White KB
Premium Member
- Joined
- Oct 3, 2017
- Messages
- 1,278
- Location
- Putting Runza on the map since 2022
- WCA ID
- 2019BRUC01
- YouTube
- Visit Channel
Misscrambles: They're a problem. You've probably heard of the recent world records and near-world record solves that were disqualified on a misscramble either during or after the competition. Often the solver is not able to attain a similar result again. There have been cases where the world record has been able to be recovered, or "clutched," like Tymon Kolasiński's groundbreaking 4.86 World Record 3x3 average. There have also been cases (that are far more common) in which the solver is unable to achieve the world record average, such as Wang YiHeng's 4.74 World Record 3x3 average, in which the solver achieved 4 sub-5 solves and one 7-second solve-- truly amazing even for modern cubing. However, when one of YiHeng's times was disqualified on a misscramble, the average was no longer the WR average. YiHeng and his mother posted several videos going over the scramble as opposed to the misscramble and found that the misscramble caused almost no difference in the solve-- only the final move in AUF. Additionally there were 18,000 attempts removed in Vietnamese competitions stretching from 2013 to 2022 for breaking similar regulations that affected the competitor's perception of the scramble, such as using scorecards as cube covers which is illegal.
To begin solving these issues that are appearing to grow more and more common as time passes, we must first clarify the main types of misscrambles.
The first type is the malicious one that can be prevented. It's when a scrambler intentionally misscrambles a cube to make it easier or harder for the solvers in that heat or round, or for other reasons that the scrambler deems acceptable.
The second type is non-malicious but can also be prevented. It's when a scrambler unintentionally misscrambles a cube and forgets to check it against the TNoodle scramble image. Often this scramble is off by just a couple moves-- usually not enough to affect anything, but still a misscramble and therefore a DNF.
The third type is non-malicious and seems like it's unable to be prevented, but there are still ways to solve this issue. It's when a move is unintentionally applied in the process of the scrambler putting a cube in a cover, the cover being carried to a table, the cube being set down on said table, and the cover being lifted to begin the competitor's inspection time. These are the hardest to detect.
For simplicity I will refer to these as Type 1, Type 2, and Type 3.
The misscramble epidemic is not new-- misscrambles have been in existence for a long time. The solutions I'm presenting today as potential solutions for them may be newer but most of them are based on things already in place.
Potential Solutions
For Type 1 scrambles the solution is pretty simple: For every heat for which scrambles are being performed, have an extra volunteer look at cubes and have their only job be to check scrambles. If it's correct, it goes on, and if it's wrong it gets solved and scrambled again (in the proper scramble this time). Then and only then does the Checker sign the scorecard. If an unusual number of cubes have incorrect scrambles, the Checker can flag down a delegate and suggest having a different scrambler. Another way to further prevent Type 1 scrambles is do what many delegates are already doing: Only let the fastest cubers out of those who are not currently competing scramble.
Type 2 can be solved using the techniques used in Type 1, which is fortunate because this means if any error is performed by the scrambler they can theoretically be solved by both methods. I will also introduce another potential alternative solution for this however. GAN scrambling robots are a relatively new (and also relatively slow) addition to cube scrambling and very expensive, but a GAN robot is able to scramble cubes with pretty good accuracy as compared to a human scrambler. There are cheaper homemade alternatives to this and I don't expect robots to replace human scramblers anytime soon, but it's a thought.
Type 3 scrambles are the most difficult scrambles to solve at the surface, but the solution for this already exists and is widely used. What is this magical solution? Well, about a year ago on Cyoubx's Friends (a popular cubing Facebook group for world-class cubers and for delegates alike), and on Reddit and our own SpeedSolving.com, a different type of cube cover was introduced which is easily 3D printed and transported to competitions for Square-1 solving. This cover has a 12-pointed star design that prevents any misalignment in the scramble state of a Square-1. Within a few months this design was altered to include a rectangular opening going through the center of this to prevent any moves being performed after Clock scrambling. One thing that you may notice is that, in the solved state, a Square-1 resembles a 3x3. This means that this cube cover could be very practical for preventing 3x3 scramble misalignment as well. The version that would likely have to be used is the one without the clock design added so as to prevent slice turns through this space in the cube cover, but this could be very practical for most 3x3s. The only issues would be ones that are inherent with any cube: size. Special care could be taken to make sure solves on different sized cubes are performed correctly as compared to other "typical" scrambles, especially if they are too big to fit the cover or too small to have any movement restricted by the cube cover.
