Practice Routines and General Cubing Tips

[gsingh]

I wanted to write a post about what I think would be the best practice routines, and how to improve, so here it is.

The General Idea:
50% Deliberate Practice, 50% Solves, but be flexible.

Once you're comfortable with the cube, I would recommend you spend the first half of your practice learning CFOP, in this order: Intuitive F2L, 2 Look PLL, 2 Look OLL. Keep in mind that you should DEFINITELY not learn all of this in one day, but take it step by step.
The second half of your practice should just be doing solves(with cfop ofc, doing more practice with the beginners method will not get you anywhere.).
After you've learned Beginner CFOP, you should take a couple of days just to solve, to get used to the method(can take up to a week). After you are comfortable with CFOP, switch back to the 50/50 practice rule, and watch this video by Lazor0monkey. Take a few days to implement these F2L solutions into your solves. After you are comfortable with these solutions, the next step would be to learn full Pll. I recommend learning 2-5 per day, depending on how much practice time you have, and then drilling them afterward. This is also the point where I recommend starting to plan cross in inspection. After you've finished learning PLL, my advice would actually be to learn psuedo slotting, and spend 1 week just doing solves using psuedo slotting, to get used to it. After this week is over, make a new session called slow solves. In this session, you should be doing slow solves, trying to be more efficient than usual. You should be doing this for 50% of your practice time, and the other 50% would be learning full OLL 25% of the time, and doing normal solves the other 25% of the time. After you've finished full OLL, learn some F2L algorithms for bad cases, and continue doing slow solves, and some normal solves. At this point, your times for slow solves should be around the same as the times in your normal solves session, so you can now stop doing normal solves completely, doing only slow solves, in your normal solves session.
That's it for now, more coming soon!

If you have any ideas for what I should add, don't hesitate to respond to this thread.

 

[Garf]

I can add a part about Megaminx:
The thing about Megaminx is that you can solve it with a method similar to 3x3. Well, as a beginner, that is what it seems like. You can make pairs a little more efficiently than 3x3, though. Some algs you use for 3x3 won't work for Mega either. Like Algorithms with middle-layer moves.
After you get comfortable with megaminx and choose to try speedsolving it, the best method to learn is Westlund for S2L. This S2L method has you make layers on different F2L centers after you build F2L. EX. Solving Yellow, then Purple, then green, then finally red/blue. It's good to have a set order of what you solve when you start, but if you want to be more flexible with Westlund as you get faster, you can. 4LLL is also really good to learn, as it allows you to solve the LL quickly.
After you get Sub-1:30 you need to change your mindset about the puzzle. You don't want to keep thinking about the mega as a 3x3, you want to think about it as its unique puzzle, with solutions way different. You also need to start thinking about block-building instead of making pairs. You could make whole blocks on the puzzle, but even a pair is considered a 2x1 block.
As for practice: you need to spend a lot of time messing around with the puzzle. It's a good idea to grind the algs, as they make you faster. The last S2L crosses also need to be efficient and fast, yet have good ergonomics as well. It is a good idea to do speedsolves as well, but most of your mega grind should be slow-solving.