[Help Thread] CFOP Discussion

[pipkiksass]

I'm aware of some because they are pretty obvious but I don't want whoever decides to watch my video to be biased, so if you spot a weakness in my solves please don't hesitate to tell me.

You're about the same speed as me, so maybe you don't want my advice !

anyway, at the end of solve 2, you AUF then cube rotate before PLL. Doing just one of the two would have saved you half a second easily.

You do a lot of y/y'/y2 (there's definitely a y2 in solve 3) during F2L. Could probably be eliminated/reduced by learning how to solve to back slots. In the 2nd solve you lose your lookahead and tilt the cube up to look at the back, when you could have worked out what was there during the previous pair by process of elimination.

Your cross is good in some solves, less good in others. I was going to say use inspection better, but the solve where you use most inspection (4) is one of the worst in terms of time. This ISN'T because you didn't inspect though, it's because you lose your lookahead.

In the 21.58, you do whole-hand turns for the D2 in the A perm. Have you tried left-handed D2 double-flick? I find it pretty awkward, but REALLY fast. Just a little thing - it's quicker than a whole-hand/palm turn if you can do it.

I think you have many of the same issues as me, and are round about the same time. I know that you know what you need to work on, but I can highly recommend doing what I intend to do to move forwards: either slow solves or metronome solves, where you concentrate on ALWAYS knowing where your next F2L pair is. I mean 100% of the time, not 80, 90, or even 99%. I won't be happy until my lookahead NEVER fails. Generally, your lookahead seems pretty good, your solves flow fairly nicely (apart from the y turns!), but it's clear that the sup-20 solves are the ones where lookahead slips a little.

 

[rj]

I recorded myself doing an average of 12 in order to be aware of all or most of my flaws by getting people better than me watch me solve. I'm aware of some because they are pretty obvious but I don't want whoever decides to watch my video to be biased, so if you spot a weakness in my solves please don't hesitate to tell me.

You rotate too much. That's pretty much your only problem. But, then again, I'm not faster than you.

 

[Morgan McDonagh]

Help with CFOP

Hi I'm a ten year old cuber my single is 48 seconds I use beginners method with some tweaked algorithms I picked up along the way, 48 is terrible so I want to master Cfop but ALL the how to guides I've looked at are really hard to 'get' could you please reply how you learned e.g sites tricks anything!!:-D
Thanks

 

[jeff081692]

I recorded myself doing an average of 12 in order to be aware of all or most of my flaws by getting people better than me watch me solve. I'm aware of some because they are pretty obvious but I don't want whoever decides to watch my video to be biased, so if you spot a weakness in my solves please don't hesitate to tell me.

~snip~

SoKeT, during F2L are you rotating to find pieces or are the rotations a part of your solution to the F2L cases you are doing? If you finish a F2L case and have to rotate to find the next pair then that is a clear sign you are not looking ahead. If you struggle to find pairs without rotating then practice without a timer and try to locate pairs without rotating and if a situation calls for a rotation then do it but the moment you rotate you should know exactly what moves you are about to perform and not have pauses when you do it.

You could also benefit from learning cases from different angles as others have said but just know this doesn't mean you have to strive for no rotations all the time but be able to justify your rotations to yourself as not being wasteful. Also a y2 is almost always a wasteful rotation during F2L.

 

[cuberman7744]

Hi I'm a ten year old cuber my single is 48 seconds I use beginners method with some tweaked algorithms I picked up along the way, 48 is terrible so I want to master Cfop but ALL the how to guides I've looked at are really hard to 'get' could you please reply how you learned e.g sites tricks anything!!:-D
Thanks

Try to watch cyoubx's videos on practice and the why your not sub-xx video

 

[SoKeT]

You're about the same speed as me, so maybe you don't want my advice !

That doesn't mean you can't recognize flaws on other cubers

anyway, at the end of solve 2, you AUF then cube rotate before PLL. Doing just one of the two would have saved you half a second easily.

This is due to my bad PLL recognizion I should AUF to execute the algorithms and then AUF to finish the cube, but as you said, I do a cube rotation instead, which is slower. I practice my recognizion by going slow during OLL and just looking at two faces until I know what PLL that is, although sometimes it is necessary to look at a 3rd one.

You do a lot of y/y'/y2 (there's definitely a y2 in solve 3) during F2L. Could probably be eliminated/reduced by learning how to solve to back slots. In the 2nd solve you lose your lookahead and tilt the cube up to look at the back, when you could have worked out what was there during the previous pair by process of elimination.

The y2 is actually to finish the cross because I didn't keep track of the last edge during inspection, I know how to insert from different angles but my look ahead could definitely use some improvement.

