3x3 Rubiks. How many algs do you need in your solution?

[shimu]

Just want to find out from you guys how many algorithms do you have to memorize for your solution. Also how fast can you complete the cube on average.

I am a newbie and my time is not fantastic. I average about 1min15sec. My solution:
1. First layer (0 algo. By intuition)
2. Middle layer (2 algo)
3. Last layer:
- First, move corner to place (2 algo).
- Flip corners to right orientation (3 algo. Each to flip 2 corners to right orientation. So if 3 or 4 corners are wrong orientation, I need to perform 2 algo. I am too lazy to memorize algo to flip 3 and 4 corners at once).
- move edge to place (2 algo. Each to move 3 edge clockwise or anticlockwise respectively. So if 4 edge is misplaced, I need to perform 2 algo).
- Flip edge to right orientation (1 algo to flip 2 edge. If 4 edge is of wrong orientation, I need to perform the algo 2 times).

So, total algo i need to memorise: 2+2+3+2+1 = 10.

How about you guys?

 

[rubiks to the third]

i dont know what the minimum is, but i can do it with 3. probly a 2 edge flip, sune, and jperm

 

[Johannes91]

It's a rather meaningless question without a proper definition for the word "algorithm" or "algo" or "alg".

 

[malcolm]

You can do it in 0 algs with commutators, this is a decent BLD method

 

[abbracadiabra]

Again, depends upon the definition of algorithm.

According to American Heritage Dictionary, an algorithm is "A step-by-step problem-solving procedure, especially an established, recursive computational procedure for solving a problem in a finite number of steps."

For the purposes of this forum, "move sequence" would be an appropriate definition. The cube cannot be solved without move sequences whether they are memorized or understood intuitively.

Either definition would be inclusive of commutators, and even my own version which is kind of like "move this out of the way, put that in place, move this back."

However, the original poster specified algorithm's which one had to memorize in order to arrive at a solution. In that case, we can both answer his question with "zero".

 

[Kenneth]

But he did not ask what the minimum number of algs needed is. The question is about how many you have memorized and the times you get from doing that.

For me, I don't know exactly, say 50-100 algs that I use to solve several hundreds of cases (and I have to memorize how to use the algos for diffrent cases). The times I get from that are around 25 seconds in the better solves.

But it also depends on the method I'm using at the moment, if I do something first, then I may need a number of algs for the rest, if I do something else instead I may need a diffrent number of algs, maybe not the same algs as in the diffrent situation.

Shimu, I strongly reccomend you to change your LL style. The method you are using is turn intensive, guess your LL are some 50-60 turns or even more.

Rather do:

Flip edges (3 algs)
Twist corners (7 algs)
Place corners (2 algs)
Place edges (4 algs)

As always you can do with less algorithms but then you have to repeat them if there is a "4 piece case".

 

[Johannes91]

However, the original poster specified algorithm's which one had to memorize in order to arrive at a solution.

Indeed, but there are still some unclear cases. For example, many cubers memorize commutators for some OLL cases without even knowing that they are commutators. But if they later learn about them and understand how they work, does that mean that they know less algs than before?

Edit: And what about symmetries? A few years ago, I thought there are 4 different G-perms and memorized an alg for each one. Now I'm comfortable at mirroring/inversing, so I only need to know one of them. I have no idea what I should answer when someone asks how many algs I know.

 

[Swordsman Kirby]

It is possible to do the cube efficiently without memorizing algorithms: Heise method.

 

[shimu]

gee... I guess I must have asked an illogical question...

I guess what I am trying to ask is how many non-intuitive sequence do you need to memorise. (^^; no sure what is commutator and all cos I am pretty an amateur. Perhaps I should go read up more on it.

By the way, is there anyway to solve the cube intuitively? I can only do so for 1st layer and after that, the rest i depend on "algo", which is memorised sequence (that make no sense to me) to move/flip corners or edges without breaking my first layer/pieces that are already in place or orientation.

again, I agree that everyone is different. something that look senseless sequence to me, might make sense to someone. he/she might be able to derive those sequence intuitively... so yes, i have asked a difficult to answer question. so sorry (^^;

 

[shimu]

It is possible to do the cube efficiently without memorizing algorithms: Heise method.

wooo... really? no memorization required? hmm... will check it out. anyway, do you know how fast can such method attain? 30sec? 60sec? 2mins?...

