Solving the Edges
Step 1: Centers | Step 2: Edges | Step 3: Fix parity
Intro | 2 pair chain solving | Other pairing methods

6 pair solving

This page definitely requires a disclaimer, so here it is. If you're reading this I assume you are interested in learning the 6 pair method, but you need to realize that I am no an expert in this method. The best I was ever able to average with the 6 pair method is sub-1:20, and I don't even think I ever made it to sub-1:15 with it. I use the 2 pair "chain" method now to average sub-1:10 frequently, and there are 6 pair solvers who also do the same. So bear in mind that my 6 pair advice might not be able to get you to sub-1:10 averages. However, other than the videos with Frank Morris on I haven't found any really good 6 pair tutorials on the internet at the time I am writing this page. I did spend a lot of time in winter of 2003 into spring of 2004 trying out the six pair method though, so here is at least one more online tutorial for this method.

Again, using what I will write on this page I was able to average between 1:15 and 1:20 for solving the 4x4x4, but please bear in mind that I am not an expert in this method. If you would like to ask tips from an expert please contact those at the top of the 4x4x4 list, or post on the yahoo group.

Even though I use the 2 pair method almost exclusively, I will be as fair as I can and try to show everything I learned about 6 pair solving from when I was still trying it out.

Setting up your cube

If you have just come from the centers page your cube is already correct, skip to the edges stuff below.

If you have the same color scheme as I do then scramble with yellow on top and green on front. To get to this scramble from a solved cube do:

b F (Ll) u2     B U' F2 R B     F' l2 D2 l2 B     L B' d2 b r     l2 f u' f ' R     D' B u2 B' R'     u F2 b2 U2 (Ff)'     F' R' D r d     F2 U2 (Rr) F B     (Rr) L2 (Bb)2 F (Uu)'     F2 (Uu) (Bb)' D' R'     (Bb) D' (Bb) L (Bb)'     L (Bb) R2 (Bb) L2     U (Bb) L (Bb) L2     (Bb)' y

Now your cube should look like this:
Front/Top View
Back/Bottom View

If your cube does not look like the diagram, please solve and rescramble with that alg above. I know it is long, but I double checked and it does get to the position shown here.

Solving 6 pairs at a time

What we are going to do for this method is solve anywhere from 5-8 pairs at a time, though generally we will solve 6. However, you can't do this twice and solve all 12 edges very often. Rarely will a solve work out that well. Generally what you are going to do is to solve 6 pairs right at the start, then try to solve 4 pairs, then finish off with whatever is left with the 2 pair method. The benefits of this method are that you solve the edges in relatively few steps, and on very rare cases you can knock out 8 edge pairs in one go! The downside is that on the cases where you can only solve 5 pairs, you had to have done the same number of moves as you would do to solve 6 pairs, so you end up wasting a few moves, which translates to wasting a second or two. So like the 2 pair method, it all sort of balances out.

Ok so we've just finished solving the centers. Now in the last few moves for the centers you should be looking for two edge pieces that belong together when the cube is solved. While finishing the centers I probably would have noticed the blue white edge pieces first. These pieces are in dFR and bDR. The first thing we need to do is to put them into the middle layers. It is in the middle layers (u and d) where we will pair up edges. The only real way to describe this is to just jump right into an example, so let's start with blue/white.

When I notice a pair I have a very regimented, unwavering set of recognition steps. I'll detail those here.

Top View
Bottom View

The above diagram just shows you where the blue/white pieces are, what I would actually notice however, is this,

Top View
Bottom View

Following the numbering scheme first I would notice the white/blue piece at dFR, since it stands out to me for some reason. I would then twist the cube around looking for the other blue/white, which is in bDR. Whenever I find a piece that I was looking for, such as the blue/white, I always look at the piece that is next to it. In this case that piece is green/white.

Now I need to get the blue/white that is in bRD into the middle layer such that blue/white piece is in the u layer. To do this I would do D2 L'.

In order to make the solve smoother though, while I am doing the moves to setup the two blue/white pieces, I go ahead and start looking for the other green/white edge, since I will want to pair up green/white as well.

Front/Left View
Front/Right View

By the time I would have completed the (Dd)2 move I would have found the other green/white at fRU. I also would have noticed that blue yellow piece that is next to it. Remember, when looking for a specific piece on the cube, once you find it always pay attention to which piece is next to it.

So far we haven't yet done our L' move to get blue/white ready to be solved. Also, we are going to want to place the other green/white piece into the BL spot, such that the green/white pieces will line up in BL as well. See the diagram below,

Notice that the alg we need to do to setup the pieces this way is L' to ready blue/white then B' U' B to ready green/white. Doing this move will set us up for solving two pairs on our first (Dd)' turn. We will solve blue/white in FL and green/white in BL. However we want to do three pairs on the first turn, and three pairs on the return. So we need to find the other blue/yellow piece while doing the alg to setup blue/white and green/white.

So go ahead and do the L' to setup blue/white. Now I would rotate my cube with y' (if you don't know rotational notation that means to spin the whole cube like your were doing the turn U'). Now I would do the alg above translated as L' U' L. I didn't notice the other blue yellow until the last L move, and it is in lDF. Also I would have noticed that white/orange is next to the blue/yellow we just found.

