• 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!

LeighSC

Member
Joined
Nov 1, 2013
Messages
18
Possible new 4x4 method for Roux?

a copy paste of a post I made on reddit:

So I had this idea, and I expect it to be picked apart very quickly, but I think that it could work.
I switched to roux recently, and wondered if I could combine hoya with roux, in a similar way to how yau can be combined with it, and I came up with this:
(just to help with my probably bad descriptions, count orange as the left block, white as the bottom, and blue on the front. just for convenience)

1) Solve 2 opposite, non block colour centres. (B/G or Y/W). So far, the same as Hoya. For the example, Y/W centres have been solved.

2) Solve the centres that will be your L block centre, and your B centre. So in the example. Orange for L and Green for B. To keep track of centre position, hold white (will be your bottom layer for 3x3) on the right, solve green, perform an x', solve the orange centre, and perform another x' for the next stage.

3) Do exactly as you would for a hoya cross, except build your first block (either by inserting the corners afterwards, or doing 2 edges, making a square, then the 3rd edge and 2nd corner). In the unused pair in the D layer, make any edge pair.

4) finish the centres as you would do on hoya. make sure to keep the block intact.

5) standard Yau/Hoya edge pairing. to avoid breaking the block, perform a 3U/' (correct notation? the top 3 layers clockwise or anticlockwise) instead of rotations to keep the corners of the block intact. after the first 3 pairs, no rotations/3U turns are required.
As an alternative,edge pairing on the M slice could be done, but I am not used to that.

6) solve roux. OLL parity could be used to get a nicer EO case for LSE, although your parity alg has to preserve corner orientation/permutation, which mine doesn't (currently)

So that's it. I think that the method is decent, and should at least compete with standard reduction with practice. Let me know how good you think it is currently, if you see potential in it, and how could it be improved. Also, if the method is decent, I want it named after me ;P

feel free to tell me how it is wrong, ineffective etc. I just think it would be pretty cool.

Thanks!
 

guysensei1

Member
Joined
Nov 24, 2013
Messages
5,143
Location
singapore
WCA
2014WENW01
Possible new 4x4 method for Roux?

a copy paste of a post I made on reddit:

So I had this idea, and I expect it to be picked apart very quickly, but I think that it could work.
I switched to roux recently, and wondered if I could combine hoya with roux, in a similar way to how yau can be combined with it, and I came up with this:
(just to help with my probably bad descriptions, count orange as the left block, white as the bottom, and blue on the front. just for convenience)

1) Solve 2 opposite, non block colour centres. (B/G or Y/W). So far, the same as Hoya. For the example, Y/W centres have been solved.

2) Solve the centres that will be your L block centre, and your B centre. So in the example. Orange for L and Green for B. To keep track of centre position, hold white (will be your bottom layer for 3x3) on the right, solve green, perform an x', solve the orange centre, and perform another x' for the next stage.

3) Do exactly as you would for a hoya cross, except build your first block (either by inserting the corners afterwards, or doing 2 edges, making a square, then the 3rd edge and 2nd corner). In the unused pair in the D layer, make any edge pair.

4) finish the centres as you would do on hoya. make sure to keep the block intact.

5) standard Yau/Hoya edge pairing. to avoid breaking the block, perform a 3U/' (correct notation? the top 3 layers clockwise or anticlockwise) instead of rotations to keep the corners of the block intact. after the first 3 pairs, no rotations/3U turns are required.
As an alternative,edge pairing on the M slice could be done, but I am not used to that.

6) solve roux. OLL parity could be used to get a nicer EO case for LSE, although your parity alg has to preserve corner orientation/permutation, which mine doesn't (currently)

So that's it. I think that the method is decent, and should at least compete with standard reduction with practice. Let me know how good you think it is currently, if you see potential in it, and how could it be improved. Also, if the method is decent, I want it named after me ;P

feel free to tell me how it is wrong, ineffective etc. I just think it would be pretty cool.

Thanks!
You could recognise OLL parity before solving CMLL, so that you can easily predict your CMLL case, while not caring about preserving the corners.
 

Vesper Sword

Member
Joined
Sep 7, 2013
Messages
72
Location
India
Possible new 4x4 method for Roux?

a copy paste of a post I made on reddit:

So I had this idea, and I expect it to be picked apart very quickly, but I think that it could work.
I switched to roux recently, and wondered if I could combine hoya with roux, in a similar way to how yau can be combined with it, and I came up with this:
(just to help with my probably bad descriptions, count orange as the left block, white as the bottom, and blue on the front. just for convenience)

1) Solve 2 opposite, non block colour centres. (B/G or Y/W). So far, the same as Hoya. For the example, Y/W centres have been solved.

