[Help Thread] 4x4 Discussion

Randomno said:
* gets to ELL * * gives up *
Click to expand...

So many crossed wires in this conversation it's hurting my brain... You are doing layer by layer, not the reduction method that AmazingCuber is describing. In reduction, you don't complete each layer in turn, what you do is you first pair every edge piece with it's matching one (you don't have to solve them in the process, just match them up), so that the cube is 'reduced' to something that can be solved using 3x3 techniques. You are trying a layer by layer approach, which requires you to use ELL to pair up the top layer edges.
ELL is the step you are looking for. To get a grip of ELL, try doing R U R' U', inner R slice, U R U' R', (inner R slice)', on a solved cube and see what happens (also, try mixing it up by using the left inner slice instead, or doing the mirror, so L' U' L U, (inner L slice)', U' L' U L, inner L slice). Then, try and use that to either completely solve the last layer, or get to a 'parity case', where you either have one 'full' edge flipped, or two edges swapped, then apply a parity algorithm (they're in the '2-cycles' bit of the ELL page) to fix the parity. It is a largely unavoidable part of 4x4 solving, you just have to practise and deal with it. :)

Will be reporting posts for being off-topic, but wanted to at least try and give some hopefully useful help before then. Maybe this can all be shifted to a 4x4 thread or something, I dunno, that's for the mods to decide. :)
 
ryanj92 said:
So many crossed wires in this conversation it's hurting my brain... You are doing layer by layer, not the reduction method that AmazingCuber is describing. In reduction, you don't complete each layer in turn, what you do is you first pair every edge piece with it's matching one (you don't have to solve them in the process, just match them up), so that the cube is 'reduced' to something that can be solved using 3x3 techniques. You are trying a layer by layer approach, which requires you to use ELL to pair up the top layer edges.
ELL is the step you are looking for. To get a grip of ELL, try doing R U R' U', inner R slice, U R U' R', (inner R slice)', on a solved cube and see what happens (also, try mixing it up by using the left inner slice instead, or doing the mirror, so L' U' L U, (inner L slice)', U' L' U L, inner L slice). Then, try and use that to either completely solve the last layer, or get to a 'parity case', where you either have one 'full' edge flipped, or two edges swapped, then apply a parity algorithm (they're in the '2-cycles' bit of the ELL page) to fix the parity. It is a largely unavoidable part of 4x4 solving, you just have to practise and deal with it. :)

Will be reporting posts for being off-topic, but wanted to at least try and give some hopefully useful help before then. Maybe this can all be shifted to a 4x4 thread or something, I dunno, that's for the mods to decide. :)
Click to expand...

I'd suggest asking for this convo to be moved to a different thread and I'll reply there.
 
So I'm really just starting out at 4x4. I've not done any timed solves yet, but I'm using Yau. After cross edges, I use 3-2-2 edge pairing, which I love thanks to the regular gifted final pair :).

Anyhoo, my question is really does anyone have any tips for centres? Everything I've done so far I've figured out myself, but what I do really is build half-centre blocks of 2, then switch these around to make full centres. I think it's pretty inefficient but, other than complete walkthrough solves, I can't find much guidance specifically on L4C.

The times when I find it most inefficient are early on, when I have no centres built, and I tend to find myself just ambling about, looking for things to work with, and I'll start building blue, then spot some orange that looks good, and build orange, then maybe green, then back to blue... Ooh, forgot red, and I've got random single red centre pieces lurking in 2 other centres...

Just an inefficient mess. I know I'm doing it, and doing it OK, but I think I could do much better. I guess picking a starting colour, and sticking with it, would probably help!? I can't seem to find any centres-specific tutorials etc., so any guidance/links/etc. would be fantastic.

TIA.
 
pipkiksass said:
So I'm really just starting out at 4x4. I've not done any timed solves yet, but I'm using Yau. After cross edges, I use 3-2-2 edge pairing, which I love thanks to the regular gifted final pair :).

Anyhoo, my question is really does anyone have any tips for centres? Everything I've done so far I've figured out myself, but what I do really is build half-centre blocks of 2, then switch these around to make full centres. I think it's pretty inefficient but, other than complete walkthrough solves, I can't find much guidance specifically on L4C.

The times when I find it most inefficient are early on, when I have no centres built, and I tend to find myself just ambling about, looking for things to work with, and I'll start building blue, then spot some orange that looks good, and build orange, then maybe green, then back to blue... Ooh, forgot red, and I've got random single red centre pieces lurking in 2 other centres...

Just an inefficient mess. I know I'm doing it, and doing it OK, but I think I could do much better. I guess picking a starting colour, and sticking with it, would probably help!? I can't seem to find any centres-specific tutorials etc., so any guidance/links/etc. would be fantastic.

TIA.
Click to expand...

Doing half-centers as opposed to one center at a time is fine. It's what I do, and it can be plenty efficient once you're good at it. I'll point out something you don't seem to have thought about yet. When you've solved three of the centers, the fourth will automatically be solved since the pieces have nowhere else to go. So when you are building half-centers, you can focus entirely on building them for three colors without ever bothering with the fourth color at all. Here's a general structure to follow: First, solve one half-center of each of three colors. Make sure that when you finish this step, those three are sitting on the left side of the F, D, and B faces in the proper order dictated by color scheme (don't worry about matching the cross just yet though, you might have picked a different three colors than the three cross edges you solved). This will leave the whole r-slice free so you can solve the remaining half-center of each of those three colors entirely with <Rw,U>.

