• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

[Help Thread] F2L Intuitive vs. Algorithm Solving

How did you learn your F2L? Intuitively or with Algs?

  • Intuitively

    Votes: 388 88.2%
  • Algorithmic

    Votes: 52 11.8%

  • Total voters
    440

avinashv

Member
Joined
Feb 5, 2009
Messages
4
I've been cubing for a few days, and have gotten my beginner's method solves to sub-1:30 with relative ease. In the effort to further improve my times, I researched speedcubing methods and decided to pursue Fridrich for the time being.

When doing the beginner method and solving the bottom-layer corners, I noticed the base F2L pair cases (such as the R U R' or F' U' F) appearing and started tracking pieces across algorithms. After very minor research, I began trying some F2L methods intuitively--I was warned that my times would, at first, increase (which they did), but the sheer number of moves seems to make the entire procedure useless to me at this point. Given how popular this method is, I'm obviously just a victim of the fact that I've attempted this on maybe 10 solves; i.e., a victim of a lack of experience, which is to be expected.

My question is this: I've seen that there seems to be a general stigma against learning F2L via algorithms in favor of solving it intuitively. While I understand the desire to engage the mind somewhat to add some intellect to what can quickly become mindless recital of algorithms, I truly think that learning the 40-odd algorithms will end up giving me a very intuitive sense of what is going on anyway.

I haven't gotten to the point where I've memorized even 2-look LL algorithms, so despite claims that this question might be premature, I'd appreciate a detailed answer as to why people prefer intuitive F2L over algorithmic-based solving.

I guess I can start this off by throwing a bone towards algorithmic F2L: (1) it has to be far fewer moves overall, (2) given that you're not thinking about tracking pairs all over the cube, looking ahead would then presumably be easier, (3) I imagine I'd see an increase in speed much earlier, and (4) I would argue that you land up "understanding" the cube to the same degree as you would by intuitively figuring F2L out.

As a sidebar, I have to express how awesome this community is. As a programmer, I've been a part of a few small forum/communities such as this one, and they've all been worthwhile time-sucks--I imagine this is going to be much the same!

Thanks.

Edit: I should add--if it wasn't obvious--that the memorization isn't really a big issue for me, and yes, I'll be doing at least 2-look LL before attempting this.
 
Last edited:

rahulkadukar

Member
Joined
Oct 11, 2008
Messages
886
Location
New York
WCA
2009KADU01
YouTube
Visit Channel
The advantage of learning intuitive F2L is that you are able to understand how the Cube actually works. This can be paricularly useful
(i)If you have F2L cases where the edge is in the 2nd Layer in the wrong position.
(ii)You can then learn to use free slots to improve your times further as you would have shorter Algorithms.

But onve you are able to do F2L intuitively thn definitely learn the Algorithms.
 

avinashv

Member
Joined
Feb 5, 2009
Messages
4
The advantage of learning intuitive F2L is that you are able to understand how the Cube actually works. This can be paricularly useful
(i)If you have F2L cases where the edge is in the 2nd Layer in the wrong position.

Hmm, I just can't see how there is that much difference. I fail to see how learning these 41 algorithms doesn't give you a solid intuitive sense as it stands.

(ii)You can then learn to use free slots to improve your times further as you would have shorter Algorithms.

Interesting; that's one of the first pieces of useful information I've heard. I see from the various F2L algorithm listings I've found that they all assume no free slots, which I guess is required so that you can use it in any situation. Would you argue that overall, using a purely intuitive F2L using free slots, for example, is more efficient (either in speed or moves) than vanilla algorithms?
 

Johannes91

Member
Joined
Mar 28, 2006
Messages
1,341
I guess I can start this off by throwing a bone towards algorithmic F2L: (1) it has to be far fewer moves overall
Not necessarily. Many of the common algs aren't optimal, because most speedcubers don't care about move count much, and they also generally assume that you're solving the last pair and have to preserve all the other 3, so there are sometimes better move sequences for the first pairs that use the empty slots. [Edit: Wow, you said exactly the same thing in your reply above.]

Someone who has a good understanding can also spot many special cases where instead of solving the pairs one by one with one alg per case, you can solve two pairs simultaneously, etc..

(2) given that you're not thinking about tracking pairs all over the cube, looking ahead would then presumably be easier, (3) I imagine I'd see an increase in speed much earlier
At first, that's probably true.

(4) I would argue that you land up "understanding" the cube to the same degree as you would by intuitively figuring F2L out.
You'll probably learn to understand the 40 algs, yes. But intuitive F2L is much more than that.

I'll also add that if you want to really understand how to solve the cube efficiently, you'll have to learn other methods, like Petrus or Heise, too. And that'll be easier if you learn F2L on your own instead of memorizing algs.
 
