Older cubers discussions

[Selkie]

Do you use ZZ/Petrus? If not how do you do the EO to make WV useful?

Also at 47... I'd hope you have plenty of finger speed left in you. Think about typing and piano players. They can go for quite a while .

I use CFOP and with some edge control durinh F2L it is not difficult to end up with a case in one out of every 2-3 solves.

 

[Jason Green]

@JanW, thanks. To set scramble lenght: left bracket change from WCA to 3x3x3, right: old style. Next type the scramble length though it's set to 25. Strange thing with your scrambles. I count sometimes F and B moves and I never noticed the things you wrote.

@Shaky Hands - this U perm hurts.

I thought you could do that! I was looking on my phone and couldn't see it.

 

[h2f]

Thanks @h2f! This works. I'll use these settings for my ZZ solves.

Maybe it was just a statistical anomaly that I saw so many scrambles like this when I first started using the app. If the scramble was a series of 20 random moves, the probability of having only one F/F'/B/B' move is about 4.4%. It happened way more often than that in my first 100 solves, but sample size is far too small to make any conclusions. With 25 move scrambles, the probability goes down from 4.4% to 1.5%, which hopefully makes it rare enough that I stop thinking about this whole issue and won't notice it when it happens. I think this was the case with the other timer I used.

For me there's one intresting thing - how solving method affects what you notice. When I was using ZZ for a while I did the same as you - counting F/B moves. When Roux - watching pairs, blocks etc. When CFOP - easy cross. Etc.etc.

 

[Logiqx]

Thanks @h2f! This works. I'll use these settings for my ZZ solves.

Maybe it was just a statistical anomaly that I saw so many scrambles like this when I first started using the app. If the scramble was a series of 20 random moves, the probability of having only one F/F'/B/B' move is about 4.4%. It happened way more often than that in my first 100 solves, but sample size is far too small to make any conclusions. With 25 move scrambles, the probability goes down from 4.4% to 1.5%, which hopefully makes it rare enough that I stop thinking about this whole issue and won't notice it when it happens. I think this was the case with the other timer I used.

It's worth mentioning that by default csTimer generates a random state scramble and gives a near optimal set of moves for it. What you observed regarding eo was most likely down to chance. The timers giving old style 25 move scrambles are just random moves.

A friend was once arguing with me that my scrambling the cube must make it easier. I refuted his claims as it makes no difference to me but with ZZ it may be true. I know I can't help but count the tips during a pyramid scramble so eo could be the same.

My suggestion to you is to toss the cube in the air and scramble from a random orientation. Toss it again before inspection.

Sent from my SM-G930F using Tapatalk

 

[h2f]

@JanW, maybe just don't look at the cube whilst scrambling it? I know I don't.

More cstimer fun (which I've only recently switched to as my browser is having problems with ctimer.co.uk which I've been using for quite a while now)...

I had a scramble earlier from cstimer where I was heading for a great time after managing to pull off a deliberate double X-cross and ended up screwing up an awkward OLL. Finished up on 52.93 seconds, which is one of my 5 highest times in the past 2500 solves. Did it again and got 18.98 seconds which would have been one of my lowest 5 times ever.

Scramble: F' B D2 L' U' L U L' D' R' U2 R' U2 B2 L2 D2 L F2

x2 // inspection
D R F2 R B' U' F' L' U L R' U2 R' // double X-cross
U' B U B' U' R' U' R // 3rd pair
B' U2 B // 4th pair
U2 R B' R' U' R U B U' R' // erm, messed this bit up, got it on 2nd attempt
U y F2 U M' U2 M U F2 // comedy U-perm, feel free to laugh

I reckon I was so happy with myself for pulling off the double X-cross that I had a brain leak!

There's a lot of possibilities in this scramble. My speedsolve:

2 y' // inspection
D' F L U R D2//cross + 2 pairs
y U L' U L //1st
F R' F' R//2nd
U2 L U L' U' L U L' //3rd
R' U2 R//4th
U R' F R U R' U' F' U R //oll (thunder)
R' U2 R' D' R U' R' D R U R U' R' U' R U //Ra perm

51 htm, with cancelation 48.

But adding 3 extra moves to your xxcross makes xxxcross:

x2 // inspection
D R F2 R B' U' F' L' U U R2 U' L R' U2 R' // xxx-cross

Of course I couldnt do it during speesolve.

 

[mafergut]

Random state scrambles are supposed to be better than old style. Maybe you can just scramble in random orientation but that would make it harder for you to reconstruct later if you want to.

