Big cubes blind

[Jacco]

Hello,
I'm seeing more and more people doing big cubes blind. I've been doing 3x3 blind for a few weeks now, and I'm quite interested in doing big cubes. When searching around on the forum I couldn't really find suitable answers. My questions are the following; what are the basics of solving big cubes blind? What alghoritms and memory methods are used, and are there any tutorials made on this subject?

For what I know, doing big cubes blind consists out of centers - (wing)edges, corners etc. but what alghoritms and setup moves are required for this?

Thanks alot,
Jacco

 

[Mike Hughey]

Probably the best guides for this are in the tutorials section here. Start with Chris's description:
http://www.speedsolving.com/showthread.php?t=201
Then you can add the centers from my description:
http://www.speedsolving.com/showthread.php?t=2207
And there's also another perspective on commutators from Daniel here:
http://www.speedsolving.com/showthread.php?t=697

However, for edges, r2 is the new thing (I've started using it a lot), and Erik has a pretty good description for that here:
http://erikku.er.funpic.org/rubik/r2.html

Another thing that really helped me a ton was Chris's actual walkthrough of a typical 4x4x4 BLD solve, which he did in the Yahoo! group somewhere, but I'm not sure where the link to that is. I'll add it here if I find it, or maybe Chris will add it if he sees this.

Maybe we should add a sticky to a thread somewhere that contains pointers to all these tutorials - this question keeps coming up.

 

[Zeinest815]

i'm actually searching the forums for BIG cubes BLD tutorials, but i can't find one. thanks for bringing this up.

 

[cmhardw]

Another thing that really helped me a ton was Chris's actual walkthrough of a typical 4x4x4 BLD solve, which he did in the Yahoo! group somewhere, but I'm not sure where the link to that is. I'll add it here if I find it, or maybe Chris will add it if he sees this.

A good bit of this writeup is outdated compared to what I do now, but this older version of my method can still get sub-10 solves on 4x4x4 BLD at least.

http://tinyurl.com/y7y3ee

Chris

 

[Mike Hughey]

A good bit of this writeup is outdated compared to what I do now, but this older version of my method can still get sub-10 solves on 4x4x4 BLD at least.

http://tinyurl.com/y7y3ee

Chris

That's exactly the one I was thinking of - thanks, Chris!

Other than the fact that I use r2 for most of the edges and some minor differences in memorization, the method you used in this example really isn't far off of what I do today. So I'm curious, Chris - what in particular is outdated for you? I know that you use commutators for the corners, so they're totally different - I'm just interested in what you do differently than this for edges and centers. For me, corners don't take very long anyway - I'm not to a point yet where I could get any significant benefit by changing corners methods. (It probably couldn't save more than 30 seconds to change that, considering I can go sub-1 on a 2x2x2 BLD, and that's not enough to worry about when you're still usually over 15 minutes, like I am.)

 

[Erik]

However, for edges, r2 is the new thing (I've started using it a lot), and Erik has a pretty good description for that here:
http://erikku.er.funpic.org/rubik/r2.html

Like 'a certain guy with all BLD records' and I always say: r2 is the thing to do!

 

[RobinBloehm]

Well, I'm not in the M2 method for 3x3 BLD, but use Macky's 3-cycle quite efficiently, so should I try M2 on the 3x3 to make it possible for me to solve edges on 4x4 with r2 method or can I learn it without being able to solve 3x3 with M2?
I can do 4x4 BLD with commutators for edges, but as you said the r2 method is much faster and I ask you for the way to change to r2, thx

 

[Mike Hughey]

I find r2 to be easier to visualize than M2. I don't know why, since they're practically the same, but I do. So I don't see any reason why you need to learn M2 before you learn r2. In fact, I didn't really understand M2 very well until I learned to do r2, whereupon M2 was a bit easier for me (I still struggle with M2, but I find r2 very easy - I can get through solves with M2, but they're too slow).