Last edited:

avinashv

Member
Joined
Feb 5, 2009
Messages
4
(4) I would argue that you land up "understanding" the cube to the same degree as you would by intuitively figuring F2L out.
You'll probably learn to understand the 40 algs, yes. But intuitive F2L is much more than that.

It's interesting that you say that, because the only reason I would choose algorithms over intuitive F2L is because I feel I can get the level of intuition required. For obvious reasons, I have zero gauge of what level of intuition is built, and clearly, given the popularity of intuitive F2L, I was missing something. I guess I assumed it was a bandwagon-thing given how everyone praises it over algorithmic learning.

Out of curiosity, what kind of timeframes would you expect to see between the two methods of learning before real speed kicked in? For example, intuitively, F2L might take 3 months to master, whereas algorithmically it takes 3 weeks to master--obviously these are made up numbers, but I'd love to hear an experienced cuber's estimate, even if it is based solely on your experience.
 

rahulkadukar

Member
Joined
Oct 11, 2008
Messages
886
Location
New York
WCA
2009KADU01
YouTube
Visit Channel
Out of curiosity, what kind of timeframes would you expect to see between the two methods of learning before real speed kicked in? For example, intuitively, F2L might take 3 months to master, whereas algorithmically it takes 3 weeks to master--obviously these are made up numbers, but I'd love to hear an experienced cuber's estimate, even if it is based solely on your experience.


I think you can master the F2L algorithms in about 2 weeks but it also depends on how much you practice everyday. I mean with 4 hours a day I was able to do F2L in 1 week and I use F2L optimized for Petrus. Hence I always have 1 Free Slot and half of my Algorithms are only based on R and U.

The entire set contains 53 Algorithms.

The advantage of using Petrus is you can do a 2 Look LL with 28 Algorithms.

The only overhead is aligning the edges which may sometimes take upto 3 seconds and you must know atleast 7 Algorithms (It is better if you know more) :).

Overall you need to know 88 Algorithms and half of these can be done with R and U. So Ultimately I prefer Petrus to Fridrich.
 

Chillum

Member
Joined
Feb 5, 2009
Messages
18
One good thing about learning f2l intuitively is that you don't need to have the cube in a certain starting position before you apply an algorithm. If you understand the situation then you can solve a slot from different directions, or mirror images of that situation, without having to memorise a new sequence of moves each time.

Having that understanding sometimes enables you to come up with creative solutions to 'unique' situations (like using multislotting or split layer insertions or making use of empty slots).

You can also concentrate your algorithm learning powers on the last layer, while still playing around with f2l.

Oh, and if you haven't already done so then check out badmephisto's videos on youtube, they're really good :)
 

JLarsen

Premium Member
Joined
Sep 30, 2008
Messages
1,880
Location
Dover, New Hampshire, USA
WCA
2009LARS03
YouTube
Visit Channel
Out of curiosity, what kind of timeframes would you expect to see between the two methods of learning before real speed kicked in? For example, intuitively, F2L might take 3 months to master, whereas algorithmically it takes 3 weeks to master--obviously these are made up numbers, but I'd love to hear an experienced cuber's estimate, even if it is based solely on your experience.


I think you can master the F2L algorithms in about 2 weeks but it also depends on how much you practice everyday. I mean with 4 hours a day I was able to do F2L in 1 week and I use F2L optimized for Petrus. Hence I always have 1 Free Slot and half of my Algorithms are only based on R and U.

The entire set contains 53 Algorithms.

The advantage of using Petrus is you can do a 2 Look LL with 28 Algorithms.

The only overhead is aligning the edges which may sometimes take upto 3 seconds and you must know atleast 7 Algorithms (It is better if you know more) :).

Overall you need to know 88 Algorithms and half of these can be done with R and U. So Ultimately I prefer Petrus to Fridrich.

Petrus is awesome, I need not know any algs for F2L, and only 28 for 2lll. I also know EJF2L so i know 16 MGLS cases too, but you need advanced blockbuilding techniques to make twisted corner blocks, and more so to make them when they're appropriate. It's REALLY nice once you get used to it though.
 

JL58

Member
Joined
Dec 15, 2008
Messages
127
Out of ignorance I took an opposite direction to F2L: learn all the algorithms. I am now practicing with intuitive, based on the numerous comments that this is the only way I could substantially impore my time (I can't consistently break 30's). I found that the algorithm approach works for speed of execution in spite of the number of moves and the lookahead as my fingers can work on the current pair and my eyes/brain search for the next. The intuitive method is helping in reducing the number of move (using empty slots and dealing with edges in the wrong slot) as well as in allowing solution from almost any angle (it would be too much for me to learn 4 times the algorithms).

This being said I have 2 major problems, that hours (days?) of practice don't seem to improve:
- I can't lookahead using intuitive
- I have the hardest time to decide to go one way or the other (intuitive/algorithm). I don't think fast enough and my mind seems uncapable of looking at a pair with both concepts at the same time.