EDIT: Sorry, I answered before reading Michael @Logiqx reply

 

[Shaky Hands]

@Shaky Hands, I never look at the cube while scrambling. There's no need to look at it to register those things I mentioned. I'd need to not think about what I'm doing while scrambling, which is easier said than done. Especially now when I became aware of and got annoyed by this issue, it's very hard not to notice if I perform the first half of the scramble and there are no F/F' or B/B' moves. Once I notice that, it's even harder to not get a count of how many there are in the second half of the scramble.

Ah I see. I misunderstood how you were determining stuff in the scramble.

There's a lot of possibilities in this scramble. My speedsolve:

2 y' // inspection
D' F L U R D2//cross + 2 pairs
y U L' U L //1st
F R' F' R//2nd
U2 L U L' U' L U L' //3rd
R' U2 R//4th
U R' F R U R' U' F' U R //oll (thunder)
R' U2 R' D' R U' R' D R U R U' R' U' R U //Ra perm

51 htm, with cancelation 48.

But adding 3 extra moves to your xxcross makes xxxcross:

x2 // inspection
D R F2 R B' U' F' L' U U R2 U' L R' U2 R' // xxx-cross

Of course I couldnt do it during speesolve.

Nice! Thanks for sharing. [/QUOTE]

 

[Logiqx]

And regarding WV, which also @Selkie Chris mentions above, I really have a hard time deciding what set of algs I should go for next. There are so many last slot sub-sets VLS, WV, SV, HLS,... not sure that I didn't make on of those up [emoji14] and there's also COLL...

I've enjoyed learning COLL (L, U, T, Pi, H). I'd recommend it to anyone who is comfortable with full OLL and looking for something new to learn which complements their existing skills. Left to chance you have a 1/8 chance of edges oriented after your last F2L pair (suitable for COLL or WV) but with partial edge control you can orient edges during your final pair and set up for COLL. I'm currently playing with the 5-move VHLS cases which often leads to a nice COLL + EPLL solution.

IMO COLL has the most benefit for OH and big cubes but it's still fun on 3x3 and the frequency of PLL skips after COLL is 6 times higher than normal (often executing a COLL alg which is no slower than the regular OCLL). I think for this reason it is also worth doing during 2H solves. It's especially good for 4x4 since the need to execute an OLL parity alg increases the chances of EO significantly. The way I handle 4x4 parity means that 9/16 solves have edges oriented prior when I begin the LL.

Even if you don't actively use COLL in a solve you'll find the corner recognition sometimes kicks in anyway. I don't always use COLL when I'm timing myself 2H (yet) but I sometimes recognise the COLL case as I start executing the OCLL and know that my corners will be fully solved. This reduces PLL recognition time and improves the flow of that solve.

Edit: I discussed COLL with Mark at the UK champs last year and our view was that COLL probably wasn't worth it for 3x3. I've changed my mind... it probably is worth it.

 

[MarcelP]

Happens to me as well. When slow solving I don't usually get any better than 17-18 seconds. It's only by combining good lookahead and a speedier TPS when I can get down to 13-15. Of course that's only in lucky solves. Most of the time trying to rush too much only ends up in bad lookahead and worse times.

I must say one more thing - a lot of people in a recent time asked me why I have a slow tps.

I've noticed that in my good solves like 15.xx or 16.xx I'm fast and my lookahead is good. In slower I lock up, lose lookahead, bad recognize olls and plls etc.

I have been struggling with breaking my sub 19 barriere. I slow solve in my good averages. With real slow cross + F2L I have my best times in averages. How ever, if I step it up I can very nice solves like this one from today:

Spoiler

First solve of the day, cube GuaGuan yuxiau
Time List:
1. 13.71 L2 U' F2 D F2 U F2 R2 U' B' U' R2 F R B D2 U' R D U2

z2 x //Inspection
L' U x' U L' U L' x' U2 x // CROSS 10
R' U R U' R' U' R // 1ST 7
U' y' R' U R U2 R' U2 R U' R' U R //2ND 13
U' R U2 R' U2 R U' R' //3RD 8
y' U' R U R' U' R U R' //4TH 9
U2 r U2 R' U' R U' r' // OLL 8
U2 R U R' U' R' F R2 U' R' U' R U R' F' // PLL 15
// 70 MOVES IN 13.71 = 5.1 tps

alg.cubing

As you can see by this reconstruction, it is far from efficient, but still a full step solve sub 14. I managed to keep looking ahead in all F2L cases, and between F2L - OLL and between OLL and PLL. Solves like this makes me think I could be sub 19 easily, if I only adjust to higher TPS. But is I practice with higher TPS, my averages go down the drain.. sup 20.. Go figure..

