My reflection on the evolution of F2L style (+ task for the community)

[teller]

I never before noticed how fast that F' can be - I shall use that increasingly now when CFOPing.

I don't see how that's any faster than thumb :/

In the J-perm that starts RUR'F'R..., for example, I find that right index push for F' more naturally leads to R.

For this algorithm I've started to do the F' by pushing the DFR cubie upwards with my index finger because I've practised OH enough with my right hand that it's kinda easy (Just an alternative)

Unfortunately I cannot do this with my left hand very easily, but I guess with enough practice, I might be able to do it...

Wow...I hadn't thought of that. Fascinating...

Ya know, I used crappy Rubik's brand cubes filled with cement for 25 years...I think part of the reason I'm attracted to fingertricks is because I always wanted to do cool flippy things, but the cube just wouldn't do it. Today, with a Sheng En or Haiyan Memory I'm like a kid in a candy store. I had these heavy barbells strapped to my fingers all those years, and now they're off. I should shoot some more video...

 

[Edward]

I never before noticed how fast that F' can be - I shall use that increasingly now when CFOPing.

I don't see how that's any faster than thumb :/

In the J-perm that starts RUR'F'R..., for example, I find that right index push for F' more naturally leads to R.

For this algorithm I've started to do the F' by pushing the DFR cubie upwards with my index finger because I've practised OH enough with my right hand that it's kinda easy (Just an alternative)

Unfortunately I cannot do this with my left hand very easily, but I guess with enough practice, I might be able to do it...

Wow...I hadn't thought of that. Fascinating...

Ya know, I used crappy Rubik's brand cubes filled with cement for 25 years...I think part of the reason I'm attracted to fingertricks is because I always wanted to do cool flippy things, but the cube just wouldn't do it. Today, with a Sheng En or Haiyan Memory I'm like a kid in a candy store. I had these heavy barbells strapped to my fingers all those years, and now they're off. I should shoot some more video...

Yes, please do!
I love your obscure style, with all the cool flippy things and what not.

 

[teller]

I never before noticed how fast that F' can be - I shall use that increasingly now when CFOPing.

I don't see how that's any faster than thumb :/

In the J-perm that starts RUR'F'R..., for example, I find that right index push for F' more naturally leads to R.

For this algorithm I've started to do the F' by pushing the DFR cubie upwards with my index finger because I've practised OH enough with my right hand that it's kinda easy (Just an alternative)

Unfortunately I cannot do this with my left hand very easily, but I guess with enough practice, I might be able to do it...

Wow...I hadn't thought of that. Fascinating...

Ya know, I used crappy Rubik's brand cubes filled with cement for 25 years...I think part of the reason I'm attracted to fingertricks is because I always wanted to do cool flippy things, but the cube just wouldn't do it. Today, with a Sheng En or Haiyan Memory I'm like a kid in a candy store. I had these heavy barbells strapped to my fingers all those years, and now they're off. I should shoot some more video...

Yes, please do!
I love your obscure style, with all the cool flippy things and what not.

Thank you...I will do it...I do have some juicy ones ready.

 

[jms_gears1]

I never before noticed how fast that F' can be - I shall use that increasingly now when CFOPing.

I don't see how that's any faster than thumb :/

In the J-perm that starts RUR'F'R..., for example, I find that right index push for F' more naturally leads to R.

For this algorithm I've started to do the F' by pushing the DFR cubie upwards with my index finger because I've practised OH enough with my right hand that it's kinda easy (Just an alternative)

Unfortunately I cannot do this with my left hand very easily, but I guess with enough practice, I might be able to do it...

Wow...I hadn't thought of that. Fascinating...

Ya know, I used crappy Rubik's brand cubes filled with cement for 25 years...I think part of the reason I'm attracted to fingertricks is because I always wanted to do cool flippy things, but the cube just wouldn't do it. Today, with a Sheng En or Haiyan Memory I'm like a kid in a candy store. I had these heavy barbells strapped to my fingers all those years, and now they're off. I should shoot some more video...

Yes, please do!
I love your obscure style, with all the cool flippy things and what not.

Thank you...I will do it...I do have some juicy ones ready.

i your finger tricks. MOAR VIDS

 

[JTW2007]

"go slow, and look ahead."

Whoa. I seriously don't think I've heard that in years. Didn't it used to be the default answer to "how do I get faster at F2L"?

Anyway, that was very interesting. I'll have to check out that F2L page more in the coming weeks.

 

[jms_gears1]

"go slow, and look ahead."

Whoa. I seriously don't think I've heard that in years. Didn't it used to be the default answer to "how do I get faster at F2L"?

Anyway, that was very interesting. I'll have to check out that F2L page more in the coming weeks.

lolyears.

 

[JTW2007]

"go slow, and look ahead."

Whoa. I seriously don't think I've heard that in years. Didn't it used to be the default answer to "how do I get faster at F2L"?

Anyway, that was very interesting. I'll have to check out that F2L page more in the coming weeks.

lolyears.

Have I just been missing it?

 

[trying-to-speedcube...]

Left index > right index

 

[cmhardw]

Wow, Macky that's a very interesting post! It's always really neat to hear of the perspective of the evolution of speedcubing from one of the masters himself

I think this idea is a very good one, and very interesting. It's as if we are taking the 40 some odd cases for F2L pairs and increasing that to maybe between 100-200 algorithms. Of course the logic behind why each alg works would be understood by the cuber, but each and every case could be executed in a brain-dead way by sheer muscle memory.

