[Unofficial] Mike Hughey 6x6x6 BLD 50:15.94

[Mike Hughey]

I really wasn't expecting to get a 6x6x6 BLD so soon after my 5x5x5 BLD, but I did this week's first competition solve successfully last night, and got it on video. So here it is! It's a pretty rotten time. I had a place where I did one pair of pieces in the wrong order, and had to backtrack to fix it (the problem with me storing 3 images at each location), and also I had a real fiasco (it's before the part on the video - I was too embarrassed to show it to non-cubers) where I finished memorizing, and then pulled the blindfold off of my forehead instead of pulling it down over my eyes. And then, to make matters worse, I put the cube down to put the blindfold back on, and when I picked it back up I discovered I had reoriented it, so I had to figure out the right way to hold it again. So my real memorization time was something like 21:30.

The video is at 4 times normal speed. Sorry about the camera being so off-center - it must have slipped right before I started.

To help you follow along, I solve in the following order (this is usually how I do 6x6x6 BLDs now):
corner orientation
corner permutation (except parity)
outer X centers
inner X centers
left obliques
right obliques
inner wings
outer wings
parity

Memorization time (after the fiasco with the blindfold): 22:53.

 

[joey]

reserved rersever ersvevsvs

 

[longshot789]

Incredible. Pretty fluid on the execution too. I'm infinitely impressed.

 

[Pedro]

how do you do the corners at the beginning? center-safe algs?

I prefer to do center first, so I don't have to worry about it anymore (not that I'm doing the 6x6, yet )

 

[Mike Hughey]

Yes - the only algorithms I use for corner permutation (other than parity) are:

A perm
(R B' R' B)x3
U2 R2 U2 (R B' R' B)x3
(R2 D R2 D' R2 U2)x2 and inverse

For 2 2-cycles that don't work easily with algorithms 2 or 3 above, I just do two 3-cycles instead. Instead of (A B), (C D), I do (A B C), (A D C). I think I remember I originally got that idea from comments by Marcus Stuhr, who apparently solved them that way (a long time ago, before I ever took up speedcubing) even on 3x3x3.

I like to do corners first so they can go in short-term memory. But I still do them first out of habit when doing multis, even though that probably doesn't make sense.

I also have what is probably a bad habit of making sure any corner parity I have left is at UFL, UFR, even though that unnecessarily takes an extra algorithm. Then I can simply remember I have parity, without having to remember where the parity is. Chris Hardwick once (years ago) said he did this, and I still do it that way. Probably kills me on a lot of 4x4x4 BLD solves, but by the time you get to the really big cubes, it's not a very big deal.

 

[Ville Seppänen]

Mike is amazing!

Were you just really careful in the end or is the cube that stiff?

 

[Mike Hughey]

I slowed the video down to normal speed right before the end, so the ending would be normal. The rest is running 4 times normal speed, you know. For you, Ville, my 4 times normal speed probably looks like normal speed for you.

But yeah, I definitely do 6x6x6 a bit slower than the others (and that's as compared to my normal slow speed), because I'm afraid of pops. I think it would be awful to almost finish a 6x6x6 BLD solve and pop at the end!

Edit: On looking at it again, I guess I was being particularly careful at the end. Just a few solves ago, I missed a 5x5x5 BLD on video simply because I messed up the T perm at the end. I didn't want that to happen again. So I was going slower than normal there.

 

[masterofthebass]

Ummm... mike, if you didn't move the cube after putting on the blindfold, then you can take it off again. I think that would've made your memo a little faster

 

[ExoCorsair]

Nicely done.

 

[Mike Hughey]

Ummm... mike, if you didn't move the cube after putting on the blindfold, then you can take it off again. I think that would've made your memo a little faster

I'm not sure I understand your comment. I did take it off again (or at least I peeked under it) - didn't you notice? Or are you saying that I should have peeked earlier? If so, that's probably true, but I was really collecting my thoughts after my earlier blindfold fiasco most of that time, not trying to remember the orientation. I probably did waste a few seconds once I started really thinking about it with the following sort of dialogue: "Should I look to be sure? But if I do, there will be non-cubers seeing this on YouTube who will think I cheated. On the other hand, if I don't, I might miss this whole 6x6x6 by just a couple of corners misoriented. That would be awful! Okay, I'll peek."

 

[masterofthebass]

Sorry Mike, I thought you tried to orient the cube without taking off the blindfold XD. Ignore that comment then.

 

[PatrickJameson]

Amazing. Just amazing.

Nice color scheme too

 

[Hadley4000]

Mike, you rock. Amazing memory.

 

[pjk]

Congrats Mike, nice job.

 

[Dene]

Yay Mr. Hughey!!!

 

[Ryanrex116]

That was amazing.

Did anyone catch what he said right at the beginning?

 

[mati rubik]

now the 7x7?

 

[badmephisto]

very impressive. I'm not really into this so forgive this newb question, but how does this compare to solving 3x3 cubes in terms of memorization? You could take 8 cubes together to form a 6x6, so is it about 8 cubes? hmm

 

[Swordsman Kirby]

Well, it's got:

24 left-obliques, 24 right-obliques, 24 outer corner centers, 24 inner corner centers (don't know the names of the latter two), 24 outer wings, 24 inner wings, 8 corners... compared to 3x3's 12 edges and 8 corners.

The centers are kinda tricky because of the ambiguity, as well.

 

[Mike Hughey]

Yeah, what Tim said. But the centers are really a win more than a loss, because once you get used to dealing with them (so they don't feel "tricky" anymore), you have the advantage of being able to pick an orientation to maximize solved centers, so there's a lot less than 24 of each type that actually need solving. (Maybe 20 of each type on average.) A nice thing about the ambiguity of the centers is that you can put your fingers on them while memorizing (I learned that from Chris Hardwick) to keep track of which ones you still need to do, so unlike the edges, you don't have to waste time figuring out which ones you've missed - you know when you're finished memorizing them all.

From a practical perspective with my particular memorization approach, when I'm doing multiBLD it typically takes 5 locations to store a 3x3x3. It took 21 locations for this particular 6x6x6, plus corners, which I do short-term with a single cube. So really it's only about 4 times as many locations, but each location is completely filled up on a 6x6x6, while that's not true on a 3x3x3.

I will try to get a 7x7x7 on video Sunday night, but I only do one per week, so if I DNF, I'll have to wait another week for another chance. So it might come soon or it might be months - it depends on how lucky I am.

Oh, and at the beginning, I was just saying "22:53". I figured that since it was on video, I wouldn't have to remember what my memorization time was.