Speedcubing statistics: preliminary report

[Dene]

Whoa that's crazy. How did you even notice that?

 

[pjk]

Ok so I just figured out something quite awesome(or horrible, depending on your view).

If we take the 'Most common .xx's' data(http://pastebin.com/f6bae0f30) we can see that all of the times are either >2582 or <901. This is interesting alone. But if we take a look on the number of times these numbers appear in a row, we find that it forms a perfect pattern. This is what we get;

112121211111212122
112121211111212122
112121211111212122
112121211111212122

That would make sense that a pattern would exist because of the previous data you sorted. There is obviously something in the timer causing this.

I contacted Bob Fox about it and he said that their lead engineer would get back with me shortly. I will let you know what I find out.

 

[qqwref]

By the way, given the numbers, I estimate that about 1/3 to 1/4 of the stackmats used do NOT have this problem. I think it must be that some versions (the most common ones) of the timer have this weird distribution, while others don't.

 

[StachuK1992]

I have an idea to quicken this process.

Does anyone have access to a server that people could upload a bunch of times?

I'm thinking that we could get a bunch of people to do an easy-ish puzzle (2x2/3x3) over and over, and have each person possibly upload their times to said database, then work from those numbers.

 

[spdcbr]

lol for the retard zone.

 

[PatrickJameson]

I have an idea to quicken this process.

Does anyone have access to a server that people could upload a bunch of times?

I'm thinking that we could get a bunch of people to do an easy-ish puzzle (2x2/3x3) over and over, and have each person possibly upload their times to said database, then work from those numbers.

Almost a quarter million times are not good enough?