I also think it would be a good idea to consider the simpler ZBF2L cases that are either super fast to execute, or conveniently easy to spot while solving. There are perhaps 20-30 algs (Chris Tran, can you confirm or better this?) that are just really neat and fast, or really useful, and would be a worthwhile addition to any cuber's F2L tool kit. Good examples of these are:
R' D' L F' L' D R2 U R'
M' U' M U2 L F' L'
F R U' R' F' L' U' L
F R U2 R' F' L' U' L
R l U' R' U l'

Of course I know not everyone would agree with including these algs in the first place, and even if they could agree on that it's hard to agree on which algs to include. Still, I think that having a couple of ZBF2L "gems" in the toolkit would not hurt for the standard CFOP solver. What do others think about this?

Chris

 

[joey]

macky: *claps* Great post.


Makes me think that I've hardly put any effort into my F2L nor fingertricks.
Lot of people seem to have tricks and nice ways to do algs.. but I just try turn quickly with disregard to execution style.
I think it may be the next step to my improvement.

 

[oskarasbrink]

I've always found Faz's sune execution in this vid pretty interesting. I'm not sure if this is how he normally does it though. Left index push for the first U, and a leftie double flip for the U2.

Forward to about 0:43

no, the first U is right index

 

[rubiknewbie]

Is there anyone in particular who you're referencing when you say "the Japanese cubers"? (Sorry for my ignorance)

This was actually really interesting to read, and when you compared the "modern" style of F2L to OLL/PLL, I started to become curious- am I wrong that the consensus is that intuitive F2L is superior to algorithmic? I've been under the impression that it was.

It is the general consensus but at very advanced level, some cubers like Breandan Vallance and Escher use some very advanced algorithms to avoid cube rotations. Some of them are totally non-intuitive.

 

[joey]

They be non-intuitive.. but they are still easy to follow when you see them.

 

[dada222]

Thanks a lot for the link. It's about time I learned something new!

 

[Dene]

Well I don't know about you guys, but anyone with short fat fingers is not going to be able to F' with right index with any deliberation.

 

[Kirjava]

Well I don't know about you guys, but anyone with short fat fingers is not going to be able to F' with right index with any deliberation.

That means we shouldn't mention or attempt the technique at all.

>_>

 

[keemy]

Well I don't know about you guys, but anyone with short fat fingers is not going to be able to F' with right index with any deliberation.

FALSE.

Also I have been using this as long as I can remember (F' push with right index). But recently someone told me it was bad technique and that I should be using left index w/o re grip to pull down the ULF sticker because pushes are inherently slower than pulls.

 

[Anonymous]

am I wrong that the consensus is that intuitive F2L is superior to algorithmic?

They're essentially the same thing.

That's true for experienced cubers, I guess.

 

[jazzthief81]

It was under such considerations that I set out to compile a collection of rotationless F2L algorithms.

[...]

After adding mirrors and rotations of existing F2L tricks from various sources, I explored variants of existing algorithms for as yet undocumented ones. The result is what should be a fairly comprehensive list that I myself have yet to sift through.

macky

This is a list I compiled a few years ago, when I started practicing solving F2L on the bottom. Maybe it can be of help to you in completing your list. It also gives you an example of how you can organize it:

http://www.cubezone.be/completef2l/index.html

Although I dubbed it "The complete F2L" it's not actually complete. However I did fully explore the following 4 groups of cases:

• Insert pair with edge oriented
• Insert edge with edge oriented
• Insert corner with edge oriented
• Insert corner with edge flipped

Each group has 48 cases because there 3 corner orientations, 4 ways to position the edge/corner/pair and 4 possible target slots.

For each of those 48 cases I may have several solutions listed, some of which may exploit an empty slot. In that case the algorithm starts with the position of the empty slot in between square brackets, eg. [FL].

The trickiest group in the bunch is Insert corner with edge flipped. For the other 3 groups it's easy to come up with good rotationless algorithms since they can be solved with R, U and L moves, but for this group you need to bring F/F' or rotations into the equation. For most cases I tried to come up with a good rotationless algorithm and a good algorithm that uses a rotation.

As for the groups I didn't explore: Insert pair with edge flipped can be reduced to Insert pair with edge oriented by means of a y/y' rotation and the same thing goes for Insert edge with edge flipped. This is also what I do on a solve 95% of the time, although I have thought about which of the rotations, y or y', would lead to the most finger-friendly continuation. For example, I'd rather do yLUL' instead of y'RUR' because then I can keep my hands in "the neutral position", meaning thumb on F and ring/middle/index/fingers on B. I find that this increases flow.

I never got around doing the Move pair cases, although I know the basic ones from all angles, like the ones where the pair is in it's target slot or where the pair is fully matched but in the wrong slot. I think it would be useful to investigate these a bit further in order to have some contingency plan in case there is no F2L piece in the top layer. (I should probably start with reading your Jeff2L page, Macky )

I also have to note that my list is not kept up to date because I printed it out a long time ago and have kept updating the algorithms on the printed sheet ever since.

Just my 2 cents on this topic. I always found optimizing F2L cases one of the most intriguing parts of cubing and I still put quite a bit of effort in to it. It's so much fun!

 

[Edward]

am I wrong that the consensus is that intuitive F2L is superior to algorithmic?

They're essentially the same thing.

That's true for experienced cubers, I guess.

Not an experienced cuber here: Intuitive F2L is basically the same thing as algorithmic FL2, except you don't purposely try to memorise agls, they just kind of end up being memorised by sheer practice.