I want to shave a few sec of my time.

As is mentioned in many other threads Fridrich is NOT the only method. There are a great number of other methods out there that are very fast methods. Petrus, Roux and Heise are few of the more popular not Fridrich Methods.
I would highly recommend reading around the web and trying a few of these methods to see what you like best. A quick google search will easily lead you to the sites to learn these methods.
 
What Arget said is important. The most important part of getting faster is NOT learning all ~150 CFOP algs... it's learning to look ahead.

I know that I'm not fast by any stretch of the imagination... but just taking a week to look ahead shaved off 20 seconds from my average. Go slow... look ahead... go slow... look ahead. It REALLY helps A LOT!
 
My suggestion is to do these steps in any order:
1. Practice looking ahead! This is a really important step that can't be stresses enough! Looking ahead cuts recognition time, which is important now but will be even more so later on (If you continue to speedsolve, which I certainly hope so!)

2. Practice. The reason everyone here (and everywhere else, undoubtedly) says this is because it is so integral to a good method. Practice makes perfect!

3. Learn the F2L. This will 'shave' many seconds off your time and doesn't even require a razor or shaving cream! (Learning F2L intuitively isn't any worse that learning algorithms. Well, in my opinion.)

4. Integrate a three-look-last-layer into your method. Basically, learn the PLLs. This set of twenty-one algorithms get the last layer corners AND last layer edges into the right location. Again, shaved seconds without lubricating with shaving cream.

5. Practice.

6. Practice.

7. Practice. (You get the point.)

Remember to get used to any new steps you learn, as taking an extra 10 seconds to recognize the various cases is going to do more harm than good.
 
Back
Top