[WCA Regulations 2012] Proposed New Event: Skewb

[Kirjava]

And you don't think that that's a problem?

Not at all. Besides, the only way to combat it is with scramble filtering - which I really do not agree with.

For the majority of events, you can't objectively say that it's impossible to beat the WR average on a random set of scrambles.

I'm not doing. But don't you think in the future that the WR for 4x4x4 avg will end up probably being the one with 0/5 OLL parities?

This is just a natural aspect of cubing.

Not to mention, one of the main arguments people were making earlier is that it will always require skill to set the WR, and based on Kris' post, I'm not really convinced of that.

Once someone good gets 'easy scrambles', I don't see someone unskilled completely fluking better than they do on equally easy scrambles.

I think it should also be considered that since tps is so low on skewb

This is a common misconception that people who dislike skewb seem to promote. It's simply not true - people are getting close to 8tps for alg executions.

 

[jrb]

@kinch2002 Your argument about how the Skewb is awkward to turn makes no sense. If speedcubing was about fast turning, then it would be called speedturning, not speedsolving. In other words, speedcubing is about solving the cube as fast as you can, not turning the cube as fast as you can.

Also, to the people that say Skewb is too easy to be made an official event, I highly doubt that Skewb WRs will reach sub-2 for single and sub-3 for average(like Pyraminx), and I have not seen one person who is complaining about Pyraminx being an official event. (I have nothing against Pyraminx, it's actually one of my favorite events).

 

[Kirjava]

In other words, speedcubing is about solving the cube as fast as you can, not turning the cube as fast as you can.

People b*tch about how speedcubing is all about tps now and that makes it boring.

Then people b*tch about it being too hard to get high tps on a new proposed puzzle.

Go figure.

I highly doubt that Skewb WRs will reach sub-2 for single and sub-3 for average(like Pyraminx),

I can see sub2 being plausable eventually for single. However, average will probably never reach anywhere near 3.

and I have not seen one person who is complaining about Pyraminx being an official event.

Dene doesn't like pyra.


I'm looking forward to being told that we can't have Skewb because it's too easy, then going and competing in magic.

 

[JustinJ]

Not at all. Besides, the only way to combat it is with scramble filtering - which I really do not agree with.



I'm not doing. But don't you think in the future that the WR for 4x4x4 avg will end up probably being the one with 0/5 OLL parities?

This is just a natural aspect of cubing.

That's a valid point, however, OLL parity adds a much less significant amount of time relative to how long the puzzle takes. What I'm seeing so far is that the variance in cases for skewb causes a much more significant range of times.

Once someone good gets 'easy scrambles', I don't see someone unskilled completely fluking better than they do on equally easy scrambles.

Easy sets of scrambles are not all equally easy, and someone unskilled getting good times isn't the only problem. What if someone who is the best at the time gets an average far, far beyond what is normally achievable, due to easy scrambles? The WR average is then completely unrepresentative of the skill of that person, or anyone else, until someone else gets a set of equally ridiculous scrambles.

This is a common misconception that people who dislike skewb seem to promote. It's simply not true - people are getting close to 8tps for alg executions.

Citation needed? I'd be interested to see that for generic algs, not just the nicest cases.

But nonetheless, prove to me that the better half of cases for methods like Kir-Meep, acubist, and rubikskewb aren't completely imbalanced in terms of speed, compared to the worse half.

 

[Kirjava]

That's a valid point, however, OLL parity adds a much less significant amount of time relative to how long the puzzle takes. What I'm seeing so far is that the variance in cases for skewb causes a much more significant range of times.


Easy sets of scrambles are not all equally easy, and someone unskilled getting good times isn't the only problem. What if someone who is the best at the time gets an average far, far beyond what is normally achievable, due to easy scrambles? The WR average is then completely unrepresentative of the skill of that person, or anyone else, until someone else gets a set of equally ridiculous scrambles.

This issue applies equally to Pyraminx, and I don't see it happening at the moment.

Citation needed? I'd be interested to see that for generic algs, not just the nicest cases.

People can nearly do U Perm sub1. U perm is 8 moves.

I think 4tps for a full solve is quite easily achievable.

But nonetheless, prove to me that the better half of cases for methods like Kir-Meep, acubist, and rubikskewb aren't completely imbalanced in terms of speed, compared to the worse half.

I do not wish to put in the effort it takes to 'prove' this, but cannot see how they would be /that/ imbalanced.

 

[JustinJ]

This issue applies equally to Pyraminx, and I don't see it happening at the moment.

It happened to 2x2. When Rowe got 2.45, nobody (including him) was anywhere near that at home.

People can nearly do U Perm sub1. U perm is 8 moves.

I think 4tps for a full solve is quite easily achievable.

Here's Kris doing 9 moves for a different alg in 2.03 (4.4 tps), which looks pretty fast to me, and he's starting holding the skewb in the right places, with zero recognition time. U perm is much more finger friendly than most skewb algs.

I do not wish to put in the effort it takes to 'prove' this, but cannot see how they would be /that/ imbalanced.

