CuBeOrDiE
Veteran
put 3 on top and 3 on bottom. then line them up to where you have a face you can easily turn. dunno, did i explain that well enough?
edit: this is not a method, just an idea. ive only done a few solves like this
Simple Little Idea for Higher-order Cube Edge Pairing
- Thread starter CuBeOrDiE
- Start date
put 3 on top and 3 on bottom. then line them up to where you have a face you can easily turn. dunno, did i explain that well enough?
edit: this is not a method, just an idea. ive only done a few solves like this
MichaelP.
Veteran
I think if you had one unpaired edge in the top and bottom, you could do FBRL turns by positioning the unpaired edges correctly. It won't be efficient though.
CuBeOrDiE
Veteran
its just an idea. but if you could solve edges and the 3x3 in one go, that would be awesome. but i haven't found a way yet, prbly because there isn't one.
edit: FYI my 5x5 has a snapped core so i can't solve right now
but im getting vcubes!
Kirjava
Colourful
You edited your posts to make it seem like this was your idea all along. Douchebag move. There was no mention of doing only 6 until I brought it up. You kept saying there were four edges to complete after the centres were complete. You made this mistake again and again and again. You would've noticed that you'd put 8 and 4 instead of 6 and 6 if you actually meant that.
You did it SO MANY TIMES. You did it in your CLARIFICATIONS.
I do not believe you.
koreancuber
Veteran
Bad idea, but stop scorching him. He just wasn't thinking enough.
CuBeOrDiE
Veteran
w/e. i don't have a 5x5 in my hands right now, so i didn't think when writing this. that is the naked truth. but when i solved like this before my 5x5 was broken i DID solve 6 tredges.
edit: i did that so new people to this thread wouldn't post the same things you posted. just to save time
editing the edit: i changed my posts so others wouldn't waste our time saying things like "hey, that doesn't work!"
Last edited:
Robert-Y
Legend
I thought of this idea before. After the 6 tredges and all of the centres have been solved, I was thinking about doing F M or F' M, then solving the 6 remaining tredges by using 2 wing chain solving. (Use the last two tredges algs if you need to...)
(Rotate the cube so that the 6 tredges which have already been solved are on the LF, LD, LB, RF, RD, and RB slots before you do the F M or F' M.)
I still can't seem to find a solution for the high move consumption though :/ ...
(Rotate the cube so that the 6 tredges which have already been solved are on the LF, LD, LB, RF, RD, and RB slots before you do the F M or F' M.)
I still can't seem to find a solution for the high move consumption though :/ ...
Last edited:
waffle=ijm
Waffo
lol stadler for 5x5 without actually making blocks.
cuBerBruce
Veteran
I don't necessarily completely solve the last 4 centers (before starting edges), but have at least horizontal lines on each of those centers, and preserve horizontal lines on those centers while solving the first 8 edge groups. Then I solve the last 4 centers before doing the last 4 edge groups.
CuBeOrDiE
Veteran
Robert-Y
Legend
I think Oliver Perge does the same thing btw.
I wonder how many moves are saved on average for solving the first 8 tredges like this and how many moves are required to restore the 4 centres...
qqwref
Legend
- Joined
- Dec 18, 2007
- Messages
- 7,834
- Location
- a <script> tag near you
- WCA ID
- 2006GOTT01
- YouTube
- Visit Channel
You can do this with pairing 6 edges. The problem is that only being allowed to use r, 3r, ..., 3l, l, U moves severely limits how efficient your centers can be, and since on larger cubes centers are most of the solve this will slow you down in the long run.
A related idea I had a while ago was to make all centers, and then allow basically yourself to mess them up in strips while solving the first 8 edges. So the idea there is you can do R2, F2, B2, L2 at any time no matter what the centers look like, and this should save a lot of moves on those edges. Then you can either fix the centers (only 4(N-2) pieces, it's quick) and do the edges, or - if you've left only the middle four edges to be solved - you can actually solve the middle part like a 3x3x(N-2) cuboid (well, if N is even, otherwise you have some flipped edges in the middle layer), which is probably faster since you can blockbuild and use domino algs and stuff. I haven't tried this enough to know if it can be as fast as (or faster than?) normal redux on big cubes, but it doesn't seem obviously slower.
A related idea I had a while ago was to make all centers, and then allow basically yourself to mess them up in strips while solving the first 8 edges. So the idea there is you can do R2, F2, B2, L2 at any time no matter what the centers look like, and this should save a lot of moves on those edges. Then you can either fix the centers (only 4(N-2) pieces, it's quick) and do the edges, or - if you've left only the middle four edges to be solved - you can actually solve the middle part like a 3x3x(N-2) cuboid (well, if N is even, otherwise you have some flipped edges in the middle layer), which is probably faster since you can blockbuild and use domino algs and stuff. I haven't tried this enough to know if it can be as fast as (or faster than?) normal redux on big cubes, but it doesn't seem obviously slower.
Dene
Premium Member
I thought of this quite a while ago.
Of course, at the time I didn't think of doing only the first 6 edges instead of 8 (silly me).
As you can see from the responses then, this has been thought of many times before in the past and has always been rejected.
In fact, PatrickJameson came out with the same idea a bit later.
My response to this method: I would not use it for anything bigger than 4x4. It could potentially be fast, but I don't think it can be faster than standard reduction.
Of course, at the time I didn't think of doing only the first 6 edges instead of 8 (silly me).
As you can see from the responses then, this has been thought of many times before in the past and has always been rejected.
In fact, PatrickJameson came out with the same idea a bit later.
My response to this method: I would not use it for anything bigger than 4x4. It could potentially be fast, but I don't think it can be faster than standard reduction.
CuBeOrDiE
Veteran
yeah, im starting to think the same thing
Kirjava
Colourful
I'm not scorching him for his idea.
They don't have the context that we have, this thread would get confusing otherwise. Luckily, I have provided it.
I still don't believe you, but that doesn't change anything - method discussion is fun regardless. ^_^
The 5x5x5 is a tricky beast. I've thrown all sort of weird ideas at it and nothing compares to redux.
The closest thing I found (aside from k4) is something like this;
LR centres
1x4x5 block or three 'cross' tredges
M centres
complete 4x4x5 block with intuition
complete F4L with intuition/pairing
pure LL
Then there's the n>4 k4 variation where you do all centres first, pair the cross, F2L with corners and midges and direct solve the rest. I'm not a fan.
And this is why my 5x5x5 times suck.
Robert-Y
Legend
One day, some someone will find and develop a method which is greater than reduction... maybe... 
CuBeOrDiE
Veteran
What are your 5x5 times?
Kirjava
Colourful
~2:00
Dene
Premium Member
Perhaps clicking the WCA profile button would help?
Kirjava
Colourful
You'd think that, but my results from the last competition are wrong for 5x5x5 XD
Similar threads
[Member Intro]
Just a simple cuber