[Unofficial] Mike Hughey Square-1 BLD avg 3/5: 8:59.83

[dillonbladez]

:O
Insane! This is absolutely amazing.
I think you once said that you'd remember the permutation and orientation changes when doing cubeshape algs, then memorize orientation (putting corners + edges to respective sides, same thing) then permutation
Is it still like that?

 

[RyanO]

As far as I know, most people who have done Sq-1 BLD before haven't used their own technique, but just speed BLD it.

So Mr. Hug-Hay would be the first to have developed an actual method for this.

Most people who have done the square 1 blind have started from cubeshape. I believe Stefan did complete a full scramble though. I would argue that speed BLD is an actual method. Just because Mike didn't do it first doesn't take anything away from this amazing accomplishment. I'm truly astonished by your accuracy at something that seems so out of reach for most of us.

 

[Weston]

Wow, there can't be very many people who have managed this! Congrats!

I'm 99.9% sure he is the only one...

Unless some crazy Indonesian decided not to tell us.

I've seen Takao Hashimoto do this a few times successfully "in competition."
He doesn't use Mike's matrices though. He memorizes the pieces where they are, and tracks them as he solves blindfolded.


Amazing job Mike!

 

[joey]

Mike has just essentially turned tracing into something you learn beforehand.

I mean, when I thought about how to do sq-1 BLD, this is the obvious way.

No offence to Mike, it's still awesome.

Still think he should try the way I suggested too ._.

 

[~Phoenix Death~]

Oh geez. Oh GODS.

 

[Mike Hughey]

Yes, a number of other people have done square-1 BLD before. Stefan was probably the first, and Takao is totally awesome at it. But I believe they have all traced the pieces during the solve. Which means that some scrambles are easier than others (depending on how many moves from square the puzzle is). With my method, they're all the same (unless the scramble is already square or almost square, in which case my method doesn't really have any advantage at all). The nice thing is that with my method, I'm easily getting sub-7 solves now, and sub-5 should be easy for someone who's actually good at it. (Hopefully I'll be there soon.) And I've never heard of anyone else getting sub-10.

Mike has just essentially turned tracing into something you learn beforehand.

That's basically right. It's the same as tracing, except that I've pre-traced all possibilities, so I don't have to do it during the solve.

Still think he should try the way I suggested too ._.

I did see your suggestion, and I don't see where the benefit is with your approach - you would just be doing the same work, but with the blindfold on instead of with the blindfold off. Am I missing something?

I hope to have the full method up in the next couple of days; I made a lot of progress today. I still need to add the last 26 cases and then check the whole thing over for obvious mistakes. I hope that once it's posted, it will answer some of the questions. But it is a bit complicated, so at first it might create more questions than it solves.

Oh, and by the way, I'm ordering some more square-1's (I only have 3). This method is VERY conducive to multi.

 

[TheAlmostMadHatter]

I loved watching you at the Cat and Mouse Mike! You were amazing

 

[Mike Hughey]

@joey: I guess I'm starting to see the potential advantage of your approach - you wouldn't have to juggle quite as much. I suppose I can give it a try sometime in the next couple of days and see if it helps or not.

I suspect I won't get as much out of it as you would, since I'm not very good at solving sighted with a BLD method. This is similar to that, except that you're not actually sighted.

 

[uberCuber]

wow impressive work :tu

can't wait to see you try multi!

 

[iRiLLL]

still downloading

240mb wtf

 

[Pitzu]

Mike, you are still awesome! :tu
Last year I started to think on an S1 blind method but then I had problems with my employment status. But I'm interested in your method.
After I consolidate my employment situation I should start again doing some crazy blindfolded solves (I have some plans... )

 

[KJiptner]

Absolutely amazing! I will watch the entire thing now, instead of working on my thesis!

 

[joey]

@joey: I guess I'm starting to see the potential advantage of your approach - you wouldn't have to juggle quite as much. I suppose I can give it a try sometime in the next couple of days and see if it helps or not.

I suspect I won't get as much out of it as you would, since I'm not very good at solving sighted with a BLD method. This is similar to that, except that you're not actually sighted.

Just give it a go
I just want to see if you can actually get similar times with it, at least.

 

[AvGalen]

Erik only rules all galaxies with light
Mike would easily kick him out of the ones where you can't see

 

[RCTACameron]

I'd be really annoyed if this became a WCA event before 2x2 BLD.

Square-1 multi blind? If someone (probably Hug-Hey) did this, I would be scared. It's probably going to happen, though. Now, it's Square-2 blind time!

 

[Forte]

I'd be really annoyed if this became a WCA event before 2x2 BLD.

Square-1 multi blind? If someone (probably Hug-Hey) did this, I would be scared. It's probably going to happen, though. Now, it's Square-2 blind time!

Sq2 is easier

 

[joey]

Sq-2 is easier.

 

[Enter]

sir Mike you are soo cool!!!!

 

[Mike Hughey]

Thanks for the compliments, everyone!

@joey: I guess I'm starting to see the potential advantage of your approach - you wouldn't have to juggle quite as much. I suppose I can give it a try sometime in the next couple of days and see if it helps or not.

I suspect I won't get as much out of it as you would, since I'm not very good at solving sighted with a BLD method. This is similar to that, except that you're not actually sighted.

Just give it a go
I just want to see if you can actually get similar times with it, at least.

This morning, I had just about convinced myself that your method would be better. But now that I've tried it, I'm back to my previous opinion. I'm disappointed, because for a few minutes there I was convinced this would take me sub-5.

I've made two tries so far, both DNFs. The first was a disaster because I got lost during execution; I think just because of unfamiliarity with the approach. The second one was off by a 3 cycle of edges, time 8:44. My memo time is significantly shorter - more like 3 minutes. That seems outrageously slow considering what little I have to memorize, but apparently it's still hard even doing that little, for me at least. Then solving is definitely more work. The worst is when you realize you have to break a new cycle, and you have to remember what you've already solved.

Still, I think it can be about the same speed, and I could see how some people might find this easier. I doubt I ever will. But I'll probably keep trying until I get a success, at least.

@Pitzu: I'm really looking forward to seeing the results of your plans!!

@RCTACameron: With this method, square-1 multiBLD is really quite straightforward. Especially if you do it my way instead of Joey's. You just store the results in long-term memory, and you're good to go. You can figure out how to get to square by feel, if you can't remember, and from there, the memo is just like 3x3x3, only less to remember. Because there is so little to remember for a square-1, I actually think it will scale better to multiple cubes than 3x3x3 does. I would say that for a small number of puzzles, like 5 or 6, I should be almost as fast with multi as I would be doing them individually.

 

[r_517]

Hey! Hug! Mike!