We need to place that blue/yellow edge into the BL spot so that on our first (Dd)' turn, the blue/yellow edge pair will form as well. To do that do the move L' D' L. Afterward spin the cube with y' (like U') again so we can track the white/orange piece better.

So here is the setup, with a (Dd)' move we will pair up 3 pairs, at FL, FR and BR.

Now in order to solve on the return d layer turn, we have to find the d layer piece in our setup that does not pair with it's mate in the u layer. This piece is white/orange.

Now go ahead and do your (Dd)' turn, but at the same time glance around frantically for the other white/orange piece.

The white/orange piece is on the back at lBU.

Now we are going to do the same thing, but starting with the white/orange edge. This will allow us to pair up 3 more edges as we fix our d layer.

Notice that the piece next to the white/orange at lBU is red/yellow. So as we are placing the white/orange piece into the u and d layers such that it will pair with the other one when we do (Dd), start looking around for the red/yellow piece. The move we need to do to line up the two white orange pieces is F U2 F'.

As you do this move you will probably spot the other red/yellow piece in the U layer. It ends up in UL after the move to ready white/orange. Also notice that next to the U layer red/yellow is green/orange.

To ready red/yellow to solve we will want to do R U2 R'. Now while we are doing that move, remember to look around for the other green/orange. Don't look too hard in the U or D layers though, because you won't find it. The other green/orange is in uBL. I don't know of a way to handle cases like this very smoothly, so here is what I am going to do.

Go ahead and do, or finish, the R U2 R' move to ready red/yellow to solve at FR right now. Now we are going to replace the solved white/blue edge at BR with any random edge, but remember which piece you end up placing into the d layer in BL. So track which piece ends up in dBR. I personally would rotate the cube as y (like U) and do R U' R'. This will insert the red/white and green/red mismatched edge pair such that the red/white piece ends up in the d layer of that spot.

Front/Left View
Front/Right View

Now we're going to step out of our 6 edge solve and use a little trick to fix this situation. Notice that doing a (Dd) turn will pair up two edges. This is the end of a regular 6 pair technique. However, ideally the edge you need for the last pair won't already be in the middle layer. So notice if we end the regular way we will only solve 5 pairs, this is the worst case scenario. But, we are going to use a little trick to salvage this situation and actually solve 7 pairs.

Notice that doing a (Dd)2 will pair up those problem green/orange pieces at BR. Let's go ahead and do that now.

Front/Left View
Front/Right View

Now our two edges that we've already setup are still setup to be solved, but we're going about solving them after we take care of this problem case of green/orange.

Remember how when we replaced our last solved edge with random edge how we marked which piece would end up in the d layer? That piece was red/white, and I've marked where the other red white is. Now if you replace our now solved green/orange edge group with the one containing red/white, such that the red/white piece ends up in the u layer, our center fixing d layer turn will solve three additional edges. That's 7 total!

So first replace green/orange with the red/white edge like this, B' D B. Now if you do a (Dd)' turn you will solve three more edges. Do that now.

So we took our worst case scenario, and luckily we were able to efficiently turn it into one of our better scenarios, solving edges at once.

Now usually we would continue from here with the 4 pair method, in this method you only pair up two edges on the first d layer turn, then you pair up two edge on the return turn. This is done in the same way as the 6 pair idea, only with one less edge at each step.

However, we started with two edges solved at the beginning of this step. So we had two solved, and we just solved 7, that leaves 3 edges left to solve.

From here we have to finish with the 2 pair method, since it will pair up the last three edges at the same time. First notice the blue/orange piece at dFR. This piece was attached to our red/white piece during our earlier 6 pairing, and it ended up at dFR. So now glance around the cube and look for the other blue orange piece.

Luckily it is in a good location at fDL. Also, the edge it is next to is orange/yellow.

Front/Left View
Front/Right View

From here do the move L' to get blue/orange ready to pair up with a d layer turn. Now as you are doing that look around the cube for the other yellow/orange piece. It is in bDR.

After you do the L' move go ahead and pair up the blue/orange edge group at FL by doing (Dd)'. Now replace the solved edge group with the edge group containing the other yellow/orange edge, such that the yellow/orange ends up in the u layer. So do the move L D' L'. Now fix your centers with (Dd) and the edges are done!

Front/Left View
Front/Right View

Shortfalls of my explanation

Again I'm no expert of the 6 pair method, so I don't have an efficient solution handy every time if we get a weird case like we had in this example solve. Now, pairing up the edge pairs can be viewed as transposing piece between edge pairs, thus solving them. This can always be simplified down to a maximum of 12 transpositions, so you're left with cycles exactly like in BLD solving!

Now what happens if you don't have a 6-cycle of edges, yet you are working on trying to solve 6 pairs at a time with that cycle? That is when you run into problems like we did above. Now sometimes, if you can look ahead well, you can see a way to fix that. I'm not very good at seeing fixes like that quickly, which is why I prefer the stability of the 2 pair method. Solving with 2 pairs uses many more substeps but it has no bad cases.

So in short, if you plan on using the 6 pair method, be ready to improvise when you try to use the 6 pair method on a cycle of edges that is shorter than 6 in length. This will give you a bad situation that needs to be fixed, and on a worst case scenario you would only solve 5 edges.

For a video explanation of the 6-pair method see the 4x4 edges page.
Step 1: Centers | Step 2: Edges | Step 3: Fix parity
Intro | 2 pair chain solving | Other pairing methods
Go back