Your cross is good in some solves, less good in others. I was going to say use inspection better, but the solve where you use most inspection (4) is one of the worst in terms of time. This ISN'T because you didn't inspect though, it's because you lose your lookahead.

I need to practice doing the cross blindfolded, but I think the reason it is not good as it could be is because I switched to doing it on the bottom about 2 weeks ago after doing it on top for ages.

In the 21.58, you do whole-hand turns for the D2 in the A perm. Have you tried left-handed D2 double-flick? I find it pretty awkward, but REALLY fast. Just a little thing - it's quicker than a whole-hand/palm turn if you can do it.

I do the D2's in both A perms with my right hand because I find it easier and faster, I tried doing it with my left hand and it feels like that hand is not mine, as if it was another guy's hand. Hard to explain, it's really awkward.

I think you have many of the same issues as me, and are round about the same time. I know that you know what you need to work on, but I can highly recommend doing what I intend to do to move forwards: either slow solves or metronome solves, where you concentrate on ALWAYS knowing where your next F2L pair is. I mean 100% of the time, not 80, 90, or even 99%. I won't be happy until my lookahead NEVER fails. Generally, your lookahead seems pretty good, your solves flow fairly nicely (apart from the y turns!), but it's clear that the sup-20 solves are the ones where lookahead slips a little.

I practice a lot with a metronome, I look ahead and the solves at 100 bpm flow really nicely, the problem is when I time myself, I completely forget about looking ahead. I know it'll come with more practice because almost 3 years ago I used to average ~16 seconds and my look ahead was really good, but I had to do 4LLL and cross on top so that slowed down my solves quite a bit. So yeah, I need to practice my look ahead everyday, even though is really boring, but I know for a fact it does help a lot.

---

You rotate too much. That's pretty much your only problem. But, then again, I'm not faster than you.

I wish that was my only problem, but yeah it is one of the most noticeable ones.

---

SoKeT, during F2L are you rotating to find pieces or are the rotations a part of your solution to the F2L cases you are doing? If you finish a F2L case and have to rotate to find the next pair then that is a clear sign you are not looking ahead. If you struggle to find pairs without rotating then practice without a timer and try to locate pairs without rotating and if a situation calls for a rotation then do it but the moment you rotate you should know exactly what moves you are about to perform and not have pauses when you do it.

You could also benefit from learning cases from different angles as others have said but just know this doesn't mean you have to strive for no rotations all the time but be able to justify your rotations to yourself as not being wasteful. Also a y2 is almost always a wasteful rotation during F2L.

I can do about every F2L case from all angles, the problem lies in my look ahead. I barely do cube rotations when going slow, and my look ahead is nice at 100 bpm, but I just forget to even try looking ahead at all under the pressure of a timer. I know I need to practice a lot more, maybe something extreme like only solving the cube slowly during a week or a month is a good idea, which reminds me of the hyperbolic time chamber

But seriously, I need to improve on basically everything, solving the cross during inspection, looking ahead, learning the remaining 31 OLLs or at least finish learning the painfully slow dots which sometimes take 3 algorithms to solve, and PLL recognizion. And I know I can get sub 15 averages because my PB (11.98) and some other 13's were full step, with flawless looking ahead, so if I can do that consistently, that will be awesome.

Thanks for your input peers.

 

[pipkiksass]

I practice a lot with a metronome, I look ahead and the solves at 100 bpm flow really nicely, the problem is when I time myself, I completely forget about looking ahead.

See, this is your problem. Let's say 25-35 moves of your solve are f2l, so at 100bpm you're looking at 15-21 seconds.

Try upping the speed of the metronome gradually. If you get to 120bpm that's 12.5-17.5 seconds for f2l.

You can't expect to consistently solve at >3TPS (assuming 60 move average, that's sub 20), when your metronome practice is much slower than this.

I'd highly recommend pushing the metronome solves up by a few bpm each week. This will help massively with the consistency of your solves, even if it doesn't give you immediate sub-10s. When you do get a nice f2l and easy last layer, you may get low teens even turning at 2TPS.

 

[mark49152]

I did lots of metronome practice and eventually concluded that it was interfering with my progress rather than helping. The reason? Metronome forces a pace on you, but doesn't train the skills you need to keep that pace.

The skills you need are (1) being able to insert every pair from muscle memory without watching or thinking, and (2) being able to find and track the next pair while doing so. Metronome doesn't help either of these things. All it does it test you. I also found that trying to lock my moves to a metronome pace was making it harder to do them purely from muscle memory, meaning more thinking about the current pair and less looking ahead.

IMHO, blind inserts and blind tracking are more helpful ways of practicing lookahead.

