Curious slow learner....

DaveyCow

Veteran
Joined
Jun 16, 2011
Messages
418
Location
Oregon
WCA ID
2012WING01
YouTube
Visit Channel
All these people recently saying they got sub 45ish in a few weeks... I'm way above that (meaning that my ave is much greater than 45 :) - not complaining and just enjoying the journey (I love taking my time working on the intricasies of cross and intuitive F2L)! But just curious as to any thoughts on how people do this so fast?
 
Last edited:
Some people are just fast improvers. Even though I average about 27 seconds per solve I still feel really slow. On the Race to Full OLL thread in the Off-Topic board, I always seem to have the slowest times. Many people here have been cubing for a long time, so they have all the knowledge they need. I for one, subbed a minute on my 3rd day with a storebought. Still using that storebought, I subbed 45 about 2 weeks later. That same cube has lasted through my entire 6 months of cubing and I'm now averaging *see sig* times.

You wanna know how I improve?

It starts with a "Prac"
&
It ends with a "tice"
 
It really depends on the way you learn your method or progress to more algorithms. For me, I didn't reach 45 seconds for about 3 months because I had a crappy Transformers storebought and I was using Beginners LBL Method F2l and two look OLL and PLL. Then I switched to a real Rubik's storebought (which I lubed) and Fridrich F2l and Full PLL and I dropped to about 30-35 seconds in a month. If I were to do it again, I would probably get a different cube to start, and also I would have learned F2L right away instead of the insert corner, then insert edge stuff.

EDIT: I forgot, I didn't actually start fingertricking until I averaged about 25 seconds lol. That can help a lot too.
 
Last edited:
Most people who are "fast improvers" simply commit more hours daily to practicing. There's no trick to speedcubing a Rubik's cube. Since it's based on speed after all, you practically want to solve as braindead as possible, so you spend less time actually thinking during your execution. In my opinion, intelligence only determines how quickly you can learn a method. But that's still nothing compared to time itself.
 
I took about 3 years to actually want to learn CFOP and I took a 1 year break in that time, where I would rarely pick up a cube, and now I'm sub-30 working towards sub-20. Just go at your own pace, if you don't feel like learning CFOP there is no rush.
 
I took about 3 years to actually want to learn CFOP and I took a 1 year break in that time, where I would rarely pick up a cube, and now I'm sub-30 working towards sub-20. Just go at your own pace, if you don't feel like learning CFOP there is no rush.

hey nice! what method did you use for those 3 years? (I'm currently using/learning cross, int-F2L, 4LLL)
 
It took me 7 months to switch from intuitive cross/ keyhole/ 4LL to intuitive F2L, 2 look OLL and full PLL. Only just started getting on with full OLL a few days ago. I feel like I wasted alot of time on basic methods though. I was too lazy and chicken to just get on with it I guess ;)...
 
I have been cubing for 1.5-2.5 months now, and I average 28-34 seconds. I have achieved this through ~40 solves a day on minimum, and focusing on learning and practising algorithms for my last layer. I messed around for about 2 weeks on Petrus and for a little while on Roux, though I currently use CFOP with 2 look OLL. Doing that, I think, has greatly improved my understanding of how the cube works. The issues with doing this is that it becomes quite a big thing in your life, and takes up a fair bit of time. Reading up on cube theory and watching lots of videos has also helped, I think. Don't think that you have to do it as fast as I have, just go at your own pace. Something that will help you a lot later on if you start it early is practising your lookahead. That, in case you do not know, is going slowly and turning at a reasonable pace, and watching for the next corner/edge that you intend to slot in. Macky explains this a lot better than I do: http://cubefreak.net/speed/articles/lookahead.html. Make sure to have fun and stay positive.

Enjoy.
 
it took me 1 month to sub45 using the basic LBL. now i average about 20 and still using LBL (with one look for all algs; 1/3 OLL & 2/3 PLL). quite a lot of people don't believe me that i can sub 20 with LBL until they see it themselves. i just don't like to memorise a lot of algs, but to improve my TPS with better and quicker observations.

maybe TPS is one of the facts which determines if you can be fast (i'm fast at many other action games/sports which requires to be fast in observation and clicking stuff), but practice and a lot of algs are always the most important facts to be really fast
 
I improved sort of fast in the beginning.
I think that's because I figured a lot of stuff out by myself.

