Speedcubing
"now let's see how fast we can go!"
Armed with the basic ability to solve the cube I became interested in the speedcubing websites and watching the amazing speedsolving videos on YouTube. I was at first merely an observer - an interested outsider with only a vague interest in improving my rather slow solve times. In late September 2008 I decided to get involved by joining the speedsolving forums and chatting with other cubers. I was most encouraged by the friendly responses to my introductory post and I found good advice which was undoubtedly inmproving my times.
At this point my 3x3x3 solve technique involved a laborious layer-by-layer approach: -
- make a white cross from "The Daisy" (a simple cross building technique)
- insert white corners
- complete 2nd layer by placing edges - using a muscle-memory algorithm I no longer recall!
- make yellow cross by flipping last layer edges - using algorithm: F R U R' U' F'
- reposition last layer edges with Sune: R U R' U R U2 R'
- move last layer corners into the correct positions - using an algorithm to swap three corners: U R U' L' U R' U' L
- rotate last layer corners to complete the cube - using algorithm R'D'RD
I quickly learned that in order to reduce my times and become consistent, I needed to adopt a solving method such as "Fridrich", "Petrus", or "Roux" that reduced the number of discrete steps by combining elements and introducing sets of algorithms that could be trained for consistent speed. I chose the more popular Fridrich, or "CFOP" method which has 4 stages: -
- C - Cross: make a cross on the bottom layer of the cube
- F - First 2 Layers (aka "F2L") : complete the bottom two layers of the cube (without wrecking the cross!) by efficiently placing the 4 corner-edge pairs
- O - Orient the Last Layer (aka "OLL") : orient the top layer pieces (without wrecking the first two layers!) so all the pices have the top face colour facing up -- 57 algorithms required
- P - Permute the Last Layer (aka "PLL") : rearrange the top layer pieces (without wrecking the first two layers!) so as to complete the cube -- 21 algorithms required
I soon discovered that I have a severe limitation in learning algorithms: I just don't seem to be able to remember them like other speedcubers can! I therefore decided to follow an easy (or "easier" :) ) introductory path to the Fridrich method: -
- learn the Intuitive F2L: i.e. how to perform F2L without any algorithms.
- learn the "2-Look OLL": a slower two-stage OLL process reducing the number of algorithms
- learn the "2-Look PLL": a slower two-stage PLL process reducing the number of algorithms
In October 2008 I learned basic intuitive F2L and most of the 2 look OLL and PLL from tutorials by badmephisto (http://uk.youtube.com/user/badmephisto) which allowed me to almost reach my goal: when it came to the UK Open on November the 8th 2008 I posted an official average time of 1:01.73 -- I was almost satisfied!
In an official World Cubing Association competition like this, the average of 5 timed solves is calculated by discarding the fastest and slowest times and taking the mean of the remaining three.
My times were 1:08.03 58.04 1:00.71 52.99 1:06.44 -- Best time 52.99, worst time 1:08.03.
...to be continued...
Cube Lube
One of the first steps recommended for improving twisty puzzle solve times is to reduce the friction on the sliding faces with a suitable lubricant. Back in the early 80's we used to use furniture polish to lubricate a Rubik's Cube but I was wary of ruining my new cube: there's plenty of warnings on speedcubing websites about which lubricants are suitable for cubes, so I waited until I could get my hands on some silicone spray from a local Maplin store (N71AN Silicone Grease with PTFE 200ml spray). This made a significant difference to the cube if not to my solve times!
More recently I've been using a few different Silicone spray lubricants: one that is proving to be very good is from UK LIDL stores for about 99p labelled "W5 Silicone Oil" and although the aerosol is very powerful (it blows cubies around!) it doesn't take a long time to dry like the Maplin silicon and the vapour isn't so nasty. This feature makes it a good one to spray directly into puzzles as it is more silicon than solvent propellant.
Recording solve times
In order to track my progress I started using a timer. I chose Rubik's JTimer (aka JNetCube), a Java application that runs well on my Linux machines as well as Windows XP at work. More recently I've been using my tiny Asus EeePC 701 with CCT (Caltech Cube Timer), again a cross-platform Java application which is amazing. Get CCT here: http://cube.garron.us/cct/ .
DIY 3x3x3
With the Rubik's "store-bought" 3x3x3, the sprung centres are riveted on so there's no scope for adjusting the range of play. A "DIY" type cube has centres attached to the core by screws that can be adjusted to you liking. The ability to "cut corners" whilst performing sequences of moves, i.e. being able to turn a face when an adjacent face is not perfectly aligned, can greatly improve the rate at which these moves are performed. If the amount of play is too great, i.e. the cube is too loose, then there is a greater risk of a "pop" where the corner and/or edge pieces become detached from the puzzle!
After much deliberation I chose my first DIY 3x3x3: I ordered a black Competition Speedcube from the puzl website as part of a package deal with a speed Megaminx. When my parcel arrived I was delighted - such a difference! My times dropped significantly and I could usually solve in under a minute as long as I made few mistakes! I set myself a target: to attend the UK Open 2008 competition and post an official one minute average!
...to be continued...