 

[Lid]

Marcel 2nd pair I'd do: y' U2 R' U2 R U R' U' R (8)

 

[Logiqx]

I have been struggling with breaking my sub 19 barriere. I slow solve in my good averages. With real slow cross + F2L I have my best times in averages. How ever, if I step it up I can very nice solves like this one from today:

Spoiler

First solve of the day, cube GuaGuan yuxiau
Time List:
1. 13.71 L2 U' F2 D F2 U F2 R2 U' B' U' R2 F R B D2 U' R D U2

z2 x //Inspection
L' U x' U L' U L' x' U2 x // CROSS 10
R' U R U' R' U' R // 1ST 7
U' y' R' U R U2 R' U2 R U' R' U R //2ND 13
U' R U2 R' U2 R U' R' //3RD 8
y' U' R U R' U' R U R' //4TH 9
U2 r U2 R' U' R U' r' // OLL 8
U2 R U R' U' R' F R2 U' R' U' R U R' F' // PLL 15
// 70 MOVES IN 13.71 = 5.1 tps

alg.cubing

As you can see by this reconstruction, it is far from efficient, but still a full step solve sub 14. I managed to keep looking ahead in all F2L cases, and between F2L - OLL and between OLL and PLL. Solves like this makes me think I could be sub 19 easily, if I only adjust to higher TPS. But is I practice with higher TPS, my averages go down the drain.. sup 20.. Go figure..

Nice solve. Regarding TPS... just remember that turning 5% faster equates to ~1s. When we turn fast we might be doing 25-50% faster which is easy for the fingers but too fast for the eyes and mind. The trick is to only turn 5-10% faster for a nice improvement.

Sent from my SM-G930F using Tapatalk

 

[h2f]

As you can see by this reconstruction, it is far from efficient, but still a full step solve sub 14. I managed to keep looking ahead in all F2L cases, and between F2L - OLL and between OLL and PLL. Solves like this makes me think I could be sub 19 easily, if I only adjust to higher TPS. But is I practice with higher TPS, my averages go down the drain.. sup 20.. Go figure..

Nice solve. Regarding TPS... just remember that turning 5% faster equates to ~1s. When we turn fast we might be doing 25-50% faster which is easy for the fingers but too fast for the eyes and mind. The trick is to only turn 5-10% faster for a nice improvement.

Nice solve Marcel. I think it's kind of forcing my mind to be faster. It's like - stop trying to hit me and hit me. I've noticed it in CS GO - I was aiming slow etc. - just aim and shoot. Wrtining it I think there will come moment when I'm back to slow turning to get better at lookahead.

 

[mafergut]

Nice solve @MarcelP, it's exactly the same feeling I have, I get a good solve or two, I try to have high TPS, times go up. It's like either that one solve was shouting at me pair after pair so I could not go wrong or I just went lucky but continuing on that high TPS makes it clear that, for the majority of solves, it is too fast for my lookahead, so average times are horrible, even sup-20. Maybe I should follow Michael's @Logiqx advice and just progressively increase TPS but it's difficult to find the correct pace to maximize speed without losing lookahead.

By the way, Shengshou is preparing a Master Kilominx, in the same desing style than their recent Kilominx which I'm still waiting to recieve. But the Yuxin 6x6 arrived this morning. It's still in the package waiting for my unboxing later this weekend.

Here is my unboxing video. Sorry because it's like 25 minutes long and still my dizzying 1st person angle. I have to think about getting a good GoPro camera that I can use with a tripod instead of this webcam. Quality of the video is not very good either but I hope you enjoy it.

 

[muchacho]

Good video, nothing needs to be changed.

That yellow of the 6x6 may be good enough for me to tell it apart from green.

I bet (and hope) your son scrambled that cuboid

 

[mafergut]

He did. And I had to look some cuboid algs to be able to solve it.
By the way, the Aurora Mega is even better than I thought. I'm not gonna buy a Galaxy any time soon.

Sent from my Nexus 4 with Tapatalk

 

[earth2dan]

Nice video @mafergut. I skimmed a bit to the cubes I was interested in . I'm in the same boat as you with the 6x6 springs. I'll have to order a second spring set, because there's no way I'm taking them out of my awesome 5x5. Cubicle sells Yuxin 5x5 hardware sets for pretty cheap, I'm sure that'll be in my next order.

