Thanks, Raúl. It's curious that all of you recommend to change events and I was thinking that maybe the cause of my dip in 3x3 performance was the time I have been devoting to 2x2 lately that obviously detracts from 3x3 practice time. Others in the forum have advised me that, at our age, we cannot pretend to be fast at all events and, hence, I should focus on 3x3 and forget completely about 2x2, 4x4... but that's not fun at all as 3x3 then becomes more of a job than an enjoyment.
What I've observed is that lately my TPS is slower, maybe the forced slow turning during F2L to try and build my lookahead is contagious and I don't turn as fast as I used to during LL. Maybe it's just that the whole concept of lookahead needs to sink in until I can start to turn a bit faster. I have realized that my fast times up to now where the result of high TPS and "lucky pair spotting" (like when the pairs keep coming up for you to spot them and you think, wow! great lookahead this solve but the truth is that you didn't find them, they found you, if you know what I mean).
Regarding Yau, I started solving 4x4 with reduction and soon changed to Yau. I'm so slow yet with either of them that I cannot certainly tell but dedge pairing lookahead with pure reduction was a nightmare for me and with Yau it is much, much better. Maybe people with great lookahead can benefit from improved efficiency with reduction. Does anybody know what Feliks uses for 4x4 (I've heard that from 5x5 and up he uses reduction)?
And, lots of luck with your 7x7 and 3BLD practice over the holidays. I'm sure you will be able to shave those 2 minutes and more off your 7x7 times. Happy Chri...., I mean, happy cubing
It can be very frustrating to be on a plateau for a long time or even regress. Trust me this is something I have a lot of experience with. Whilst I am faster than average for my age at a global average of sub 16.5 at 46 years old, this has taken a LOT of practice and work. Oh when I said a LOT I meant a
LOT!
I was sub 19 4 years ago
For me the first single recommendation I could make is to read this post by a UK cuber called Rowan Kinneavy. The post may be 4 1/2 years old now but is still as strong today as it has ever been:-
https://www.speedsolving.com/forum/showthread.php?27870-How-to-practice
Ok, now my 2 cents for what its worth. Practice that consists of just a number of complete solves may work for some but it will not work for all and I put myself in this category. I can do an Ao100 a day and not get any improvement in months. Why? I firmly believe we reach a stage where our solves are more sub conscious than conscious. More reaction than thinking. If at some stage we make a mistake or do not optimise something then we then start repeating these mistakes automatically and we can end up doing 100 solves in a row, all with the same reasons why they are not sub "insert your goal here". There may be some fundamental issues you have which are really easy to address but you are just repeating them and reinforcing them over and over again.
For me the first step is to understand yourself what your weak points are. For this I highly recommend filming some solves and trying to be very critical about your own solves. If you dont have a tripod ask someone to assist you. Look at solves of those who are faster or those you aspire to solve like and try and find the differences between these and the videos of your own solves. I like to watch similar aged solvers, my inspiration comes from Ron Van Bruchem. I like to watch guys on this thread who are of a similar age. I will also conversely watch solves by Breandan Vallance, Rob Yau, Faz etc etc (albeit slow motion). There is something to learn from them all, even at my turn speed.
Is your cross to F2L transition really slow? Is you second roux block slow? Is your F2L turn speed fast but with lots of pauses? Are you using lot of AUFs for recognition of LL algs? Are you getting a lot of lockups? There are almost an infinite number of these question you can be asking yourself.
Once you have your list of "problem" areas and for assistance and transparency I will share my current worst bad points with you with you:-
- Cross - Slow cross to F2L transition and I miss easy X-Cross cases
- My turn speed could be faster even at my age
- I still have a lot of pauses in F2L which I would like to eliminate
- I need to learn 2 face PLL recognition
- I need to hold the cube less tightly when solving
- I need to give concentrated ppractice to algs with D or D' in them which I have most difficulty with
- I need to drill my algorithms more often
- I need to minimise cube rotations even more in F2L stage
The next bit is actual easy, research or practice ways to improve on your worst points in your solves. Do concentrated sessions on these. So for me today I might do an Ao100 or if pressed for time I might do an Ao12 but I will do 15 minutes drilling G perms and find a new alg for an F2L case where I currently rotate. Ideally tomorrow I will do something else.
Keep reassessing your solves. If the practice is working then the weak point may become stronger and you will have new weak points to address.
Even in full practices mix it up a bit. I like to do an Ao12 very very slow, then an Ao12 as fast as my hands will go, then a usual speed average. Even the change of speed pulls you out of that automated solving mode.
It is my honest opinion that this targeted practice not only obviously concentrates your time and effort in the part of the solve that need it but it also gives you time practicing that is not sub conscious, that makes you think rather than just repeating solves.
Hope this helps in some way and good luck
P.S.: Another vote for Yau here on 4x4. I started with reduction and got down to probably about 1:40 quite a few years ago. I now average globally about 1:10 with Yau. I might try Yau on 5 as my 5 times are not great. Another very good alternative for 5 is AVG and I was lucky enough to spend some time with Arnaud Van Galen a few years ago discussing his method at a comp. Worth a look, especially with age with a lower move count but does require a lot of lookahead for maximum results.