I use r2 as Erik describes it, except when I have middle slice pieces to contend with, I will use commutators to solve the next 2 pieces, instead of using the special algorithms for r2. I had those special algorithms memorized from Erik's page for about 2 weeks, but since then I've forgotten them, since I don't bother with them now. Since you can do commutators on edges, you might consider my approach as an alternative.

 

[masterofthebass]

My experience with M2 / r2 is mainly the opposite. I found M2 really easy to learn, without ever looking at r2. When I went to r2, the only thing that i needed to learn to do was the middle slice edges, but the "normal" edges were just as fluent as my 3x3 ones.

 

[cmhardw]

I can do 4x4 BLD with commutators for edges, but as you said the r2 method is much faster

I admit that I am biased in my method (commutators), and that I also have very little experience with actually using r2 during a blindfolded solve, but I disagree with the above statement. I don't think that r2 is always "much faster" than commutators for wings of a big cube.

In this thread: http://tinyurl.com/2v4kxz I averaged 1:37 for solving the 4x4x4 edges while actually blindfolded. I admit that I believe Matyas' r2 is faster than my commutators, and possibly also Erik as well. However, I don't believe that commutators are always slower than r2. At 97 seconds to solve 24 wings I am taking roughly 4.04 seconds on average to solve each edge on 4x4x4, and again this includes recall time as well.

I'd be curious if some of the r2 guys would post their times for solving wings. I seriously doubt that I'm the fastest at big cube edges while blindfolded, Matyas is just such a juggernaut at this event, and so many other guys are really really fast now (Rafal, Tim, etc.). But I think I am not at the bottom of the totem pole either, even though everyone says commutators are much slower than r2.

I'm tired of commutators getting such a terrible rap all the time. Someone who solves with r2 please post your times (thank you masterofthebass for already doing so), as I am curious how commutators really do compare to r2. To me commutators are just as "mindless" as I hear people say r2 is since I've been using these algs for so long I don't have to think to recall which commutator to use in nearly all situations.

Chris

 

[Pedro]

I agree, Chris

well, I'm not using commutators myself...not yet...still need to make a system for memorising the centers (I'm thinking about using some "groups of 4", like the Beatles, the Fantastic 4, the Ninja Turtles...how bad can that be? )

but I believe commutators can be very fast when you can use them without thinking at all, like you said...it becomes so easy and natural that you can do it really fast (not my case yet )

 

[Stefan]

I'd be curious if some of the r2 guys would post their times for solving wings.

I'm very interested in this, too. I don't have a good 4x4 so my twisting is slow, and my memorization/recall is terrible so my own times would be meaningless for this. But I'm already using r2 and would like to know how fast the fastest guys are with it.

 

[Mike Hughey]

I agree with Chris as well. I was very impressed watching Daniel Beyer doing commutators at the Virginia Open. (Unfortunately, I never really got to watch Chris do them, but I'd assume he's probably just as fast or faster.) Daniel really flies through the commutators, and takes very little time to figure them out before doing them. I switched to r2 for the non-middle-slice edges simply because they come out probably about a minute faster for me. That's just because I'm still not all that good at commutators. But I'm still faster with commutators than I am with the middle-slice algorithms - I just can't execute all the moves for a pair of r2 middle-slice algorithms (or one middle-slice algorithm and a non-middle-slice one) as quickly as I can think of and execute a commutator.

I still believe that ultimately, for someone with LOTS of practice, commutators will always beat r2. But the problem is that it takes LOTS of practice (or perhaps talent at a level I certainly don't have) to get that way; r2 is mindless very quickly, and doesn't take much practice to get good at. I really wonder if Chris could ever be faster with r2 than with commutators, since he's already so good with commutators.

I'm pretty happy with my method - I'm still doing enough commutators to be getting better at them, and yet I'm getting more satisfying times than I would be getting with pure commutators. Besides, with all the centers work on 5x5x5 solves, I'm still doing more commutators overall than r2.