For Kir-Meep, for the CLL step (I don't think people use EG, since the algs are bad, I could be wrong though), There are two cases, Pi and L, which I believe have about a 50-50 occurrence rate. Pi is half the moves of L and doesn't require a regrip, and Kris says that L is horrible compared to Pi.

For L5C, he also told me that U can be done about twice as fast as H or Z, which can be done about twice as fast as every other case.

Considering U perm is fairly common, and can be made even more common by forcing them, it seems that the best method for getting a good official average with Kir-Meep will be to just go to competitions until you get an average with 4 U perms, and every other average you do is worthless.

I don't know much about other methods, so I had to focus on Kir-Meep, but I wouldn't be surprised if it was similar for others.

 

[da25centz]

It happened to 2x2. When Rowe got 2.45, nobody (including him) was anywhere near that at home.



Here's Kris doing 9 moves for a different alg in 2.03 (4.4 tps), which looks pretty fast to me, and he's starting holding the skewb in the right places, with zero recognition time. U perm is much more finger friendly than most skewb algs.



For Kir-Meep, for the CLL step (I don't think people use EG, since the algs are bad, I could be wrong though), There are two cases, Pi and L, which I believe have about a 50-50 occurrence rate. Pi is half the moves of L and doesn't require a regrip, and Kris says that L is horrible compared to Pi.

For L5C, he also told me that U can be done about twice as fast as H or Z, which can be done about twice as fast as every other case.

Considering U perm is fairly common, and can be made even more common by forcing them, it seems that the best method for getting a good official average with Kir-Meep will be to just go to competitions until you get an average with 4 U perms, and every other average you do is worthless.

I don't know much about other methods, so I had to focus on Kir-Meep, but I wouldn't be surprised if it was similar for others.

So pyraminx shouldn't be an event because, say you use lbl, you could get a LL skip and that would give you a great solve?

 

[Kirjava]

It happened to 2x2. When Rowe got 2.45, nobody (including him) was anywhere near that at home.

2x2x2 is much easier than Pyraminx and Skewb.

Here's Kris doing 9 moves for a different alg in 2.03 (4.4 tps), which looks pretty fast to me, and he's starting holding the skewb in the right places, with zero recognition time. U perm is much more finger friendly than most skewb algs.

Z perm is much harder than U perm.

For Kir-Meep, for the CLL step (I don't think people use EG, since the algs are bad, I could be wrong though), There are two cases, Pi and L, which I believe have about a 50-50 occurrence rate. Pi is half the moves of L and doesn't require a regrip, and Kris says that L is horrible compared to Pi.

People use EG. A face is often nicer than a layer.

For L5C, he also told me that U can be done about twice as fast as H or Z, which can be done about twice as fast as every other case.

Doublesune, sune, fruruf etc vs every other CLL.

Considering U perm is fairly common, and can be made even more common by forcing them, it seems that the best method for getting a good official average with Kir-Meep will be to just go to competitions until you get an average with 4 U perms, and every other average you do is worthless.

Same with Pyra and L3E cases.

 

[JustinJ]

2x2x2 is much easier than Pyraminx and Skewb.

From what I've seen, Skewb times are just as variant.

Z perm is much harder than U perm.

As is every other L5C alg that's not U perm.

People use EG. A face is often nicer than a layer.

Examples of fast people who do?

Doublesune, sune, fruruf etc vs every other CLL.

Not nearly as common as U perm on Skewb. Especially if you force the top center solved, which is really easy to see/do in some cases, apparently.

Same with Pyra and L3E cases.

Which is bad. I don't think the time differences are as high though.

 

[Tim Major]

When I learnt algs for Skewb, I did Layer, CLL and force one centre, then L5C using U-perms. Luck happens, but the average world record always goes to the best. Look at Pyraminx's history, WR has mainly swapped between Odder and Yohei.
And when/if Skewb becomes official people will practise it more. I know a Japanese guy uses Layer -> rest of Skewb.
If we get the best in the world at Pyra, Skewb and 2x2, Pyra would be most efficient, then 2x2/Skewb. Skewb tps is hardly half as good as 2x2.
I for one used to average 10~ and I probably had a pb average of 8/9. I do not remember U-perms being the deciding factor in solves.
Edit: Kirjava, 2x2 may be faster but the current WR pyra methods are more efficient.

 

[Kirjava]

Justin:

The points you are making also mostly apply to existing events. Until there is a serious movement for removing these events the complaints you have are invalid.

 

[JustinJ]

Justin:

The points you are making also mostly apply to existing events. Until there is a serious movement for removing these events the complaints you have are invalid.

Fair enough. I still think it's an issue, but I guess it's a necessary one. Don't get me wrong, I'll certainly enjoy watching skewb solvers, and to be honest, this isn't really a problem for anyone but the top, say, 0.5% of competitors. I guess we'll see how it goes.

