twisted cc or ccw as originally. Also it's funny that the inverse can be done as you suggest as I had just inverted literaly, giving
L' (R U R' U')x2 L (U R U' R')x2
instead of (U R U' R')x2 L' (R U R' U')x2 L
both work equally well

Yes you're right. The thing is that I dont memo them as being CW or CCW but where the white or yellow sticker is. This means if it's in position "to me" I do the alg with UR/UL begining. But If I do a setup or y/x/z I do inverserly. I used to the way I deal with it.

Aobut memeo twists - I do the same - I memo it with one or two letters and it was also a main cause of my dnfs.

 

[mark49152]

Problem with two-target memo is that then you have to think during execution whether that was a cw or ccw twist in order to do the trick right.

I have a simple trick for that - if my pair is in alphabetical order it's clockwise, if counter-alphabetical then ccw. For example PK/KP, BQ/QB, DI/ID. I always use the same pairs/words so don't need to think about it any more. I know a bush in my scene means UBR clockwise.

Also, if you have two twisted corners I assume you can do a setup to put them both at UBL and UFL and just fix both of them with one alg.

Usually (unless it requires too much thinking) I'll rotate to get one of the corners in buffer position then set up the other. No need to undo a rotation .

 

[mafergut]

Yes you're right. The thing is that I dont memo them as being CW or CCW but where the white or yellow sticker is. This means if it's in position "to me" I do the alg with UR/UL begining. But If I do a setup or y/x/z I do inverserly. I used to the way I deal with it.

Aobut memeo twists - I do the same - I memo it with one or two letters and it was also a main cause of my dnfs.

Okay, I got it now, so every quarter turn would make you swap the alg and a half turn would make it remain the same. So, for the corner at the usual OP Y-perm target (DFR) you would do an F2 setup and do the same alg.

I have a simple trick for that - if my pair is in alphabetical order it's clockwise, if counter-alphabetical then ccw. For example PK/KP, BQ/QB, DI/ID. I always use the same pairs/words so don't need to think about it any more. I know a bush in my scene means UBR clockwise.

Usually (unless it requires too much thinking) I'll rotate to get one of the corners in buffer position then set up the other. No need to undo a rotation .

Nice trick!! I didn't think about that. Regarding using the same words every time I need to start improving that. It's the cause I tend to forget my memo because some letter pairs are very difficult to make a word for them, even using a mixture of Spanish and English. So BQ is a bush for you? You just decided that randomly or where is the relation of a bush with a Q that I can't see? I would have thought of maybe a BBQ

And, if you are doing corners last yes, you don't need to undo a rotation. If you are doing edges last then you have to Well, if the corners are at DFR and DBR then with a z2 you can setup both. That would be very convenient but I'd probably have done B2 F2

 

[h2f]

Okay, I got it now, so every quarter turn would make you swap the alg and a half turn would make it remain the same. So, for the corner at the usual OP Y-perm target (DFR) you would do an F2 setup and do the same alg.

Exactly! I must found something more usefull to memo that.

 

[EvilGnome6]

@Mike D: Great job on all the PBs and I note you've edged past me in the elderly rankings for 4x4 average, which gives me a bit of extra motivation for my April comp

You had me excited but I don't think my 1:17.77 quite beats your 1:17.69. I'm right on your heels, though, and the game is on.

It looks like my biggest gain in the elderly rankings was with the 2:15.60 5x5 single.

 

[Logiqx]

I'll update the over 40's rankings when the results are in the WCA export.

Please let me know when your results have been visible on the WCA site for 24 hours.

I will update the over 40's thread at some time after your 24 hour notification(s).

 

[muchacho]

The Yuxin 5x5 arrived this morning, sooner than I expected, but just in time, no more megaminx for now since I was finally able to get a sub-5 ao12: 4:50.50

Not that I have solved it already but the Yuxin feels good, and the colors of the stickers are fine for me, I don't need to change them, which is unusual.

 

[mark49152]

You had me excited but I don't think my 1:17.77 quite beats your 1:17.69. I'm right on your heels, though, and the game is on.

There you go, an elderly moment. I misremembered my PB as 1:17.9x. Still, having you on my heels is some motivation!

 

[newtonbase]

Thanks for all the corner twisting advice everyone.

 

[Lid]

Very impressive times, can I ask what method you use. Well the method is obvious, what sub steps and how many algs you know?

As you can see from the reconstruction just standard Vandenbergh, and I do each step in one look, which means full EP also. Plus some random tricks here and there. Currently planning to learn CP+Parity. Most of the algs are on my webpage.

Here you can practice CS: http://mzrg.com/js/cubeshape_trainer.html

edit: I know it's alot of algs ... but I had so far 7 years with the Sq-1 & the QiYi had made it fun solve once again.