 

[pipkiksass]

I did lots of metronome practice and eventually concluded that it was interfering with my progress rather than helping. The reason? Metronome forces a pace on you, but doesn't train the skills you need to keep that pace.

The skills you need are (1) being able to insert every pair from muscle memory without watching or thinking, and (2) being able to find and track the next pair while doing so. Metronome doesn't help either of these things. All it does it test you. I also found that trying to lock my moves to a metronome pace was making it harder to do them purely from muscle memory, meaning more thinking about the current pair and less looking ahead.

IMHO, blind inserts and blind tracking are more helpful ways of practicing lookahead.

I agree in part; however I'd dispute that ALL that metronome training does is test lookahead.

Metronome training requires you to already be able to do the inserts blind and on autopilot (Mark's skill #1) - if you can't it's pointless. If you can, on the other hand, it helps train lookahead by giving you more time to look around the cube, and forcing you to make use of that time to locate your next pair (Mark's skill #2).

So in a way it tests skill 1 while teaching skill 2. The challenging part of metronome practice shouldn't be solving the current pair, it should be finding the next. If you're finding solving the current pair hard, metronome practice is attempting to test a skill you don't yet have, which is pointless and counter-productive.

Personally, I find slow solving more useful than metronome, as I'm not locked in to a specific pace, but can still force myself to be as smooth and consistent as possible, without forced pauses interfering with my muscle-memory.

Variety, as they say, is the spice of life: don't practice any one thing too much, and try a range of practice methods. No one thing will work for everyone, but with enough trial and error, you'll find something that works for you.

 

[SoKeT]

I did lots of metronome practice and eventually concluded that it was interfering with my progress rather than helping. The reason? Metronome forces a pace on you, but doesn't train the skills you need to keep that pace.

The skills you need are (1) being able to insert every pair from muscle memory without watching or thinking, and (2) being able to find and track the next pair while doing so. Metronome doesn't help either of these things. All it does it test you. I also found that trying to lock my moves to a metronome pace was making it harder to do them purely from muscle memory, meaning more thinking about the current pair and less looking ahead.

IMHO, blind inserts and blind tracking are more helpful ways of practicing lookahead.

I can insert any F2L case on any given position without looking, so that's good. How would you recommend practicing F2L then? And may I ask what your times are?

 

[pipkiksass]

I can insert any F2L case on any given position without looking, so that's good. How would you recommend practicing F2L then? And may I ask what your times are?

I'd recommend what I'm doing now, which is BLD cross, followed by smooth f2l, concentrating on watching the next pair 100% of the time. I find it really useful. I'd try Mark's recommendation of blind tracking a well, so spot your next pair, and insert BLD, trying to track where the next pair will be in your head. If you can already do blind inserts, you don't need to practice that.

I currently average 19, with an appalling (3-5s!) cross and poor lookahead. Unfortunately my life is too busy for me to practice - I know exactly what I need to do and how, but just don't have the time to do it!

 

[SoKeT]

I'd recommend what I'm doing now, which is BLD cross, followed by smooth f2l, concentrating on watching the next pair 100% of the time. I find it really useful. I'd try Mark's recommendation of blind tracking a well, so spot your next pair, and insert BLD, trying to track where the next pair will be in your head.

That sounds really hard but I will give it a try.

 

[mark49152]

I can insert any F2L case on any given position without looking, so that's good. How would you recommend practicing F2L then? And may I ask what your times are?

Slow because I'm old and still learning. Pip already summarised blind tracking pretty well. Over the last few weeks I've been doing lots of that, and it's improved my F2L from a jerky ~16 to a much more fluid ~12, so I believe it works and should help faster people too.

To make practice more regimented, I keep a score - for each of the first three pairs, 3 points for tracking both pieces, 2 points if one wrong, 1 point for losing both, and 0 if I get the insert wrong even if tracking was successful. For the last pair, 1 point for successful blind insert. So each solve gets a score out of 10, and I do 10 at a time and track my progress. At the moment iI'm getting around 70/100.

 

[SoKeT]

That is unnecessarily complicated in my opinion. Why try to track the next pair without looking when you can just look at the cube while you insert the previous one? Ideally, if I get no pauses, the result would be the same, but the process is much easier. I think I will just go slow, with or without metronome, and keep practicing tracking the next pieces.

 

[jeff081692]

That sounds really hard but I will give it a try.

That is great that it sounds hard because that shows you will improve a lot when you master it.

As for tracking a pair without looking, this would help allow you to turn faster having confidence in where pieces would end up during F2L solves. The general thing is predicting > tracking > no look ahead. This is why we try to see at least the full cross in inspection and not 3 pieces and track the last one.

