Okay, so I just tried this "experiment" where I record a controlled session that I solve normally and a session where I stop for about a second (I think may have stopped for 2 seconds once or twice). Since it's an experiment, I've decided to organize it like one (terribly, I might add) to make it easier to read and so if anyone else wants to try it. Note that this is with the Fridrich method of solving. The Petrus method may have different variables associated with it than the ones I noted below. Also note that I'm terrible at writing up lab reports, so expect to see some very messy organization and rules of thumb below.
Intro: The purpose of the experiment is to record the influence of Rubik's Cube solving times that stopping for a second has on the cross-F2l transition.
Materials:
- Rubik's Cube
- JNetCube timer
- Microsoft Excel
Procedure:
1. Solve a session of twelve solves using the JNetCube timer as you normally would (without stopping for a second on the cross-F2L transition if you don't need to). This is the "control" session, so pay attention and stick as close to the conditions below as possible (or alter it, but stay consistent).
2. Reset the recorded times on JNetCube
3. Solve a session of twelve solves using the JNetCube timer while stopping for a second on the cross-F2L transition, even if you don't need to. This is the "variable" session. Try the keep the solves in accordance with the conditions below.
4. Discard the fastest and slowest times and calculate the average again as well as the standard deviation in Excel. (Since I think JNetCube takes your fastest and slowest times into consideration when calculating your average.)
Conditions of the sessions:
- 15 second inspection time
- Discard ALL lucky cases
- If you know all of 3 or 2-look LL, then only use that. If you only know all of 3-look LL, then do not use shortcuts that will leave you with 2-look LL. Otherwise, discard all instances of a 2-look LL if only know 3-look LL and got lucky with a 2-look.
- If you know 4-7 look LL, then only use that. No shortcuts to make it 2-look or 3-look LL (Not the best way to get a controlled session). Again, discard all instances of 3-look or 2-look LL if you only know 4-7 look and got lucky.
- If you POP, discard it.
- If you lose track of an algorithm and will have to sacrifice about 5-10 seconds to solve the cube, discard the time.
*If one of the above conditions are not met and so the times are discarded, solve the cube again in place of that solve while adhering to the conditions above. Overall, just stay as consistent as possible.
Data:
- Control:
-- Times: 30.63, 34.36, 38.80, 32.02, 33.20, 34.84, (26.76), (40.62), 38.19, 32.70, 33.41, 35.44
-- Average (discarding fastest and slowest solves): 34.36
-- Standard Deviation (discarding fastest and slowest solves): 2.59
- Variable:
-- Times: 37.14, 38.48, 34.25, 33.31, 29.39, 39.70, 35.55, (41.22), 36.34, 39.62, 37.27, (25.98)
-- Average (discarding fastest and slowest solves): 35.11
-- Standard Deviatoin: (discarding fastest and slowest solves): 3.47
Conclusion:
I'm terrible at writing conclusions, so just read my casual comments below.
All right. So my control was faster than my variable. This either means that my scrambles were easier in the control, I generally did better, or I usually do wait one second. On the variable, I may be spending too much time on the pause, my scrambles weren't as easy, or I just did worse. However, there isn't much of a difference in times. I don't think a one second difference is that big in my speed range, so I think the consistent sub-20 people should give it a try.