 

[Selkie]

@Stefan - Great stuff. Not sure I would learn full EP, maybe in the future but definitely will get straight on more optimised Cube Shape and practicing bottom later EP as I currently switch layers. Thanks for the links :tu

@Race to Sub 1 4x4 Guys

Well I spent some time yesterday changing my method slightly. For some reason with Yau I had already put cross on the bottom to solve last cross edge and it threw my look ahead which is not good on 4x4 anyway. I have also been trying to persevere with my AoSu mini as I can turn faster with it. Just got the following:-

Average of 5: 1:04.80 (First 5 solves)
Average of 12: 1:08.63
Single 55.29 Personal Best
First time I have ever had a counting sub-1 in an average of 5 too

Spoiler

1. 1:09.31 r' F2 U' r' u L f2 r' D F' B2 L2 F' U2 f2 u R' U' f2 r u r2 U D2 r2 u2 B' f2 D' r2 D' f' u B D2 L2 U2 u' B' F2
2. 58.69 R2 F2 B' R2 U L2 f u' B' F f2 L D u r' R L f' u U' r' u2 F2 B' L' D' L f u' F2 u L R B R2 f2 r2 R L2 B
3. 1:12.75 u L2 U B L2 R u2 L2 B' r B' F2 r' R2 f2 r2 L' R f' F' r F L2 u' U' r u2 B' f' D2 u F' D2 L2 D' R' u' f' R L
4. (55.29) F2 U2 F r D u f2 B' r U2 B D2 R' U' L' u R' u2 U' r F2 B' D' u' r U2 F U' R2 D2 F B2 u F' u U' D B' f2 F
5. 1:06.41 F' u r L' D F2 f2 r' L2 f' D2 R2 F2 u' U2 B2 u U B L' B2 f R2 f D' L f' B' U' L u F R' f u2 U2 f2 u2 f F'
6. 1:10.75 F' f' U' F' L2 f B' L2 f2 U2 F2 U' D r2 U' F r L2 D' L2 D F U' F' f2 U' B' U2 D2 f2 D2 F2 r' R f2 r D u2 r2 L2
7. 1:04.70+ F f L2 f' F2 R D U R' F' u2 U2 R L2 r2 u U2 D2 R2 D L' r2 F' D' U u' r2 R D' u2 U B L' r' U' L2 u F D2 f
8. (1:27.64) f' B' U2 B L2 D' R r' U B2 F2 L' f2 B r' L2 D2 r' f2 D2 r2 L2 U u F' L2 f' R u B' U' R' U r' u' R2 F' D2 u F'
9. 1:10.21 R' F f2 r2 B2 F U' R F f D' u' U2 B D r u' F f U' L2 B2 L2 u R D2 R2 L' f R2 U2 F R' f r' F2 f2 D2 B D2
10. 1:14.37 F' D r' F2 R2 r' D' U' f r' f2 F B D' u2 B2 U' F' r2 D2 r' B' L D' F2 f r2 F f' R' D2 F2 U' D' F2 D' r2 U' u2 f2
11. 1:11.06 F' D2 f2 u' L u' D f2 U D' f' D' B2 L' F' B2 D' R F u f R' L2 r F r F' D R U2 D f' L2 U2 u f2 B2 U2 f D2
12. 1:08.05 B D2 B U r2 R F2 U2 R' r f' L B2 L' u L' R' U2 F2 D U2 L2 F2 L R' f' L R' f2 r D' B2 f2 u' R' L D U' f2 U

I did film this so if I can edit the video to be small enough I will upload it and edit this post to embed.

 

[mafergut]

I did film this so if I can edit the video to be small enough I will upload it and edit this post to embed.

If the full Ao12 is too big I for one would like to watch the Ao5 or the 55.29 single, at least.

 

[Selkie]

If the full Ao12 is too big I for one would like to watch the Ao5 or the 55.29 single, at least.

Just managed it but had to kill the scrambling and most of the inspection Even with the scrambling at 8x it was too long. The scrambles are in the credits at the end though. Uploading now

 

[Jason Green]

Just managed it but had to kill the scrambling and most of the inspection Even with the scrambling at 8x it was too long. The scrambles are in the credits at the end though. Uploading now

What is "too long"? YouTube used to limit upload length but doesn't seem to for me anymore.

 

[Selkie]

What is "too long"? YouTube used to limit upload length but doesn't seem to for me anymore.

Didn't even realise there wasn't a limit anymore. I'll put it down to a senior citizen moment

 

[Shaky Hands]

Didn't even realise there wasn't a limit anymore. I'll put it down to a senior citizen moment

https://www.youtube.com/watch?v=teKD0aAtGo0

Excellent stuff Chris. Really good. Did you edit out an exclamation of happiness at your PB?