This method takes a lot of practice and experimenting to get efficient at, but you must focus on efficiency before speed. Realize that it is not always best to follow the strict pattern of (1st half of three centers) -> (2nd half of those three). It will often be possible to preserve extra bars that you need for the second step (the second bar of a color you've already put in place in the first "layer") while you're still finishing the first step, or even to insert an extra U move somewhere to set up an extra bar or two to be formed. And occasionally you may find it best to ditch that strategy entirely and finish in the style of the one-center-at-a-time approach. Also note that although the three bars in each step need to *end up* in the correct order, they don't need to be built in that order.

I'm not going to try searching for a video to show you, both because I don't even know if there's a good one (I've never watched any 4x4 video tutorials), and because I don't think it's needed. You'll understand things better if you discover them yourself through experimentation. It's important that you don't let yourself stick to too strict of a pattern, and that's something it would be difficult for any video to teach.
 
uberCuber said:
I'll point out something you don't seem to have thought about yet. When you've solved three of the centers, the fourth will automatically be solved since the pieces have nowhere else to go.
Click to expand...

I had thought about it, but it's normally something I don't consider until I have only 2 centres remaining - I hadn't considered just ignoring a colour from the start. Generally I'll just be indecisive and flitter between the 4 centres! I think a bit of decisiveness, and focussing on just 3 colours, should help a lot!

uberCuber said:
This method takes a lot of practice and experimenting to get efficient at, but you must focus on efficiency before speed.
Click to expand...

Yeah, tis why I've not timed any solves yet :-) This method makes a lot of sense to me, in terms of efficient evolution of my own method.

uberCuber said:
Realize that it is not always best to follow the strict pattern
Click to expand...

Yeah, similarly to edge pairing, where I'm generally using 3-2-3, but being flexible where I can/need to be. It's one thing I really enjoy about 4x4 - that you actually need to think about what you're doing. For me at least, every part of 4x4 is still intuitive, up to the 3x3 stage and fixing parity.

Many, many thanks for your detailed response and suggestions. :-)
 
Last edited:
I am currently using yau for 4x4, but I tend to have issues with last four centers. I have thought about and looked at videos on them, but have yet to find a good way to solve them for me. Does anyone know of any videos of methods or explanations of methods that could help me get faster?

Also, does anybody know of a good way to distinguish if you have oll parity while you are still doing F2L? I tend to have issues with this and knowing ahead of time might cut a few seconds off of my time because I don't have to orient the top layer to set up any easy OLL case. Thank you.
 
Awesomecuber27 said:
Also, does anybody know of a good way to distinguish if you have oll parity while you are still doing F2L? I tend to have issues with this and knowing ahead of time might cut a few seconds off of my time because I don't have to orient the top layer to set up any easy OLL case. Thank you.
Click to expand...

You can look at how many oriented/unoriented edges you have. Oriented edge = one that can be solved without using F and B quarter turns (of course, rotating may changes orientation). Look up a ZZ tutorial for more information.
In general, people kind of have a sense for oriented edges by the 3rd or 4th pair just through repetitive solving and becoming familiar with patterns and cases. I don't think you need to worry about it too much - trying to use "rules" for whether an edge is oriented or not will take quite a bit of time, especially at first, and not really save you very much time overall.
 
I am averaging around 1:20 on 4x4 for about 1 month. I am using yau and doing first 2 centres in about 7 sec,three cross edges in about 16 sec,last four centres in 12-13 sec,last cross edge 5-6 sec,last eight edges-15 sec and 3x3 stage around 25-30 sec depending upon parity.How can I improve on these substeps?
 
Awesomecuber27 said:
I am currently using yau for 4x4, but I tend to have issues with last four centers. I have thought about and looked at videos on them, but have yet to find a good way to solve them for me. Does anyone know of any videos of methods or explanations of methods that could help me get faster?
Click to expand...

I always hated using Yau...try the Hoya method, it is much easier for lookahead (at least for me). It is the same concept but only requires you to solve 2 ceters after pairing edges.

**Edit: forgot your video :)

http://www.youtube.com/watch?v=TwH9LUnmtCM
 
Last edited:
You must log in or register to reply here.

Similar threads

Turns_And_Twists2024
4x4 OH discussion
2
Replies
24
Views
2,532
AJT17
AJT17
C
  • Poll Poll
4x4 General Discussion
2
Replies
22
Views
2,760
Artem Tabaliuk ( Артём )
Artem Tabaliuk ( Артём )
Kenneth
  • Sticky
Blindsolving Discussion and Help
294 295 296
Replies
5,908
Views
729,739
abunickabhi
abunickabhi
Robert-Y
[Help Thread] Yau4 Discussion
3 4 5
Replies
80
Views
27,849
MeSub20
MeSub20
B
[Help Thread] Puzzle Setup / Tension Discussion
28 29 30
Replies
592
Views
93,781
Fukuoka Kengo James
Fukuoka Kengo James
Back
Top