Toronto results

[Pedro]

If you can do 17/18, handle the disappointment, decide to try again (and get 14/18) you are a hero!

surely...and there's the mental stress/tiredness...

 

[Lt-UnReaL]

But then someone could ask for 238428394 cubes and solve 20 of them...

 

[qqwref]

You still have to solve 20 of them, no matter how many cubes you say you are going to do.

I heard a suggestion a while ago that was based on the WCA time limit for cubes. Remember how for multiple blindfold you get 15 minutes per cube for each cube up to 6, and 10 minutes for each after that? The idea was that you give a large time penalty, say about 10 or 20 minutes, for each cube that is unsolved at the end, and then check if the competitor is still under the time limit for the number of cubes that ended up solved. Presumably if they are not the attempt would count as a DNF or something.

The way this would work is - suppose you did 20 cubes out of 100. Well you didn't solve 80 of them, so you add on 80*20 minutes = 26 hr 40 minutes penalty. Now no matter what your time is this is well over the time limit for solving 20 cubes, so your attempt wouldn't count as a successful solve. That's why you can't just do a huge number of cubes. But if you look at Momonga's 17/18 in 2:10, that would be turned into a successful solve of 17 cubes, out of 18, in a time of 2:30 including the penalty - although he made an honest mistake he did solve 17 cubes correctly in under the time limit for solving 17 cubes, so in this system of measurement it would count as a record.

I don't do multiple blindfold, but I've done enough statistics to know that, even if someone has a constant rate of success no matter how many cubes they attempt, 17/18 actually represents a higher rate of success than 2/2. If you get a cube right 71% of the time (very low), you can get 2/2 half of the time, but the chances for 17/18 are practically infinitesimal... It seems odd to me to give 2/2 a higher ranking given that it is both more likely for a cuber to get it, and requires less skill.

 

[Pedro]

I don't do multiple blindfold, but I've done enough statistics to know that, even if someone has a constant rate of success no matter how many cubes they attempt, 17/18 actually represents a higher rate of success than 2/2. If you get a cube right 71% of the time (very low), you can get 2/2 half of the time, but the chances for 17/18 are practically infinitesimal... It seems odd to me to give 2/2 a higher ranking given that it is both more likely for a cuber to get it, and requires less skill.

don't think that idea is good...

if you do X cubes, in almost the time limit, and make a mistake, you'd add 10 mins and go over the limit...so it wouldn't count...and I don't see that as fair

I used (or still do) to agree with you that something like 17/18 is "better" than 2/2, just because it takes much more skill...but...indeed, one who solves 2/2 did no mistake, and one who solves 17/18 did (at least) one mistake...so...

the bolded part...
71% of accuracy is very low? lol...then I really suck...my accuracy is like 50-60%, going for speed...

and...I don't think the chances of getting 17/18 are infinitesimal...people can use stronger/better memory techniques for multiple cubes, so they can remember it better and for a longer time...

 

[Dene]

I personally think that if someone could do 2/2 cubes in 2 and a half minutes, then they are amazing, and just as good as the person who does 20/20 cubes in 4 hours.

 

[Mike Hughey]

I would actually have a chance (an outside chance, but a chance) at succeeding at 20/20 cubes in 4 hours. I'd have no chance at all of doing 2/2 cubes in 2 and a half minutes.

 

[tim]

I personally think that if someone could do 2/2 cubes in 2 and a half minutes, then they are amazing, and just as good as the person who does 20/20 cubes in 4 hours.

o_O. Multi bld isn't about the time. It's about the amount of cubes, so 20 cubes in 4 hours is muuuuccch better.

Stefan once suggested a rule similar to the hour cards disciplin in memory sports. Every competitor gets one hour and x (x is determined by the competitor) scrambled cubes and has to memorize + solve as much of the cubes as he can within the hour. To avoid the "ask for 1000 cubes and solve the 2 easiest ones" problem, the competitor should solve at least x*50% cubes.
I really like this idea. This will make life easier for judges (just one hour instead of 3 or even 4 hours) and competitors (no need to borrow 40 cubes from other competitors). And really slow people have the chance to solve 2/3 cubes within the hour and don't have to take care of the 15 minutes/cube rule.

 

[Dene]

But, I think time is important. You don't think someone doing 2/2 cubes in 2 and a half minutes would be amazing? We all know that you can solve many cubes, so speed is going to be the next factor to aim for, right?

 

[Mike Hughey]

o_O. Multi bld isn't about the time. It's about the amount of cubes, so 20 cubes in 4 hours is muuuuccch better.

At EPGY a few weeks ago they had a 10 minute time limit on the attempt. There, it WAS about the time. So it depends on the competition.

I do like Stefan's idea, though. We need to come up with some way to make the event more practical, and that would keep the focus more on the number of cubes than on the time (at least until people get too good at it).

 

[tim]

But, I think time is important. You don't think someone doing 2/2 cubes in 2 and a half minutes would be amazing? We all know that you can solve many cubes, so speed is going to be the next factor to aim for, right?

Of course, that's amazing, but again: It's called multi bld, not multi speed bld or anything like that.

btw. i wouldn't call 12 many...

 

[Pedro]

But, I think time is important. You don't think someone doing 2/2 cubes in 2 and a half minutes would be amazing? We all know that you can solve many cubes, so speed is going to be the next factor to aim for, right?

Of course, that's amazing, but again: It's called multi bld, not multi speed bld or anything like that.

btw. i wouldn't call 12 many...

hahah...no, 12 is not many...lol