Analysis — Feliks Zemdegs’ official 4.75
Cubing Diaries #2 — The former 4.73 world record’s little brother can teach you three valuable lessons
Good day (or evening or night)! Welcome to another Cubing Diaries.
In today’s article, we’ll analyze the official 4.75 that our Feliks Zemdegs, keeper of all solves, got in the first round of Camberra Spring 2017.
Even though this solve is relatively old, I still think it deserves to be remebered. At, the time, it didn’t get as much repercussion because it wasn’t a world record. Then, it was the 5th fastest official solve ever, and Feliks’ second fastest.
What makes this solve exceptionally especial is the hypercreative solution found by the prodigy kid to finish the last layer, demonstrated by the low 8.21 TPS in this solve, compared to the average ~9.0 TPS of world class cubers.
The solve is going to start with the green cross, so if you’d rather follow along with your cross color, just put it in the front before scrambling. Here’s the scramble:
F2 R2 D2 R U2 F2 U’ F’ D L B’ F D L2 D’ L2 F U
If you scrambled with the standard orientation, your cube should look like this:
Cross
During inspection, Feliks made an x’ intending, as mentioned, to execute the green cross.
Before we start to understand this solve, we should take a quick glance at a possible beginning that Feliks didn’t catch. If he had executed the green cross with L U R’ F’ Uw’ L’ he’d have a x-cross in six moves with the solved pair in the back left, which is the worse slot to fill, with an easy cancellation into the second pair in the front left slot, the second worse slot to fill. But, in the heat of competition, this didn’t happen, so let’s keep going.
The cross he planned and executed was
U’ U’ R’ F y (D’ U’) U’ R’
It’s worth noting that even though the solution he chose wasn’t a x-cross, he was able to use the cross to prepare the execution of the first pair. With the U2 before the final R’, he guaranteed that the red-yellow pair would end up in an easy three move case, as we’ll see further ahead. He made, if we’re being generous, a little-x-cross.
I also find it interesting to point out that this solve demonstrates beyond doubt that, in some situations, it’s worth it to rotate with y and y’ during the execution of the cross so that your solve is as finger friendly as possible.
First three pairs
As I mentioned before, Feliks made sure, during the cross execution, that the first F2L pair would be as easy as possible. So, with
U y’ U R’ U’ R
he inserted the red-yellow pair in the back.
Note that this y’ is a consequence of the y he made during the execution of the cross, returning to the initial position after inspection. Feliks, knowing where he would need to execute the first pair, still chose to rotate during the cross, which shows the value a comfortable execution has, even if it costs two rotations.
After inserting the first pair, the second pair, white-red, formed in the back right. Therefore, right after, with
U2 R U’ R’
Feliks inserted the formed pair in the front right.
In this case, it would have been more interesting to insert this pair with U’ R U2 R’, because it would made it so that the next pair, orange-yellow, could be easily solved with a three move insertion. This is idea of multislotting, where you solve a pair while influencing the next one, makes your F2L a lot more efficient. As we’ll see further ahead, to fix this pair, Feliks spent four extra moves than he’d need had he seen this possibility.
Obviously it’s not easy to see these kind of solutions on the fly, after inspection, but I felt it was important to mention this detail so that we realize that even the faster solutions can be improved upon.
Moving on, with
y’ U’ R U2' R’ U U L’ U’ L
Feliks fixed the flipped pair, as mentioned, and inserted it in the front left slot.
He was able to efficiently rotate so that the final pair would be solved in the front right slot. Even if the first two pairs were solved in the back right and front right slots, respectively, which is not a good sign for the third and fourth pairs, during the solve made it so that the last slot would be filled in the front right, the ideal position.
Fasten your seatbelt, for now comes the magical part of this solve.
Last slot and last layer
After reaching the fourth and last pair, Feliks had some options. With the corner already in place, he could’ve inserted the edge with U F’ R U R’ U’ R’ F R, preserving the edge orientation, or he could have solved it with beginner’s method, taking the corner out and joining with the edge (😷not), but none of this is good enough.
I believe that when trying to predict the OLL or even the ZBLL, depending on the solution he chose, the king of zeroing realized, after a quick pause, that with
R U’ U’ R’
he’d be able to solve the five remaining edges. Not only that, but two corners would also be put into place. That’d leave just three corners to be solved. And what do we use to solve only three corners? Commutators!
And that’s exactly what he did. He used a commutator to finish the solve. Even if he did an unnecessary move at the start, he was able to quickly come up with the commutator he needed and so with
y’ U U’ R D R’ U2 R D’ R’ U’
he finished the solve, even canceling a move on the AUF.
The fluidity and spontaneity with which Feliks was able to execute this crazy solution are fascinating. Not only was he able to see that he’d be able to solve all the edges easily, but also that two corners would fall into place and that the last three remaining corners would be solvable with a commutator, and all of that in less than ~two seconds.
