qqwref wrote:Incidentally, I think an M slice off (or r on a big cube) should be +2, not DNF. The reasoning is the same. I think in all situations where a slice move is misaligned, it was executed as one move, so it should be treated as such.
The reasons for "Slice moves counted as 2 moves"This question has been discussed a lot in the past years and it is a very legitimate question (I personnally "think slice" and often perform such moves physically as one move).
Maybe it is necessary to explain again the reasons why slice moves are counted as 2 moves as if they do not exist.
It is important to know what a move is for different purpose:
1)
Solved state and penalties.
- Misalignement threshold of puzzle parts beyond which at least a move is considered.
- Since we got "+2 penalty" for "1 move away from solved state" and "DNF" for "2 moves away or more", it is important to know what falls into the "1 move" category.
2)
"Fewest moves".
Using the "Half Turn Metric" ("Half turn"? What a bad name!) is an elegant solution to our problems.
Take any twisty puzzle, it basically is:
-
a volume,
-
cut planes that make parts of the volume free to move.
This is my vision of twisty puzzles. WCA won't ask you to change your way of considering puzzles, but what's interesting for WCA is:
- it is
easier to see when the puzzle is "one move away" from solved state, especially on big cubes, just focus on a cut plane,
- it is
generic (same rule for any twisty puzzles, you only need to set specific limits in article 10f),
- it is
consistent (same rule for penalties and "fewest moves").
I can understand you don't care about genericity and consistency (by the way, is UD' a slice move even on Megaminx?) because you focus on speed cubing and penalties.
But
if you want to count slice moves as 1 move only, I can show you pictures where the cube state is
impossible to judge.
Take a 7x7x7 and apply an "E" move that rotates together layers 2-3-4-5-6.
When looking at the resulting cube, if it was a clean move, you can visually say there's only 1 move applied. But penalty cases in such circumstances rarely are that clean. Firstly, you would have to make groups of compatible adjacent layers. Then, decide if there's only a group misaligned between two others that are aligned.
If, on your 7x7x7, layers 2-3-4-5-6 are vaguely aligned, between layers 1-7 very loosely aligned, you can't decide if it's a slice move.
It's easier to see on a 3x3x3 I admit, but I pretend you can only compare correctly two adjacent parallel layers.------------------------------
No more penalties for misalignments!Now, we would not be talking about this if the penalty for misalignment did not exist.
In my opinion, this penalty is a mistake and bloat, like the extra attempt for pop rule was. When you don't cleanly finish your solve, you deserve a DNF, just like when you stop the timer with a popped piece.
With puzzles that take long, you can afford a check. On fast puzzles, well, it's part of the dexterity challenge! And if your 2x2x2 get misaligned everytime you drop it on the table, because it's too loose and you put too much lubricant inside, blame yourself.
You may consider it would be a bit crual. I think cubers now are mature enough to accept it.
Rules would be more simple, and easier for judges (Bilbao...).
No more questions like:
- Is it fair to have +2 for all puzzles?
- Why don't you allow slice moves?
- Why is U2 not DNF? It's not just a slight misalignement, the cube is clearly missing a full big move!
- Why is a cube lacking (U/2)R2 (orthogonal slices) only a penalty case?
And no need to define a metric for speed cubing anymore. The only metric is time, because it is speed cubing, not "fewest moves".
I can't read the proposal for getting rid of this +2 penalty in the draft. Could WCA explain why it has been rejected?Gilles.
