JohnnyA
Veteran
How could I go about calculating an average move count for the following steps?
Fridrich F2L
EOLine
ZZ F2L
Fridrich F2L
EOLine
ZZ F2L
Move Count
- Thread starter JohnnyA
- Start date
Muesli
Premium Member
Count your turns lots of times, take an average.
Lucas Garron
Premium Member
Write a solver and calculate mean solution length.
But first, think about why you care to know the movecount. Is it because you want to be fast, and you're assuming fewer moves=faster? (Not necessarily a bad assumption)
But first, think about why you care to know the movecount. Is it because you want to be fast, and you're assuming fewer moves=faster? (Not necessarily a bad assumption)
Cride5
Premium Member
Practice each method for at least a year, then do an avg of 100...
EDIT: The average optimal move-count for EOLine is ~6.127 moves.
The problem with this is that the specific order and strategy with which blocks are solved in ZZ is not very easy to determine, and will vary between solvers of different experience. When starting out in ZZ I would generally stick to building a 1x2x2, followed by the 1x1x2. Nowadays I take advantage of 'multiblocking' scenarios more often, meaning that sometimes I'm effectively building an entire 1x2x3 side all at once. The optimal move-count for completing a whole side is usually somewhere around 9 moves, making an optimal F2L using the 'whole side' strategy around 24 moves - far fewer than an the average optimal Fridrich solve. However, if the F2L is broken down into a series of 1x1x2 + 1x2x2 blocks then the move count is higher. In reality the move count a solver achieves will largely depend on their experience with the method. My strategy is somewhere between doing each side as 2 blocks and doing whole sides, and I'd guess that the move count reflects this.
I'd imagine ZZ F2L is roughly comparable to Fridrich in terms of move count. The difference is that F2L is executed with LU and RU-gen algs and no rotations. This is part of the reason I originally adopted the method. My opinion is that its faster, not because its fewer moves, but because the moves are easier to execute and there are no interruptions. However, I can't make a fair comparison with Fridrich because I don't practice it. I think I average ~35 with Fridrich, compared to ~20 with ZZ.
Last edited:
Johannes91
Veteran
- Joined
- Mar 28, 2006
- Messages
- 1,341
The hardest part is deciding what exactly it is you want to know. The way you phrased the question is really vague. Are we talking about a typical speedsolve by a typical sub-15 cuber? Linear fewest moves? FMC? Optimally one substep at a time with no look-ahead? Optimally everything at once?
Fridrich F2L and EOLine are fairly clear, but the substeps of ZZ F2L need to be defined. I personally use many ad-hoc techniques after EOLine.
Getting approximate (or sometimes exact) averages using a computer simulation is easy if you can formulate the question clearly (remember that computers can't think and don't understand what they're doing). The results you get may or may not have any correlation with human solutions, however, so think carefully before drawing any conclusions.
Fridrich F2L and EOLine are fairly clear, but the substeps of ZZ F2L need to be defined. I personally use many ad-hoc techniques after EOLine.
Getting approximate (or sometimes exact) averages using a computer simulation is easy if you can formulate the question clearly (remember that computers can't think and don't understand what they're doing). The results you get may or may not have any correlation with human solutions, however, so think carefully before drawing any conclusions.
Last edited:
JohnnyA
Veteran
I accidentally put this in the wrong section. This should have been in puzzle theory, and is about a theoretical essay I am writing for comparing Fridrich and ZZ mathematically.