This solve is a testimony to the flexibility that distinguishes Feliks from many others world class cubers, giving his solutions an elegance rarely found.
Where can you go after you’re able to smoothly incorporate commutators in your solves? TCLL? Full 1LLL? The next step is being able to see the AB3C in the middle of the solve allowing you to cancel into a four move insertion somewhere along the execution. A few years from now, we’ll look back and see how we really knew nothing.
Conclusion
Initially, I thought that Feliks had been able to plan the solve until the third pair in inspection. In a “similiar” situation, in his 4.73 world record, in the previous year, where the solve began with a x-cross and two formed pairs, he stated that he was able to predict until the second pair with certainty and to have an intuition of where and how the third would end up.
However, after comparing the number of moves each solve took to get to the third pair (9 vs 17), I realized that this hypothesis wasn’t that realistic.
What I think actually happened that allowed for this unusual solution to be developed so quickly is that, while solving the third pair and after predicting the execution of the fourth, Feliks started to look ahead even further to predict the OLL or the ZBLL he’d get. Then, he realized the edges would be solved with R U2 R’. The small pause before the final execution could be Feliks confirming that indeed only three corners would remain and that they could be solved with a commutator without any setups.
This solve can teach us a few things.
Firstly, it shows us that your inspection doesn’t need to be perfect, especially in competitions. Even if you can’t find the ideal cross, like the six move x-cross on this scramble, your performance in comps will be similar to the your performance at home if you’re able to consistently plan a decent start, with at least the cross and a first pair planned out. Don’t spend 10 minutes on inspection on every solve at home because you’re trying to always find the optimal cross, but learn to be consistently decent with your inspection.
Secondly, this solve also shows, as mentioned beforehand, that using rotations in the execution of the cross and F2L isn’t necessarily bad, in contrast to the myth floating around that you should aim for a rotationless F2L. Unless you use ZZ, Roux or have eight fingers in each hand, rotations can be a very important tool to guarantee fluidity to your solves. In this case, Feliks rotated four times: to finish the cross with and R’ instead of a B’; to execute the first pair in the back right; to execute the third pair in the front left and to finish the solve with the commutator.
None of these rotations are bad or condemnable. The important point is that they are used moderately, in the exact spot where they are necessary. Rotations are bad when they are used as an excuse for not learning to execute F2L pairs in different angles or when they used in excess. Because of this, learn to rotate efficiently, guaranteeing a comfortable solution and positioning F2L slots conveniently, in the order back left -> front left/back right -> front right, if possible, if you’re predominantly a right hand solver¹.
Finally, this solve demonstrates an ability that every cuber should have, which is being versatile and capable of dealing with different situations in the best way possible. Explore the Rubik’s Cube with the most random methods you can find. Learn a bit of Roux if you’re a CFOPer and vice-versa. Investigate exactly what was the corner first method they used in the 80s. Practice Megaminx and see what it can teach you for 3x3 F2L. Invent a new method!
Perhaps one of the most interesting aspect of the cube is exactly the fact that, after thousands of solves, we can still find new details that didn’t even pass through our heads. Look for the unexpected, constantly searching for new ways to expend your horizons. Try to navigate the cube, finding new solutions to old challenges. Don’t be afraid to adventure into techniques that may seem obscure. Ask yourself: “what if?”. In a word: dare, for you won’t grow while you don’t get out of your comfort zone and learn².
1: There’s some discussion, which I won’t get too much into in this article, about what order should you fill the back right vs. the front left slot in. Just know that, as a rule of thumb, the back left slot should be filled first and the front right slot should be filled last, with some degree of interchangeability between the back right and the front left slots.
2: I should also quickly mention that you shouldn’t really dare in competitions. In them, you should stick to what you already know. Probably, Feliks has played around with this idea many times at home before, and that’s why it came to him so “naturally”. He dared, messed up and finally got it right many times at home before trying it in a comp. Therefore, dare freely at home. In competitions, dare with caution. ❤️
The complete solve, along with the scramble:
F2 R2 D2 R U2 F2 U’ F’ D L B’ F D L2 D’ L2 F U
x’ // Inspection
U’ U’ R’ F y (D’ U’) U’ R’ // Cross
U y’ U R’ U’ R // 1st pair
U2 R U’ R’ // 2nd pair
y’ U’ R U2' R’ U U L’ U’ L // 3rd pair
R U’ U’ R’ y’ U U’ R D R’ U2 R D’ R’ U’ // LSLL
39 moves
8.21 Turns per second
That’s all for today. Thank you for reading.
What do you think of this solve? Could you have executed it faster? Would you prefer it if I had analyzed another one? Hated the article and want to scream at me? Hit me up!
Just DM me on instagram, where you can also always be notified when there’s a new article and be cheered up in that gloomy sunday night with energizing stories
In case you missed my first article, I explained where I want to take this blog and told a little bit about my history with the Rubik’s Cube:
