New type of 2x2 scrambler

[qqwref]

http://mzrg.com/miniSites/scramblers/222 11moves.html

I'd like to propose for this to become WCA-official but this type of thing isn't under consideration right now so there is no point in posting it in the WCA forum yet. Anyway this is a modification of the current WCA-official scrambler. Here are the changes:
- The official scrambler does 500 moves to find the random position, but this one is better because it simply picks a random position directly.
- This scrambler will discard scrambles that are 7 or fewer moves from solved (this happens <20% of the time). This means that with this scrambler there are no incredibly easy solves! I think this could restore some legitimacy to the 2x2 single solve event.
- The scrambles here are always 11 moves (the maximum number of moves a 2x2 scramble could require) so there is no way to tell how easy a scramble is by looking at the moves.
- I removed the "minimum scramble length" thing because that was not necessary.
- I cleaned up the code a LOT, removing any unused functions and rewriting some things to make them smaller without changing what they did. (For comparison's sake, this is 10kb and the official one is 16kb.)

 

[esquimalt1]

I'm liking it a lot

 

[Tim Major]

This means that with this scrambler there are no incredibly easy solves! I think this could restore some legitimacy to the 2x2 single solve event.

But, but, but... I like easy scrambles...

Edit: 1st scramble I tried: F U F' R F U' F2 R U R' U'

A bar properly already made, do maybe the scrambles aren't such a bad idea.

 

[hr.mohr]

- The scrambles here are always 11 moves (the maximum number of moves a 2x2 scramble could require) so there is no way to tell how easy a scramble is by looking at the moves.

Very nice, great work!

 

[qqwref]

This means that with this scrambler there are no incredibly easy solves! I think this could restore some legitimacy to the 2x2 single solve event.

But, but, but... I like easy scrambles...

Edit: 1st scramble I tried: F U F' R F U' F2 R U R' U'

A bar properly already made, do maybe the scrambles aren't such a bad idea.

Careful. I said there no *incredibly* easy solves. This does not mean there are no easy solves ever. There is still a component of luck and there will always be some scrambles that are easier than others. It is even possible to get a scramble with one layer solved on this scrambler, but the CLL will be at least 8 moves including AUF ;-) I just want to prevent scrambles that are so easy that slower competitors get sub-2 second times because they see it all from inspection. Look at this: sub-1.5 single, 7 second average - is that fair, and is the single solve representative of that person's skill? No, it is not.

 

[esquimalt1]

I agree with the example of slower competitors getting sub 2 singles. Hopefully if the WCA accepts this scrambler, 2x2 single will involve more skill.

 

[fundash]

I like it!

Now all i need is an east sheen 2x2...

 

[Jake Gouldon]

Aren't some states impossible to reach because the scramble is always exactly 11 moves?

 

[qqwref]

Aren't some states impossible to reach because the scramble is always exactly 11 moves?

There's no proof that every position can be solved in 11 moves, but it seems to be possible most of the time.

 

[DavidWoner]

This scrambler will discard scrambles that are 7 or fewer moves from solved (this happens <20% of the time). This means that with this scrambler there are no incredibly easy solves! I think this could restore some legitimacy to the 2x2 single solve event.

....................

Wrong.

R U R' U R U2 R' U
F R' F' R U R U' R'
F U R U' R' F' R U
etc

All incredibly easy, all >7 moves.

F R2 U' F' R'
R U' F2 R F
U F2 R U'
etc

All short, none of them are obviously easy, except to experienced solvers.

And what happens to the old records? As with the change in multiBLD, this would make the old records a lot harder to beat. I like the idea of outputting all 11 move scrambles, I suggested something similar to Arnaud for when an easy scramble came up in the weekly comp. However, dropping all <8 move positions is stupid.

 

[qqwref]

R U R' U R U2 R' U
All incredibly easy, all >7 moves.
[...]
However, dropping all <8 move positions is stupid.

I think you are completely misinterpreting what I mean by "incredibly easy". I mean stuff where the solution is very few moves, not stuff where you happen to know how to solve it. The cases you gave are easy but not "incredibly easy". I'm not talking about the kind of scrambles where you get really nice times, I'm talking about the kind of scrambles that are not just "really nice", but completely unfair to anyone who was not at the competition.

The reason I chose <8 was because it is not arbitrary. Obviously we NEED to have something that prohibits stupid **** like F R U2 R' from showing up in competition, but there's no way the scrambler can decide how easy a scramble will be for a human solver and still be method-independent. A scrambler doesn't know the difference between a sune and R U' R U2 R U' R'. Discarding everything 7 moves or fewer will prevent most stupidly easy scrambles from showing up while discarding less than 20% of the legit ones. If you have such a problem with easy 8+ move scrambles existing, why don't you suggest a better way to get rid of the incredibly easy ones?

F R2 U' F' R'
R U' F2 R F
U F2 R U'
All short, none of them are obviously easy, except to experienced solvers.

It's not entirely about whether you can see the solution. Take R U' F2 R F for instance: this can be solved by doing z' y L' U' L R U R'. Even if someone can't SEE the solution from inspection, the scramble is objectively easy and thus they have a higher chance of getting a really lucky skip (in this case, LL skip in 6 moves qtm!) and then achieving a time that I don't want to have to compete with. Similarly the first one could plausibly be done with guimond as y2 R U R' R2 F2 U'. The fact is even though these scrambles might be hard to see from inspection they are still objectively easy and provide an extra-high chance of an extra-low time to an inexperienced solver.

And what happens to the old records? As with the change in multiBLD, this would make the old records a lot harder to beat.