You could practice without the blind tracking stuff but as you get faster you might see the benefit more clearly and I encourage you to try it if you hit a wall with just tracking normally.

 

[mark49152]

That is unnecessarily complicated in my opinion. Why try to track the next pair without looking when you can just look at the cube while you insert the previous one?

The purpose is to develop your blind insert skills to the point that you need zero concentration to do it. Just closing your eyes and inserting a pair is easy if you allow yourself to track the inserted pieces mentally, and you won't fully break the habit of tracking them when your eyes are open. Just try blind tracking and see how much harder it is.

 

[SoKeT]

That is great that it sounds hard because that shows you will improve a lot when you master it.

As for tracking a pair without looking, this would help allow you to turn faster having confidence in where pieces would end up during F2L solves. The general thing is predicting > tracking > no look ahead. This is why we try to see at least the full cross in inspection and not 3 pieces and track the last one.

You could practice without the blind tracking stuff but as you get faster you might see the benefit more clearly and I encourage you to try it if you hit a wall with just tracking normally.

Makes sense, but it would be a lot of effort to master that. I would say you learn that when you're trying to be sub 10, but it's not really needed to be sub 15.

The purpose is to develop your blind insert skills to the point that you need zero concentration to do it. Just closing your eyes and inserting a pair is easy if you allow yourself to track the inserted pieces mentally, and you won't fully break the habit of tracking them when your eyes are open. Just try blind tracking and see how much harder it is.

It really depends on the case, sometimes when you insert a pair the next one doesn't move at all, but sometimes both pieces move a lot and to be able to know where they will end up seems like it would take a ton of effort for not that great of an improvement (compared to looking ahead fluidly, not to my current solves). There are easier ways to get faster. It's like learning full OLL when you do F2L in 25 seconds; you may solve the cube 3 seconds or so faster on average, but if you spent that time getting better at F2L you would solve the cube faster overall. I may be wrong and it might not be as hard as I'm making it out to be, but that's just how I feel.

 

[mark49152]

It really depends on the case, sometimes when you insert a pair the next one doesn't move at all, but sometimes both pieces move a lot and to be able to know where they will end up seems like it would take a ton of effort for not that great of an improvement (compared to looking ahead fluidly, not to my current solves).

The improvement doesn't come directly from knowing where those pieces will end up, although that probably helps. The improvement comes indirectly, from developing a generally better ability to find and track pieces while inserting other pieces. As Jeff said, the fact it's hard is reason to do it, not to avoid it. If you think about it, how can you expect to achieve expert lookahead for as long as this exercise seems hard to you?

There are easier ways to get faster. It's like learning full OLL when you do F2L in 25 seconds; you may solve the cube 3 seconds or so faster on average, but if you spent that time getting better at F2L you would solve the cube faster overall.

That's exactly why F2L lookahead training is worth doing. There might be easier ways to speed up your F2L, but I don't know what those are. Improvement comes very slowly for me, and I only recommend this because I've tried lots of different practice techniques and this is what proved most effective. Everyone's different and it might not be as effective for you. All I can say is, give it a chance - why not spend an hour practising blind tracking, then do a few solves and see if you notice a difference.

 

[Nilsibert]

My problem at the moment is consistency. Sometimes my look ahead is spot on and I get solves around 14-16. In the video I recently posted, I got a 14.55 even though I messed up my LL(mostly OLL) pretty badly, so it could easily have been faster, maybe sub 13.
When doing averages, I occasionally get 16/17/18 seconds solves, sometimes I get up to over 25. Overall I always average low 20's or 19.xx, about 50:50.

I don't know why this is. Maybe my look ahead is only pretty good when I get lucky and the F2l transitions are pretty nice/it's easy to follow the next pair? Or is it just that I'm still far away from having learned to look ahead and that I just have "kind of an idea how to"?
I'm purposely not including LL in this because I know what I have to improve there; basically recognition, especially PLL.

Another problem is lock ups and fumbling around. I can get an easy OLL and a T-perm already AUFed afterwards, but that doesn't help when I then get "nervous" and do some weird accidental D move in the middle of the alg, or just lock up.

I always thought being just over 20 is the worst, but kinda hanging around "on the fence named 20s" is even more frustrating. It's like I am currently getting sub 20 and I just never arrive.

 

[aspetuck]

Learn F2L algorithms or full OLL/PLL?

Currently I use 2 step OLL/Pll and intuitive F2L. I solve OLL/PLL in a couple of seconds but F2L takes me 10-20. The general vibe I got was that full OLL/PLL was a better use of time than F2L algs, which aren't as necessary, but since I find F2L more challenging should I work on that instead?