Oh well, I guess I am stuck in the 30's...
 
N

nitrocan

Guest
There are 41 F2L algoritms, but about 35 of them are intuitive algorithms.
 

Lord Voldemort

He-Who-Must-Not-Be-Named
Joined
Nov 24, 2008
Messages
492
WCA
2009PRAS02
I always felt that after a while, even with intuitive, the cases become like algorithms for me. After you do them so many times, you'll know exactly what to do for every case without having to think about it.

Save your algorithms for the last layer... 57 OLLs are already bad enough (I have 31 down though :) )
 

DarkLightA

Member
Joined
Jan 21, 2009
Messages
21
I think it would be smart to focus on 4-look LL first. Are you still doing the U R U' L' U R' U' L to get the corners in the right spot before excecuting R' U' R U? If so, learn the 4-look LL (2-look OLL, 2-look PLL)
 

JohnnyA

Member
Joined
Dec 19, 2008
Messages
250
Location
London
WCA
2009ADLA01
I personally think it's better to use the algorithms to see how to solve the pair. So, you set up the position and use the algorithm to solve it - i.e. the algs on cubestation. This way, you get a good/optimal solution and you understand why you do it. E.g do R U' R' U2 R U' R' U' to set up an F2L case. The algs you might learn for this case might be U' R U2 R' U R U R' and U F' U F U2 F' U F. If you look at these algs, you will see that the first step is setting up the pairs as efficiently as possible to slot them in. By looking at these concepts, you learn how to do it intuitively but efficiently. Try this one.

Setup: R U' R' U2 L' U L

This gives you the same position, but in the wrong slot. Intuitively, a beginner would put both pieces on the top, set them up, slot ... which is horrible inefficient. If you knew the algs above, you know the concept of "wrong slot" (< also wrong term :p) and you could do L' U' L U2 R U R' which is faaar better than the intuitive way. Of course, intuitive users will work out these tricks eventually, but why work them out yourself when you can use the knowledge of others to see it for yourself?

I think algorithms make it easier to learn lookahead as well. With intuitive you need to think about the two pieces you are working with, when you start anyway. If you use the algorithms you don't need to look at them and instead can look ahead.

I personally used the algorithms to create a pseudo-intuitive F2L and I really don't regret it. I average about 18-20 and have a F2L of 9-12.
 

Weiseguy

Member
Joined
Jan 25, 2009
Messages
55
Location
California/Arizona
YouTube
Visit Channel
That's what I was thinking too, was to learn all of those "bad" or "weird" cases but in the topic I posted, everyone says no! lol

I felt that if I learn an algorithm, it would form an intuitive thing anyway because of the way I learn things. Muscle memory takes over, but also to remember things like Sune, I track the F2L pair around, I don't remember the algorithm. Tracking an F2L pair would be intuitive solving, but I learned it from an algorithm.

My F2L sucks right now, I get around 50s solves with it which was an increase of 10s or so over my LBL method. I am working on more intuitiveness with a metronome.
 

avinashv

Member
Joined
Feb 5, 2009
Messages
4
I personally used the algorithms to create a pseudo-intuitive F2L and I really don't regret it. I average about 18-20 and have a F2L of 9-12.

I like this, and it is what I was going for--enough practice with the algorithms should make them understood to the point that they are essentially intuitive, especially if 35 of the 41 are intuitive for the most part anyway (as said by a previous poster).
 

Johannes91

Member
Joined
Mar 28, 2006
Messages
1,341
... especially if 35 of the 41 are intuitive for the most part anyway (as said by a previous poster).
To someone who doesn't know anything about the cube, 0/41 are intuitive. To someone else, it can easily be 41/41.
 

Alpha Shenron

Member
Joined
Feb 8, 2009
Messages
18
I do both really. I do mostly algorithms but if I find myself in a situation where I can kill 2 birds with 1 stone then I'm going to kill 2 birds with 1 stone. Algorithms are the best if you ask me because you get a feel for how the cube moves while at the same time learning something. Your brain will figure out short cuts to the algorithms eventually.

I learned the LBL method with purely algorithms and I found short cuts (though only found with luck) that had got me 40 second solves.
 

veazer

Member
Joined
May 31, 2009
Messages
116
Location
Virginia, USA
I'm just gonna throw it out there that I learned intuitive F2L via youtube in <45 min... so Alg. take longer to learn, and they're only really useful to me now if all but one corner-edge pair is solved.

[Yes, I know this thread is dormant... idc...]
 
Last edited:

GalPro

Member
Joined
Dec 31, 2008
Messages
98
First of all, you are taking it way too fast, you are cubing for just a few days and want to learn the fridich method?

I'd say that you shoul first work on speed with begginer method, until you get to 1:00 or 1:10.

and Intuitive f2l is better for me, I do have my own tricks that I kind of figured out, how to pair every edge and corner in just a few twists,
 
Top