I figured out the first layer myself(when I learned what a layer is) corners first and probably because of that my first timed solve was 1:15. It took no time at all to get sub minute.
I used the rubik's.com website to learn the second layer and LL

Then I found many ways to improve the ll such as mirroring algs.

After that I saw nakaji's cutex.info and saw all the OLLs and PLLs and realised I could learn a certain group and do what I now know is 2 look.

Then I just slowly learned the OLLs and PLLs and practised.

Then I learned intuitive F2L
Then I learned full fridrich.

Its a bit harder to improve now.
 
I went and looked at my race to sub-30, old posts, and my prisma history, and did some estimates, and I progressed something like this:
1 week: 2:30
2 weeks: 1:20.xx
3 weeks: 1:00.xx
4 weeks: 53.xx
5 weeks: 45.xx
6 weeks: 38.xx
7 weeks: 32.xx
8 weeks: 29.xx
9 weeks: 27.xx
10 weeks: 25.xx
11 weeks: 23.xx
12 weeks: 22.xx
13 weeks: 22.xx
14 weeks: 21.xx
15 weeks: 19.xx
16 weeks: 17.xx

I only progressed so fast because, I practiced waaaay too much in the first 2 months or so. And I think that is exactly what it comes down to, practice. Although doing things like PLL time attacks, and looking-ahead are what has gotten me down to a 16-17 average though. My suggestion, is to just try to be as efficient a possible, which it sounds like you are if you like working out the more difficult cases, and then just keep practicing look ahead, and turn as fast as possible while still being able to look ahead.

And that's were I'm at now...Sorry about posting all that, but I was interested in looking at how I improved. :D
 
Last edited:
I think a lot of is is determined by what method you use. I used Keyhole with a 4 Look Last Layer which usually ended up in repeating Sune too often. I dropped to under a minute in about a month, then to about 45 in probably 1.5-2 months (this was 8 years ago now so my memory is somewhat faded). I practiced probably over an hour a day, but what got me down to consistently sub45 was learning the other OCLLs. I didn't ever really want to learn full OLL or PLL, but after I finally decided to switch to Fridrich (2 years ago now) I also went ahead and learned full PLL then OLL.

Everything depends on how you practice. I once heard a quote from a coach "Practice doesn't make perfect, perfect practice makes perfect." and I agree with that to at least a certain degree. Practicing without focusing on improving specific areas are basically just doing repetitive actions which eventually you will plateau with until you work on making those specific actions more efficient.

If I had it to do all over again, I would have learned the 7 OCLLs right off the bat, then learned CFOP just afterwards. However, in 2003, there weren't all these youtube tutorials and such, I remember seeing Jessica Fridrich's website and not wanting to learn the algs for F2L because I was severely lazy.

In terms of CFOP/Beginners, here is how I would suggest learning the methods:

Beginners Method (First layer, 2nd layer edges, EO, EP, CP, CO; although I much prefer CO then CP)
Beginners Method with 4 Look Last Layer (2LOLL, 2LPLL) - To be honest, I would probably omit this step if it was myself, but that's only because I feel like beginners LL can be fast enough (sub10) with practice, I feel like you'd be saving a lot more time at this point by giving yourself the jump start on F2L.
F2L with 4LLL
F2L with 2LOLL and Full PLL
F2L with Full OLL and Full PLL

Note here how I didn't put "learn this when you get so fast" type of field. I did this because I honestly think it is silly to not learn something until you get to a certain speed. I kept hearing "You don't need full OLL until you're sub20." You also don't need full PLL until then, either. My advice is once you've got all the PLLs memorized and your recall is at least decent, start right in with full OLL. You'll thank yourself if you can already recognize all 57 cases when you're say ~18s and not take a time hit, compared to learning full OLL at ~18s and then you start messing up recall and your times start jumping back up to 22-24.
 
I took about 3 years to actually want to learn CFOP and I took a 1 year break in that time, where I would rarely pick up a cube, and now I'm sub-30 working towards sub-20. Just go at your own pace, if you don't feel like learning CFOP there is no rush.

Wow...that's almost exactly the same as me...

3 years with LBL (didn't even know much about speedsolving)
Learned F2L with a weird variation of OLL/PLL
Quit for 11 months
Started up again last year (which is when I started getting serious about speedsolving)

To the OP:
I probably practiced an hour or two each day to get to sub-45. I actually practice more now, that I did back then. I do about 200 solves each day to get what little improvement I can :)
 
Back
Top