Let me know about what you think, WCA. I spent 59 minutes on this post and if these are all implemented they could very well prevent much more than 59 minutes of grief for solvers who would otherwise get misscrambles and have them changed to DNFs.
To begin solving these issues that are appearing to grow more and more common as time passes, we must first clarify the main types of misscrambles.
The first type is the malicious one that can be prevented. It's when a scrambler intentionally misscrambles a cube to make it easier or harder for the solvers in that heat or round, or for other reasons that the scrambler deems acceptable.
The second type is non-malicious but can also be prevented. It's when a scrambler unintentionally misscrambles a cube and forgets to check it against the TNoodle scramble image. Often this scramble is off by just a couple moves-- usually not enough to affect anything, but still a misscramble and therefore a DNF.
The third type is non-malicious and seems like it's unable to be prevented, but there are still ways to solve this issue. It's when a move is unintentionally applied in the process of the scrambler putting a cube in a cover, the cover being carried to a table, the cube being set down on said table, and the cover being lifted to begin the competitor's inspection time. These are the hardest to detect.
For simplicity I will refer to these as Type 1, Type 2, and Type 3.
The misscramble epidemic is not new-- misscrambles have been in existence for a long time. The solutions I'm presenting today as potential solutions for them may be newer but most of them are based on things already in place.
Potential Solutions
For Type 1 scrambles the solution is pretty simple: For every heat for which scrambles are being performed, have an extra volunteer look at cubes and have their only job be to check scrambles. If it's correct, it goes on, and if it's wrong it gets solved and scrambled again (in the proper scramble this time). Then and only then does the Checker sign the scorecard. If an unusual number of cubes have incorrect scrambles, the Checker can flag down a delegate and suggest having a different scrambler. Another way to further prevent Type 1 scrambles is do what many delegates are already doing: Only let the fastest cubers out of those who are not currently competing scramble.
Type 2 can be solved using the techniques used in Type 1, which is fortunate because this means if any error is performed by the scrambler they can theoretically be solved by both methods. I will also introduce another potential alternative solution for this however. GAN scrambling robots are a relatively new (and also relatively slow) addition to cube scrambling and very expensive, but a GAN robot is able to scramble cubes with pretty good accuracy as compared to a human scrambler. There are cheaper homemade alternatives to this and I don't expect robots to replace human scramblers anytime soon, but it's a thought.
Type 3 scrambles are the most difficult scrambles to solve at the surface, but the solution for this already exists and is widely used. What is this magical solution? Well, about a year ago on Cyoubx's Friends (a popular cubing Facebook group for world-class cubers and for delegates alike), and on Reddit and our own SpeedSolving.com, a different type of cube cover was introduced which is easily 3D printed and transported to competitions for Square-1 solving. This cover has a 12-pointed star design that prevents any misalignment in the scramble state of a Square-1. Within a few months this design was altered to include a rectangular opening going through the center of this to prevent any moves being performed after Clock scrambling. One thing that you may notice is that, in the solved state, a Square-1 resembles a 3x3. This means that this cube cover could be very practical for preventing 3x3 scramble misalignment as well. The version that would likely have to be used is the one without the clock design added so as to prevent slice turns through this space in the cube cover, but this could be very practical for most 3x3s. The only issues would be ones that are inherent with any cube: size. Special care could be taken to make sure solves on different sized cubes are performed correctly as compared to other "typical" scrambles, especially if they are too big to fit the cover or too small to have any movement restricted by the cube cover.
Let me know about what you think, WCA. I spent 59 minutes on this post and if these are all implemented they could very well prevent much more than 59 minutes of grief for solvers who would otherwise get misscrambles and have them changed to DNFs.