2) Solve the centres that will be your L block centre, and your B centre. So in the example. Orange for L and Green for B. To keep track of centre position, hold white (will be your bottom layer for 3x3) on the right, solve green, perform an x', solve the orange centre, and perform another x' for the next stage.

3) Do exactly as you would for a hoya cross, except build your first block (either by inserting the corners afterwards, or doing 2 edges, making a square, then the 3rd edge and 2nd corner). In the unused pair in the D layer, make any edge pair.

4) finish the centres as you would do on hoya. make sure to keep the block intact.

5) standard Yau/Hoya edge pairing. to avoid breaking the block, perform a 3U/' (correct notation? the top 3 layers clockwise or anticlockwise) instead of rotations to keep the corners of the block intact. after the first 3 pairs, no rotations/3U turns are required.
As an alternative,edge pairing on the M slice could be done, but I am not used to that.

6) solve roux. OLL parity could be used to get a nicer EO case for LSE, although your parity alg has to preserve corner orientation/permutation, which mine doesn't (currently)

So that's it. I think that the method is decent, and should at least compete with standard reduction with practice. Let me know how good you think it is currently, if you see potential in it, and how could it be improved. Also, if the method is decent, I want it named after me ;P

feel free to tell me how it is wrong, ineffective etc. I just think it would be pretty cool.

Thanks!

IMO doing it till step 4 using yau and just replacing cross with first block and a random solved edge pair in place of the last cross edge is faster than this although this could also be pretty fast but I'm pretty slow at it which is why I use the yau approach instead.
 
Last edited:

TDM

Member
Joined
Mar 7, 2013
Messages
7,006
Location
Oxfordshire, UK
WCA
2013MEND03
YouTube
Visit Channel
Possible new 4x4 method for Roux?
Or, you could solve like Yau, except when you solve the three cross edges you solve the corners too, making a Rouxblock then. You then do centres like normal Yau, and don't do the last cross edge, but instead continue with normal edge pairing (although it helps to put a paired edge in where the last cross edge would usually go).
 

LeighSC

Member
Joined
Nov 1, 2013
Messages
18
IMO doing it till step 4 using yau and just replacing cross with first block and a random solved edge pair in place of the last cross edge is faster than this although this could also be pretty fast but I'm pretty slow at it which is why I use the yau approach instead.

The yau-roux method is what made me think of this. Hoya would be easier for me as I solve hoya. The hoya approach is basically hoya, but done differently. move count will be increaded by corners, but decreased by roux efficiency, and the random edge not having to be a specific pair.
 

LeighSC

Member
Joined
Nov 1, 2013
Messages
18
Or, you could solve like Yau, except when you solve the three cross edges you solve the corners too, making a Rouxblock then. You then do centres like normal Yau, and don't do the last cross edge, but instead continue with normal edge pairing (although it helps to put a paired edge in where the last cross edge would usually go).

That method is what inspired me. I solve Hoya though, so doing Yau would be hard to pick up again, plus I dislike Yau centres, so a hoya equivalent would be nice.
 

guysensei1

Member
Joined
Nov 24, 2013
Messages
5,143
Location
singapore
WCA
2014WENW01
My new 'method' is a combination of yau5 and K4 for big cubes.
1) solve 2 opposite centers
2) build 3 cross edges and match them with the center
3) solve the rest of the centers
4) finish the last cross edge and place it in to complete the cross
5) freeslice the 4 F2L edges
And here is where it differs from yau5.
6) finish all 4 F2L pairs (instead of 2 in yau5)
7) solve LL corners with CLL
8) solve the remaining edges with commutators like in K4.


Is this feasible?
 

Hypocrism

Member
Joined
Aug 4, 2012
Messages
316
WCA
2009ADLA01
My new 'method' is a combination of yau5 and K4 for big cubes.
1) solve 2 opposite centers
2) build 3 cross edges and match them with the center
3) solve the rest of the centers
4) finish the last cross edge and place it in to complete the cross
5) freeslice the 4 F2L edges
And here is where it differs from yau5.
6) finish all 4 F2L pairs (instead of 2 in yau5)
7) solve LL corners with CLL
8) solve the remaining edges with commutators like in K4.


Is this feasible?

From my perspective, the centres are inefficient in yao variations for big cubes, and the LL will be either difficult or inefficient. But it's plausible.
 

lerenard

Member
Joined
Sep 5, 2014
Messages
274
Location
Tennessee
Petrus/Roux combo

So I love Petrus, but I find that while I can complete everything after the. 2x2x3 block very quickly (I actually use 2 look cfop ll) I find the 2x2x3 block difficult to complete quickly, especially because it's hard to find all the pieces. I love how easy it is to find the pieces in Roux, however. Thus, what if you made 2 2x2x1 blocks in BLD and BRD. Then you just place the BD edge piece (extremely easy to find and place) and then you move onto EO like normal Petrus. I don't know if anyone will like this idea, and I haven't even practiced it enough to have it be faster than normal Petrus, but I think it could be very effective. Tell me what you think.
 