 

[shimu]

However, the original poster specified algorithm's which one had to memorize in order to arrive at a solution.

Indeed, but there are still some unclear cases. For example, many cubers memorize commutators for some OLL cases without even knowing that they are commutators. But if they later learn about them and understand how they work, does that mean that they know less algs than before?

Edit: And what about symmetries? A few years ago, I thought there are 4 different G-perms and memorized an alg for each one. Now I'm comfortable at mirroring/inversing, so I only need to know one of them. I have no idea what I should answer when someone asks how many algs I know.

i think i know what you mean... like if do a simple sequence like (R2U2)x3, it swap 2 edges on the U face and 2 edge on the R face. It is very simple sequence so it is kind of like take "zero" effort to remember it now. but frankly, i still dun know how it works... (^^; It is something i cannot reason and show people step by step how the piece finally got swap.

 

[guusrs]

I define an algorithm as a sequence of moves you memorized, so triggerde by recognition, not logic.
I use about 120 algs while speedcubing and 200 algs while doing FMC
(mirror & inverse not counted)
Gus

 

[hait2]

my method is pretty standard

cross - 0algs
3 f2l pairs - 0algs
4th f2l pair + LL edge orientation - 6 algs (technically 8 but i don't count R U' R' or R' F R F')
LL corner orientation - 7 algs
PLL - 15 algs (most mirrors and inverses i didn't memorize a separate alg for, the one exception being the reverse Gperm)

which totals up to about 28. not too much and this method can get you quite fast times if you practice (then again, so can any method)
i solve somewhere low 30s now, but im severely out of practice. it's probably possible to get this method sub25 at least

 

[alexc]

Let's see, I'm switching to VH from Fridrich. I don't know all the VHF2L or COLL algorithms yet, but here is the count.

Cross-0 (intuition)
3 F2L pairs-pretty much intuitive (used to use algs, but now I am exploring different ways of pairing up c/e pairs, etc.)
4th F2L pair-16 (32 counting mirrors)
COLL-40
EPLL-4
Total-76

 

[Swordsman Kirby]

Let's see, I'm switching to VH from Fridrich. I don't know all the VHF2L or COLL algorithms yet, but here is the count.

Cross-0 (intuition)
3 F2L pairs-pretty much intuitive (used to use algs, but now I am exploring different ways of pairing up c/e pairs, etc.)
4th F2L pair-16 (32 counting mirrors)
COLL-40
EPLL-4
Total-76

Unless you plan on doing ZB at the end, VH is not useful.

 

[masterofthebass]

it is if you're lazy about learning OLL algs... You could also learn COLL cases and get easy PLLs. VH can be useful, but I don't think it's worth it.

 

[Lofty]

I agree learn VH to go to ZB. I stopped with it mainly because it isn't the best OH method. But VH can be useful to learn. COLL is like 1/15 or something like that of ZB and the F2L can give you a good understanding of fixing orientations.

 

[alexc]

Perhaps VH is really not that useful. I don't think I'm going to do ZB anytime soon, so maybe I should stop learning it for now and finish the last 8-10 OLL's I still have to learn. Maybe I could use COLL if the LL edges happen to be oriented on that solve.

 

[shimu]

Let's see, I'm switching to VH from Fridrich. I don't know all the VHF2L or COLL algorithms yet, but here is the count.

Cross-0 (intuition)
3 F2L pairs-pretty much intuitive (used to use algs, but now I am exploring different ways of pairing up c/e pairs, etc.)
4th F2L pair-16 (32 counting mirrors)
COLL-40
EPLL-4
Total-76

!!! gosh... how do u manage to remember so many alg?... (^^;

 

[shimu]

I define an algorithm as a sequence of moves you memorized, so triggerde by recognition, not logic.
I use about 120 algs while speedcubing and 200 algs while doing FMC
(mirror & inverse not counted)
Gus

200... amazing... now i know why you guys can do the cube soooo fast (^^;