• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

Ideas for Slice Move +2, with a Fair Judging System for Awkward Slices!

OreKehStrah

Member
Joined
May 24, 2019
Messages
1,435
YouTube
Visit Channel
Hey everyone!
I've had some ideas for how to make slice moves not be an auto-dnf that I wanted to share!
Also, before we start discussing, I'd like to remind everyone that we are just sharing ideas here, and to not be rude, since I know people will have differing opinions.

I made a video to go along with my idea explanation.

The only weird things I can think of is a partial slice being a DNF, while a double slice isn't, but I see it as no different from U2 being safe while certain partial turns aren't.

If nothing else, I'd like case 1 slices where the cube is fully cubic, with the only misalignment being by some full slice move being a special cases for a +2 instead of a DNF.

I'd love to hear what you guys think!
 
Joined
Sep 15, 2017
Messages
672
WCA
2017BRYA06
YouTube
Visit Channel
This isn’t necessarily a bad idea, but the way the wca counts moves is with obtm (outer block turn metric, basically just htm with a different name) and this way of counting moves makes slice moves count as 2 moves (R L’, U D’ etc...) which is why is is a dnf not a plus 2. Implementing this would complicate everything as it would technically be allowing a 2 move case to be treated as a 1 move case, whereas other 2 move cases would be considered 2 moves. This would also complicate the scrambling program because (due to a reg) you can get a 2 move scramble (technically but it will never happen due to its rarity) so would we count a M slice as 1 or 2 moves in that case? Lastly a U2 is considered 1 move in OBTM which is why it’s a plus 2 not a dnf. Also as a roux user, I rarely get M slice dnfs, I actually get way more +2s than M slice dnfs, so I don’t think roux is at a disadvantage.
 

Kit Clement

Premium Member
Joined
Aug 25, 2008
Messages
1,631
Location
Aurora, IL
WCA
2008CLEM01
YouTube
Visit Channel
This video really makes me want to channel my inner Stefan Pochmann.

You're not even considering any real difficult case here. Every single case you have shown in your video involves the outer layers being perfectly aligned relative to each other, and honestly, if we lived in a world where every cube had that happen, then slice moves would be easy to judge being within 45 degrees. Not sure why you need to put up cards to determine whether a slice is closer to solved or unsolved, it's pretty easy to see.

Additionally, it's quite easy to determine the line between 45 degrees on outer turns. There are tough calls, but it's actually physically impossible for it to be perfectly 45 degrees due to the measurement being on a continuous scale (and because irrationals are dense in the real numbers). Might be too close to call by the human eye, but in theory there's always a correct call for a given alignment. In practice, what appears to be exactly 45 degrees is called solved.

On a semi-related note, if you look at the cube that you applied R L to in the video, you could potentially make an argument that you could make a legal 45 degree turn of the slice to get it to a point where there are two legal outer misalignments (45 degrees or under), making the cube solved.

To get back to my other point, there's absolutely no consistent, and easily understood way to deal with cases where the outer two layers have any misalignment. You have no consistent point of reference. Make a 30 degree Uw move and then a 60 degree U' move to see an example where it is impossible to pick a baseline to compare the middle slice to, and each of the outer layers gives you a different ruling. There are surely ways to come up with rules, but getting them to be easily understood and applied consistently is a major challenge that is likely not to be overcome.

To illustrate that, I've made several points over the years about how people's "intuitive" rules for when a cube should be solved, +2, or DNF break down for consistency purposes pretty quickly. Here's an example of that, and I guarantee you that people will interpret this cube incorrectly even based on their own rules for slice-move misalignments.


Let's see if anyone takes the bait for the third time - should this be +2, DNF, or solved? Be sure to explain your reasoning for what the ruling is.
 

cubeshepherd

Member
Joined
Feb 18, 2018
Messages
2,258
WCA
2016STEE01
Let's see if anyone takes the bait for the third time - should this be +2, DNF, or solved? Be sure to explain your reasoning for what the ruling is.
I think you should add this question to a Q&A forum for possible delegates or to a judging tutorial, just to make sure that everyone knows what something like this should be counted as. Especially since not everyone gets it right all the time.