Last edited:

GuRoux

Member
Joined
May 6, 2013
Messages
1,712
Location
San Diego, California
WCA
2014TANG03
YouTube
Visit Channel
So I love Petrus, but I find that while I can complete everything after the. 2x2x3 block very quickly (I actually use 2 look cfop ll) I find the 2x2x3 block difficult to complete quickly, especially because it's hard to find all the pieces. I love how easy it is to find the pieces in Roux, however. Thus, what if you made 2 2x2x1 blocks in BLD and BLR. Then you just place the BD edge piece (extremely easy to find and place) and then you move onto EO like normal Petrus. I don't know if anyone will like this idea, and I haven't even practiced it enough to have it be faster than normal Petrus, but I think it could be very effective. Tell me what you think.

that's seems like a good idea to reduce cube rotations and still maintain movecount efficiency. maybe also 1x2x3 to 2x2x3 might be better.
 

TDM

Member
Joined
Mar 7, 2013
Messages
7,006
Location
Oxfordshire, UK
WCA
2013MEND03
YouTube
Visit Channel
Suggestion: add some 'frequently suggested methods' to the OP to hopefully stop people posting the same thing again and again?
(stuff like Roux where you do the D layer edges then solve pairs like CFOP, WV+CP -> E(P)LL, WV -> 1lLL, opposite-swap cross, keyhole etc.)
 
Last edited:

elrog

Member
Joined
Jan 31, 2013
Messages
518
Location
U.S.A.
YouTube
Visit Channel
So I love Petrus, but I find that while I can complete everything after the. 2x2x3 block very quickly (I actually use 2 look cfop ll) I find the 2x2x3 block difficult to complete quickly, especially because it's hard to find all the pieces. I love how easy it is to find the pieces in Roux, however. Thus, what if you made 2 2x2x1 blocks in BLD and BLR. Then you just place the BD edge piece (extremely easy to find and place) and then you move onto EO like normal Petrus. I don't know if anyone will like this idea, and I haven't even practiced it enough to have it be faster than normal Petrus, but I think it could be very effective. Tell me what you think.

This has been thought of many times and I believe it is the most efficient way to build a 2x2x3 block at the start of a solve.

@ TDM: Will do.
 

lerenard

Member
Joined
Sep 5, 2014
Messages
274
Location
Tennessee
I take it you mean BLD and BDR.
it seems like a good idea.

Oh yeah xD thanks for catching that.

@GuRoux: Now that's an idea! I will most definitely be trying that out. And the cube rotation thing was another reason I thought of doing it this way.

@elrog: Well, that gives me two thoughts immediately:
1) I must be smart/creative for having thought of it on my own then! :D
2) If it's been thought of many times, why haven't I ever read anything about it?? Stupid Internet doesn't solve all my problems.
 
Last edited:

davidx233

Member
Joined
Sep 28, 2013
Messages
5
Location
Millburn, New jersey
Possible Equator Method

I have recently been researching lots of speedcubing methods.(I solve Using Petrus) and I was wondering, is there a way to solve a 3x3 using an equator method(Solving the whole middle layer first)? It. Is easy enough to solve the middle layer, but after that, I don't know what to do. Are there any algorithms or does anyone have any ideas? By the way, this is just for fun and not to speed myself up.
 

goodatthis

Member
Joined
Jan 11, 2014
Messages
841
Location
NY
WCA
2014CAVA01
YouTube
Visit Channel
So first there is already a thread for ideas like this, it's called The New Method/Substep/Concept Thread.

and an excerpt from the OP in this thread:

Do Your Research

There are a lot of different methods out there. Please try to make sure your idea is new/original before posting. You should check out the methods pages on the wiki.

Here is a list of commonly suggested methods:
Belt - Anything that solves the cube like this (the belt does not always have to be made first / many times EO is solved with it). This is a broad category and there is a large variety of belt methods already out there and there is a good chance you will be repeating something.

I know it sounds good at first, but believe it or not, tons have people have "invented" this method before, including myself. Luckily I found out about this before I posted anything about it.
 

davidx233

Member
Joined
Sep 28, 2013
Messages
5
Location
Millburn, New jersey
I have a possible idea. You solve a 2x2x3 block on the bottom of the cube. Then solve the other two first layer corners. From this point I only have an idea of what to do. You have to place the four corners on the top lavers. They don't have to be correctly oriented though. Then you would just solve the edges. I don't know the algorithms for this because I am a Petrus solver. Does anyone know how you could solve the cube from here.
 
Top