I was thinking we could just discard all solves below a certain threshhold - 2.00 seconds, say, or the best time in the WC09 final, which was 2.46 seconds. I don't know what the best 8+ move solve in competition was but it's probably around there.

 

[DavidWoner]

Obviously we NEED to have something that prohibits stupid **** like F R U2 R' from showing up in competition

Obviously? No. I can't think of a single reason that doesn't amount to whining about fairness.

The reason I chose <8 was because it is not arbitrary ... Discarding everything 7 moves or fewer will prevent most stupidly easy scrambles from showing up while discarding less than 20% of the legit ones.

Uhh... <8 IS arbitrary, you never go about explaining where that particular number comes from. As I recall, Javier's 1.63, Patrick's NAR, Eric's current NAR, and this most recent scramble from WC were all FOUR moves. Not five or six or seven, but FOUR. So wouldn't discarding all positions <5 moves from solved make a lot more sense?

why don't you suggest a better way to get rid of the incredibly easy ones?

Because I honestly don't give a **** if some random noob has a better 2x2 single time than me. I don't understand why you seem to.

And what happens to the old records? As with the change in multiBLD, this would make the old records a lot harder to beat.

I was thinking we could just discard all solves below a certain threshhold - 2.00 seconds, say, or the best time in the WC09 final, which was 2.46 seconds. I don't know what the best 8+ move solve in competition was but it's probably around there.

14 Erik Johnson 1.72 USA US Nationals 2009
20 Phil Thomas 1.86 USA Wisconsin Open 2008

maybe more.

If you want to kill all the sub 2.46 times, there are a ton more.

 

[blade740]

I think the scrambler is a good idea, but I don't think there's a fair way to remove old records.

 

[Edmund]

Yeah, if a solution has to be x amount of moves how the hell are we supposed to beat .96?

 

[esquimalt1]

Yeah, if a solution has to be x amount of moves how the hell are we supposed to beat .96?

I believe Erik's 0.96 solve was 8 moves.

 

[DavidWoner]

Yeah, if a solution has to be x amount of moves how the hell are we supposed to beat .96?

I believe Erik's 0.96 solve was 8 moves.

Scramble #2 of finals: F' U' F' R2 F2 R U2 R' U F U R2 U F' U2 F2 R' U R F' R U2 R2 U' F
Solution: x2 U R' U' R U R' U'
hint: got to know your sexy moves to see what will happen in inspection...

 

[esquimalt1]

Alright, so then it was 7 moves.

 

[Erik]

Ya 7 moves is not that uncommon at all. We can start the discussion again about easy scrambles if you want... but move count is no indicator for easy-ness. If you have a J perm as scramble it's also like 8 or 9 moves, but ridiculously easy... at the same time Derrick and me often race 5 moves scrambles for fun... I can tell you they can be VERY hard...

 

[TomZ]

Nice work! The credits list is growing nicely...

I strongly disagree with setting a minimum move count. Maybe it is a good idea to block some scrambles, but setting the limit at 7 moves is ridiculous. That represents nearly 8% of all scrambles. This means that in over 33% of averages of five there would be at least one seven move (or less) position. Scrambles of six moves or less occur in 8% of averages of five, still quite a lot to discard. Setting the limit at 5 might make sense, but 7 is just wrong.
Getting lucky is part of the game, after all.

"The scrambles here are always 11 moves (the maximum number of moves a 2x2 scramble could require) so there is no way to tell how easy a scramble is by looking at the moves."
How is that a new feature?
Secondly, setting a 11 move minimum is just a waste. Move count is a terrible indicator of difficulty. It would seem like a good idea to mask 4-move solves (if they remain) but there is no need to hide a solve is 8 moves. And (in theory at least) the solvers should not even be able to find this out in the first place.

Good work on the cleaning up. I didn't really bother to do that as it doesn't make much of a difference and consumes heaps of time, but it's nice.

 

[qqwref]

"The scrambles here are always 11 moves (the maximum number of moves a 2x2 scramble could require) so there is no way to tell how easy a scramble is by looking at the moves."
How is that a new feature?

It's not new in the sense that it has never been done before, it's new in the sense that the current WCA scrambler does not do it (the current WCA scrambler is optimal). Hence it is something I changed.

Secondly, setting a 11 move minimum is just a waste. Move count is a terrible indicator of difficulty. It would seem like a good idea to mask 4-move solves (if they remain) but there is no need to hide a solve is 8 moves. And (in theory at least) the solvers should not even be able to find this out in the first place.

You're right that making every scramble 11 moves is not necessary if all scrambles are guaranteed to be 8+ moves from solved anyway. The main reason I did this is because people tend to ask for it; it would be simple to remove.

Besides in practice there is no way to prevent people from finding out some data about the scrambles because the scramblers are in plain view at a table somewhere, as opposed to (say) in a completely different room. If they see a 4-move 2x2 scramble, they can't help but be a little excited... but, worse, if people solve a 4-move scramble, they tend to go tell their friends even if it is against the WCA regs to do so.


For those who are criticizing the idea of throwing out scrambles with <8 moves: do you at least agree that we need to remove very very easy scrambles? I think we need to discard ridiculously easy scrambles simply because they are massively unfair to anyone who does not get an attempt to solve them. You probably don't care about this issue but in my opinion having stuff like this makes cubing seem less like a carefully regulated sport and more like some kind of amateur club activity. I know most people treat the 2x2 single rankings as completely pointless, but I don't think it has to be that way; I think we treat it as useless because of the problem of insanely easy scrambles. As I said before, if you don't like the idea of discarding all scrambles under 8 moves, I would like to hear a better idea. TomZ's idea of discarding scrambles with 5 or fewer moves is a good one although it is still somewhat arbitrary.