• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

Limitations of Visual Memo? (And my new BLD method!)

Jude

Member
Joined
Mar 27, 2008
Messages
291
Location
Sheffield, England
WCA
2008WRIG02
Hey, after recently trying some Multi Blindfold attempts I was curious about the limits of visual memo. Basically, what I find at the moment is that visual memo is OK for multi BLD (so far the most I tried was 3, and I had no problems doing that with visual for both corners and edges) it just takes a hell of a long time to get it fixed in my memory. For 1 cube it takes me about a minute to memo visually, for 2 it takes about 6 and for 3 it takes me about 17. I don't even want to try 4 at the moment as although I'm confident I could do it, it would probably take me about half an hour to memo :( So, where do you think the limit of visual memo is? (Number of cubes? Time taken to memorise the cubes? etc)

Also, (it's only slightly related but I didn't want to make 2 threads) I recently made up my own method for solving edges. However, after talking to Joey and Ville about it, it turns out it's just like a simplified version of Erik's TuRBo method (which I knew nothing about before). I'll quickly run through my method (note, I'll talk in terms of stickers, not pieces, like in Old Pochmann (now on referred to as OP)):

Use UR as buffer, find the sticker it want's to shoot to and put it to either UL or LU (whichever is easier, but if it's UL call it a good edge, if it's LU call it a bad edge). Then, find the sticker the piece now at UL/LU wants to shoot to, and put that to UF or FU (whichever is easier, and again if it's UF call it a good edge, if it's FU call it a bad edge)

Then, choose one of 4 counter clockwise edge cycles (which either flip UL UF, UF UR, UR UL or none) The algorithms I use are here.
Code:
First edge      Second Edge            Alg
Good               Good    Any U perm which cycles UR>UL>UF
Good                Bad           M U' M' U2 M U' M'
Bad                 Good       U' R U R' U' M' U R U' r' U
Bad                  Bad        U' r' U' R U M' U' R' U R U

Anyway, my problem with this method is that I am currently much faster with OP. This is because in OP as the set up moves are incredibly easy and only 1 piece is solved at at a time, I can mindlessly solve 1 piece while looking ahead to the next edge so that I can either start the set-ups for the next piece instantly after or even get cancellations between the set ups. With my new method (doesn't have a name yet, any ideas? :p) I have to have a gap of at least 2 or 3 seconds to think about both of the next 2 pieces and how to set them up. My question is, will this problem go away with practice, or is there a problem with my memo system (pure visual - and finally, the 2nd part of this thread relates back to the thread title!)?

Thanks for any help you might have, and well done for reading such a massive post :p
 
Last edited:

fanwuq

Member
Joined
Dec 5, 2007
Messages
2,831
WCA
2008FANW01
YouTube
Visit Channel
This is TurBo edges, except you can also use the mirrors of your algs.
I made a thread call "Wrong setups on purpose?" when I first started BLD in March where I think I pretty much said the same thing.
I don't see any difference between this and TuRBo. I tried to use it when I was just starting, but It takes me like 10 seconds to think about each setup. I was using letters too, so I had to waste time decoding that. I think this method is easy to visualize with visual memory. It is move efficient, so you can try to get used to it.

I don't know the limits of visual Multi.
I think Tim Sun has done 6, Rafal 12+, and FP 15+. I'm very bad at visual, I could barely do 2 2x2s using pure visual.
All were correct, I guess I should change my memo system for corners though...it's visual and it might get confusing with more cubes...I'm gonna try 20 scrambles next week and see if it's still accurate.
Well, at least corners anyway.
 
Last edited:

cuBerBruce

Member
Joined
Oct 8, 2006
Messages
914
Location
Malden, MA, USA
WCA
2006NORS01
YouTube
Visit Channel
Then, choose one of 4 counter clockwise edge cycles (which either flip UL UF, UF UR, UR UL or none) The algorithms I use are here.
Code:
First edge          Second Edge            Alg
Good                    Good                  U perm
Good                    Bad         M U' M' U2 M U' M'
Bad                     Good        U' R U R' U' M' U R U' r' U
Bad                      Bad        U' r' U' R U M' U' R' U R U

Clearly, by "U perm" you mean the U-Permutation that cycles (UR UL UF). There are 31 other permutations that are also U perms, but not the one you mean.
 

F.P.

Member
Joined
Mar 25, 2008
Messages
207
Location
Austria
FP? I thought used loci.

Rafal only does corners visually.


As fanwuq already said (after editing), I use visual for corners.
I'm working on visual for edges as well...I'm basically just visualising a string which loops around the cube to the certain stickers I need to memorise.
Not sure if I'm going to use that for multi though.
 

cuBerBruce

Member
Joined
Oct 8, 2006
Messages
914
Location
Malden, MA, USA
WCA
2006NORS01
YouTube
Visit Channel
Clearly, by "U perm" you mean the U-Permutation that cycles (UR UL UF). There are 31 other permutations that are also U perms, but not the one you mean.
There are 31 :O
I only know 8..

Yeah, most people only think about the ones that have no effect on the corners. Oh, and I forgot about the 160 other U perms if you consider the other 5 faces.
 

cuBerBruce

Member
Joined
Oct 8, 2006
Messages
914
Location
Malden, MA, USA
WCA
2006NORS01
YouTube
Visit Channel
Well, when we say U-perm, we mean a 3-cycle of edges in a single layer, that "keeps orientation".
Well, I would say a U-perm cycles 3 edges in a single layer with respect to the 4 corners of that layer (while those 4 corners remain fixed relative to each other), while, of course, also maintaining orientation (with orientation being defined in the way we usually define it when we only care about that one layer). Why would you define a U-perm in a way that only includes 25% of them? :)

Using another PLL as another example, the Speedsolving.com Wiki shows a G-perm as a 3-cycle of corners and a 3-cycle of edges. The Speedcubing.com site shows a G-perm as a 2-cycle of corners and a 4-cycle of edges. Which is right? Both are right. A G-perm could also perform a 2-cycle of edges of a 4-cycle of corners. As far as I've always known, an alg the solves any of the 64 G-perm cases can be called a G-Perm. In the same way, I assume any alg that solves any of the 32 U-Perm cases can be called a U-Perm.
 

Jude

Member
Joined
Mar 27, 2008
Messages
291
Location
Sheffield, England
WCA
2008WRIG02
Right, thanks for the feedback. I think I'll probably stick with this method, and once I'm more used to it add the 4 inverses into my repetoir, and hopefully I'll be even faster. I guess what this method lacks in mindless turning without any pauses, it makes up for in the amount of moves and algorithms it requires to solve. (Probably less than half of Old Pochmann on average)

Oh and also,
Then, choose one of 4 counter clockwise edge cycles (which either flip UL UF, UF UR, UR UL or none) The algorithms I use are here.
Code:
First edge          Second Edge            Alg
Good                    Good                  U perm
Good                    Bad         M U' M' U2 M U' M'
Bad                     Good        U' R U R' U' M' U R U' r' U
Bad                      Bad        U' r' U' R U M' U' R' U R U

Clearly, by "U perm" you mean the U-Permutation that cycles (UR UL UF). There are 31 other permutations that are also U perms, but not the one you mean.

Uh, I kinda thought that was obvious but I've edited it into my original post in case anyone else doesn't understand what I mean.
 
Top