 

[Meep]

While the points that Justin is making do apply to existing events, they currently are issues with existing events. I believe he's trying to say that it's worth considering if what people want to add has the same, or a worse version of said issues. It's not really about removing the current events that have those problems but to prevent more of it from happening. It's valid, just a different perspective (As most of the posts in here are opinion based and not about actual statistics).

 

[Escher]

If you classify that as an opinion rather than a reason then I fear that I could never actually come up with a reason rather than an opinion.

I cannot off the top of my head solve a skewb, however the only time I have ever encountered a skewb I looked up a 4 minute tutorial on youtube and learnt how to solve it in that 4 minute period. I played with it for a bit, realised how stupid it was, and decided I had absolutely no interest in playing with one ever again.

By competitive spirit I mean creating some sort of competition, something challenging and not something that will turn into another joke fest. I already talked about this, I don't feel like repeating myself.

I'm not exactly sure on this "inclusion argument" that you mention, but if I'm getting the gist of it correctly, you're saying we should include things that take less time, but I don't see any particular reason to support that reasoning. Please correct me if I'm way off.

A reason might be something that wasn't essentially 'I don't like this puzzle because it does not fit my personal definition of what speedcubing should involve'.

My mention of inclusion wasn't supposed to imply an 'argument from time', more a counter to "including new events imposes some level of strain on those who run competitions" - of possible new events we are discussing, skewb does not pose much of a strain.

Again you're thinking about events that fit your definition of 'joke fest', or apparently 'stupid'. Right now 3x3 could be considered a joke fest, since times depend a lot on last layer case variance and ease of start.


I wish people could just accept that luck is an intimate part of cubing, and efforts to control or reduce it are pointless, unless you start out an event this way or wipe the results database clean. If people really hate luck that much, they should go practice magic.

 

[Tim Major]

Dene: cool story. I don't like 7x7. It's time consuming and extremely simple. Oh it's one of your favourite events? How could that be?
I got my 7x7 for Christmas, and solved it on the same day. No tutorial. If Skewb is so simple why didn't you work it out for yourself?

 

[Dene]

To be quite honest, skewb is more of a competitive event than feet and magics, and also all of what you have posted is opinions, and are not particularly relevant to this discussion.

The way I see it, the only way someone can object to introducing this event is to go against the "constructive discussion" thing. And it is relevant. This discussion should not be about what the regulations for skewb should be, this discussion should be about whether skewb should be an official event or not.

A reason might be something that wasn't essentially 'I don't like this puzzle because it does not fit my personal definition of what speedcubing should involve'.

I wasn't talking about speedcubing, I was talking about official WCA competitions.

My mention of inclusion wasn't supposed to imply an 'argument from time', more a counter to "including new events imposes some level of strain on those who run competitions" - of possible new events we are discussing, skewb does not pose much of a strain.

Oh right I see. Hmm well, I have no problem with adding events as a general statement, just adding events that I don't believe add to the competitive spirit sufficiently. Also, it might be worth my pointing out that the events that use up the most time at our competitions in Australia are the puzzles that take the shortest times to solve (except magics). This is because more people compete in these events. I assume the reason that more people compete in these events is because either the puzzles are simpler to learn and get fairly good at, or most people just don't have enough patience for puzzles that take longer to solve.

Again you're thinking about events that fit your definition of 'joke fest', or apparently 'stupid'. Right now 3x3 could be considered a joke fest, since times depend a lot on last layer case variance and ease of start.

Well it's a good thing that I am of the opinion that it might be worth not recognising the 3x3 single WR as official in any sense. Of course there is still the risk of an average being lucky too, but it still won't be a good average unless the person is very fast. I'm also open to the consideration of making averages for faster puzzles that are more than 5 solves.

Dene: cool story. I don't like 7x7. It's time consuming and extremely simple. Oh it's one of your favourite events? How could that be?
I got my 7x7 for Christmas, and solved it on the same day. No tutorial. If Skewb is so simple why didn't you work it out for yourself?

I figured out most of skewb by myself playing around with it, but I'm the first to admit I was never very good at creating algorithms or doing comms or that sort of thing, which is pretty much essential to figuring out a puzzle by yourself. And your 7x7 reference is completely irrelevant. If I gave you that 7x7 before you had ever solved a Rubik's cube you would never have been able to solve it without a tutorial. Please think before you post.

 

[mitch1234]

Meep
Sarah
Chris Bird
Odder
Lars V N
MaeLSTRoM

Just to mention some of the nerds

Chris doesn't really practice, but I'm one of those that does.

 

[Escher]

Well it's a good thing that I am of the opinion that it might be worth not recognising the 3x3 single WR as official in any sense.

I guess you're at least consistent then.

WCA/speedcubing whatever - you've said 'Skewb is silly' not 'Skewb as an event is silly'.

The 'luck is bad' view really annoys me, I think I'm going to stop posting itt and just hope that Skewb is put in the next regulations.

 

[Dene]

It's not about luck being bad, it's about the records being meaningless.

 

[Escher]

It's not about luck being bad, it's about the records being meaningless.

And by meaningless you mean luck-based?