It's a great example though.
 

Kit Clement

Premium Member
Joined
Aug 25, 2008
Messages
1,631
Location
Aurora, IL
WCA
2008CLEM01
YouTube
Visit Channel
+2 becuz the black U layer aligns with the E move layer over 45 degrees. Therefore plus two. I may be blind though

What are your envisioned rules for determining solved/+2/DNF with slice move misalignments? It could be whatever you want it to be where we're changing rules to allow for slice moves.

I think you should add this question to a Q&A forum for possible delegates or to a judging tutorial, just to make sure that everyone knows what something like this should be counted as. Especially since not everyone gets it right all the time.

It's a great example though.

Under actual WCA rules, this is pretty easily seen as a DNF. It's only difficult to interpret with slice misalignments.
 
Joined
Sep 15, 2017
Messages
672
WCA
2017BRYA06
YouTube
Visit Channel
What are your envisioned rules for determining solved/+2/DNF with slice move misalignments? It could be whatever you want it to be where we're changing rules to allow for slice moves.



Under actual WCA rules, this is pretty easily seen as a DNF. It's only difficult to interpret with slice misalignments.
I think if this were to be implemented you would follow slice rules if the U/D layers are within 45 degrees of eachother (aka disregard inner slice and tread U/D as like a 2x2 kinda) if UD layers are over 45 degrees apart, and both layers are over 45 degrees away from the slice, it’s a dnf, else check if 1 or more layers are 45 degrees off from the slice, if so +2.
 
Last edited:

cubeshepherd

Member
Joined
Feb 18, 2018
Messages
2,258
WCA
2016STEE01
Under actual WCA rules, this is pretty easily seen as a DNF. It's only difficult to interpret with slice misalignments.
True, but do all delegates know that as a certain and act on that each time something like that comes up? (which I know is not that common). If not I think that should be something to make compeltly clear across the board to avoid getting different opinions. Granted I know that they can give provisional extras and address the issue after the competition if needed, but I do not think all delegates will act that way at each competitions. They will instead (and please correct me if I am wrong) act/judge in a manner that they think is correct in their eyes even if a different delegate thinks opposite of them and not even worry about mentioning that in the competition report since they see no reason to think is a "issue".
 

Kit Clement

Premium Member
Joined
Aug 25, 2008
Messages
1,631
Location
Aurora, IL
WCA
2008CLEM01
YouTube
Visit Channel
I think if this were to be implemented you would follow slice rules if the U/D layers are within 45 degrees of eachother (aka disregard inner slice and tread U/D as like a 2x2 kinda) if UD layers are over 45 degrees apart, and both layers are over 45 degrees away from the slice, it’s a dnf, else check if 1 or more layers are 45 degrees off from the slice, if so +2.

When the two outer layers are still within 45 degrees of each other, which one is used as the baseline for determining the alignment of the slice?
 

OreKehStrah

Member
Joined
May 24, 2019
Messages
1,435
YouTube
Visit Channel
I think as cubes get better, and algs progress, it wouldn’t be surprising to see more algs that use slice moves including e and a slices. As a result, I think as a community we should try to find some way to handle and possible state a cube to end in as far as alignment goes that is fair and universally agreed upon.
Because the WCA counts slices as two turns, like M2 is a R2 and L2. I personally don’t think this is fair since you can consider any outer layer turn to be equal to a wide move, which also moves two layers. A wide move is counted as one move and a U2 is counted as one move because it’s executed as one move. As a result, I think slices should also be counted as one move since they are also executed as one move.
I think what Tipster suggested above about using 2x2 rules on the outer layer, then checking the slice is a great idea and could be applied to big cubes as well. For 3x3, a potential end state check could be check the outer layers 2x2 style. If that DNFs then the solve DNFs, else check the slice layer in between, and use some sort of check to compare how the slice is relative to the outer sides to determine +2,DNF, or solved