I have both the stickered and stickerless AuFu GT. I quite like the shades on the stickerless. They're not the best, but they're definitely good. Needs a darker blue would probably be my biggest complaint, if I was going to complain. The only downside for me is that I cannot seem to get my stickerless to move like my stickered one. The stickered one feels so nice, but I much prefer solving stickerless. I might have to take it apart, clean it, and lube it. But that's a big task...

I just got the dayan ridged mega recently and I'm not really good enough yet to warrant expanding that collection looking for a better one. Question, do you put your mega on the table when doing RU algs while solving? I find myself doing that when I solve. Not sure if that's good practice or if I should work on correcting that and keeping it in my hands.

 

[mafergut]

Nice video @mafergut. I skimmed a bit to the cubes I was interested in . I'm in the same boat as you with the 6x6 springs. I'll have to order a second spring set, because there's no way I'm taking them out of my awesome 5x5. Cubicle sells Yuxin 5x5 hardware sets for pretty cheap, I'm sure that'll be in my next order.

I have both the stickered and stickerless AuFu GT. I quite like the shades on the stickerless. They're not the best, but they're definitely good. Needs a darker blue would probably be my biggest complaint, if I was going to complain. The only downside for me is that I cannot seem to get my stickerless to move like my stickered one. The stickered one feels so nice, but I much prefer solving stickerless. I might have to take it apart, clean it, and lube it. But that's a big task...

I just got the dayan ridged mega recently and I'm not really good enough yet to warrant expanding that collection looking for a better one. Question, do you put your mega on the table when doing RU algs while solving? I find myself doing that when I solve. Not sure if that's good practice or if I should work on correcting that and keeping it in my hands.

I can't blame you for skimming. The video was way too long. Last time I checked on TheCubicle I did not find a Yuxin 5x5 hardware set on sale, maybe I just didn't find it or they just put it later due to overwhelming request from the community

For Megaminx I usually put it on the table for the whole LL part. Up to then I just hold it in my hand. I have an old sub-4 solve in my youtube channel. I hope to get closer to sub-3 with the Aurora

 

[earth2dan]

I can't blame you for skimming. The video was way too long. Last time I checked on TheCubicle I did not find a Yuxin 5x5 hardware set on sale, maybe I just didn't find it or they just put it later due to overwhelming request from the community

For Megaminx I usually put it on the table for the whole LL part. Up to then I just hold it in my hand. I have an old sub-4 solve in my youtube channel. I hope to get closer to sub-3 with the Aurora

Cubicle Yuxin 5x5 springs. I guess it's not a complete hardware set, it's just the springs. That's probably what everyone is after anyway.

That makes me feel a little better, I do pretty much the same with mega. Though, not nearly as fast. I'm still a long ways off 4 minutes

 

[h2f]

I had a nice ao100 in the mornign. Started with 20.30 ao50 and finished with 19.20 ao50 after 80 solves and 19.80 ao100. Nice solves in the end: 13.13 FS and 15.69 but it didnt give nice ao5 due to bad other 3 times. I'm really happy.

Funny fact - I could have been at the comp when FZ beaten WR but thought it's a fake when I heard he will come. After few days I've registered but missed it by few hours.

 

[h2f]

Sorry for double post. But it has passed 22 hours of silence. Next ao100 sub20 (19.82). First ao12 of session was 21.30 and I thought I have a bad day. But next solves were better. After 20 solves I've started to look ahead. I've finished with 20.30 ao50. Next 50 solves were nice. There was a moment when ao50 was 19.07 (new pb!). But specially I'm glad of few nice solves with xcrosses I try to make.

But thing is that I got few times pairs of 15.xx solves and 3 next were bad - over 20 or around. If one or two were 15-16 I could make ao5 sub 17.

Spoiler

stats

solves/total: 100/100

single
best: 12.60
worst: 27.02

mean of 3
current: 18.44 (σ = 1.68)
best: 17.36 (σ = 4.18)

avg of 5
current: 19.02 (σ = 0.77)
best: 17.45 (σ = 2.33)

avg of 12
current: 19.37 (σ = 1.38)
best: 18.48 (σ = 2.17)

avg of 50
current: 19.26 (σ = 1.81)
best: 19.07 (σ = 1.96)

avg of 100
current: 19.82 (σ = 2.00)
best: 19.82 (σ = 2.00)

Average: 19.82 (σ = 2.00)
Mean: 19.86

Time distribution:
12+: 1
14+: 5
16+: 19
18+: 26
20+: 33
22+: 10
24+: 4
26+: 2