Chris, it might be worth your while to try r2 a bit. You'll find it's REALLY easy to learn - I learned it and was faster than commutators in just one weekend (undoubtedly because I'm so slow with commutators). For the 5x5x5, I solve the wings in between the x centers and the + centers (it's silly - I like to solve the 4x4x4 part of a 5x5x5 first, then solve the rest - it's a weird habit I picked up when I started, and I've never quit since then), and I find that r2 is a nice mental break from the commutators in between the centers phases. That's probably the biggest advantage for me when doing the 5x5x5 - the mental break.

(Pedro, the "groups of 4" idea is pretty cool. Good luck with it!)

 

[Pedro]

(Pedro, the "groups of 4" idea is pretty cool. Good luck with it!)

I like it too just don't know if it will be effective...but probably will, as it's just a different approach for images...

I'm using persons, as I do with edges on the 3x3 (just for multi), the only change is the way I assign those persons to the pieces...

I could have the Ninja turtles as the green centers...orange, red, blue, purple...hmm...that's not very good...I have the "green(yellow), green(white)" and so on...how am I going to do this?

well...I guess it's just about defining a way and rehearsing it a lot until I can say which one is which without thinking...

so...
Ninja Turtles, Fantastic Four, The Beatles (but they all look the same )...I thought about using the Backstreet Boys, since they're just four now haha...

I'll find my groups...hopefully I can get a successfull solve in a month or less...

 

[Mike Hughey]

I'd think you'd want to make sure they're as different as possible in your mind - I don't see how you'd ever keep the Backstreet Boys straight. But yes, I also use persons - I use them for the start and end of cycles. And I have a different category of persons for each type of piece I solve. + centers are musicians or musical groups, x centers are friends or close acquaintances, wings are cartoon characters (my favorite), and central edges on 5x5x5 are actors.

 

[Pedro]

I'd think you'd want to make sure they're as different as possible in your mind - I don't see how you'd ever keep the Backstreet Boys straight.

didn't really get what you meant here...

well, for the edges on the 3x3 (just for multi), I use PA, each stickers has a person/character and an action...so I get 1st piece person doing the 2nd piece action (I don't memorise my starting point, as it's not necessary)

I can use that for the 4x4 edges...just need something for the centers (and for the + edges and centers, but I'm not planning to do 5x5 yet )

 

[tim]

I can use that for the 4x4 edges...just need something for the centers (and for the + edges and centers, but I'm not planning to do 5x5 yet )

You can take your edge images for the + centers + edges.

 

[Pedro]

I can use that for the 4x4 edges...just need something for the centers (and for the + edges and centers, but I'm not planning to do 5x5 yet )

You can take your edge images for the + centers + edges.

well...I can...but I think that can make me confused...

I have 4 groups now, need 2 suggestions

 

[Ville Seppänen]

You could use your family members, if it fits you (and if there are 4 or more in you family ). I thought about this groups-of-4 thingy and then saw that you have the same idea. My plan is to use different families, relatives and people that are known. I now have all the 6 groups, I will probably modify them since I just recently came up with them. I'll probably use same groups for X and +centers. Right now I don't have time to try a 5x5x5 bld. I would try a 4x4x4 if I just had one :\. I'll probably try it this weekend, let's see what happens. I did try it a few weeks ago with purely letters for everything, was off by 4 centers (pretty good I think)

 

[tim]

I'd be curious if some of the r2 guys would post their times for solving wings.

I'm very interested in this, too. I don't have a good 4x4 so my twisting is slow, and my memorization/recall is terrible so my own times would be meaningless for this. But I'm already using r2 and would like to know how fast the fastest guys are with it.

It took me about 2:20 to solve all edges with r2 in this video: http://youtube.com/watch?v=p-JSlgZ7wDg.
My recall was almost perfect (just one small delay), but my execution time is very bad, because i can't turn my 4x4 fast. That's probably because i almost never do 4x4 speedsolves. So i'm not one of "the fastest guys", just a random guy. Maybe you should ask Mátyás .