When the two outer layers are still within 45 degrees of each other, which one is used as the baseline for determining the alignment of the slice?
This is definitely a challenge to agree on here. I just spent a few minutes thinking about it and here’s what I thought of:
Pick a color that isn’t on the flat sides of the outer layers/ is misaligned. Then, Look at the middle edges of the outer layers that have the color you chose. Next check the slice layer and see which of the two outer layers the chosen color’s center is closest to being aligned with the middle edge, and use that side as the reference.
 

kubesolver

Premium Member
Joined
Nov 15, 2019
Messages
425
I realy don't think there are hard cases. The easiest and IMO fair rule to apply would be to first "round the cube" to the closest cube shape (by using similar reasoning like with determininig what is the current cube position for +2's) and then then apply whatever penalty rule we find more fair. There aren't really any tough corner cases this way.
 
Joined
Sep 15, 2017
Messages
672
WCA
2017BRYA06
YouTube
Visit Channel
@Kit Clement this should help explain any questions

8916A92F-85CC-4140-821F-A351AFC9F430.png

Anyway I think it’s possible to implement this, but we shouldn’t because it would complicate everything and be unfair to all the results that came before the rule. Plus the “round to the nearest cube” method and this method can give completely different results further complicating it. As a side note wide moves are allowed because Rw2 is literally just L2.
 
Last edited:

Pyjam

Premium Member
Joined
Oct 8, 2010
Messages
2,142
Location
La Baule, France
1) If any two parallel layers are misaligned by more than 90° : DNF.
2) If any two parallel layers are misaligned by more than 45° : +2
3) Other cases : Solved.

Does it work?
 

xyzzy

Member
Joined
Dec 24, 2015
Messages
2,877
1) If any two parallel layers are misaligned by more than 90° : DNF.
2) If any two parallel layers are misaligned by more than 45° : +2
3) Other cases : Solved.

Does it work?
A 91° misalignment on one layer would be a DNF by these rules, but only a +2 under our current rules.

@Kit Clement this should help explain any questions
View attachment 11359
anyway I think it’s possible to implement this, but we shouldn’t because it would complicate everything and be unfair to all the results that came before the rule. Plus the “round to the nearest cube” method and this method can give completely different results further complicating it. As a side note wide moves are allowed because Rw2 is literally just L2.
I haven't thought too carefully about this but I find it amusing that your flowchart has all the arrows labelled as "No", even the ones coming out of the Yes boxes!
 

Pyjam

Premium Member
Joined
Oct 8, 2010
Messages
2,142
Location
La Baule, France
A wide move is counted as one move and a U2 is counted as one move because it’s executed as one move.
The reality is even worse: even with a double trigger, U2 is actually executed as 2 moves, unless the finger pushes the layer for half a turn.
I would say R2 is one move, but not U2.
But – I suppose – even the worst cuber does M' as one move.
So, U2 is officially 1 move when actually it is 2, and M' is officially 2 moves when actually it is 1.
I don't get the logic behind it.

A 91° misalignment on one layer would be a DNF by these rules, but only a +2 under our current rules.

Is it +2 up to 135° ?
If so, replace 90° by 135° in my question.
Also, the whole set of rules might be changed.
 

kubesolver

Premium Member
Joined
Nov 15, 2019
Messages
425
I don't get the logic behind it.
Even if you don't like the current rules then the simple logic behind them is actually their strong point.

- if the cube is not solved, but can be solved by turning one face of the cube it's considered +2 otherwise it's a dnf.
 

LNBFilms

Member
Joined
Feb 5, 2020
Messages
268
Location
In Honours Agebra I (Sucks just like Chem) DND!
WCA
2019BATE05
I think it should still be a DNF, because they (WCA) would have to change from outer block turn metric, which also affects FMC. So, there is my stance. By the way, if they do decide to make it a +2, they should make the way they count moves, as ”Inner and Outer Block Turn Metric System”, or ”IOBTMS”, and add ”M” related moves into FMC.


---------------------------------------
Co-Founder of WeCubeTeam
---------------------------------------
Lukas Batema
---------------------------------------
 
Top