• 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!

Briggs2 (B2) (3x3x3 CP method)

shadowslice e

shadowslice e

Member
Joined
Jun 16, 2015
Messages
2,924
Location
192.168. 0.1
YouTube
Visit Channel
My latest method developed. It also has an improved CP step.

1) CP+FB (12/12)
2) 1x2x2 in BRD (equivalent to Roux square) and a pair (8/20)
3) BLS (Briggs last slot)- finish SB while solving CP and forcing a 4 flip. (10*/30)
4) LSE (10/40)

*This is the average of the algs I've generated so far which does not use wide moves (for the most part).
So this step is likely by far the most complex- the reason why it is step 1.

First, you choose the side you wish to build FB (duh) and from there you can plan the first square as normal. You then must predict which piece is in DFR and in the final spot in the FB (DFL) then place the final FB pair in in ULF, the DFL opposite in DFR and DBR opposite in UFR. Now do an F'. You can also do the mirror and use F or B or B' etc. You get the point.

Sorry if this sound complicated. It's not really as bad as it sounds. I'll do a video later this week to clear up any confusion. In the meantime, ask anything you want in the thread.
You can actually have a series of different positions based on the different 2-gen corner permutations
This is pretty self explanatory isn't it? Make sure that the pair is with an oriented edge and that you only have one misoriented edge in D so that BLS works

Unless you want to do mirrored BLS in which case you could also do the square in DFR. In which case you have an even more efficient step (although I may have been slightly optimistic and you may get the same efficiency)
This step is only possible due to CP first step. It is essentially a modified form of WVLS so you orient the corners with the last slot as well as performing either a 4Flip or 0 flip on U so that you get an arrow 4flip for LSE which is arguably the best case.
There are 54 algs for all the cases and they average 10.3 moves. It can be more efficient as well as more ergonomic as the algs I have generated do not use wide moves (mostly).
There is a list of them is the spoiler below. The designation of each alg reflects the number of oriented corners as well as the orientation of each starting from BL. For example, E1xxr indicated the edges are oriented, 1 corner is misoriented , BL is oriented, BR is oriented and FR is facing to the right (equivalent to blinkers).
E0 R' F L' F2 R2 U L' U R' U2 L2 F

E1lxx R U R' U' R U' R'
E1bxx (U') R U' R' U2 R U' R' U2 R U R'
E1xbx SA- R' U' R' U2 R U2 R' U' R'
E1xrx (U2) R' U2 R' U2 R U2 R' U' R2 U R
E1xxr R U D' R D' R U R' D R' D R'
E1xxf R U' R'

E2lbx (U) R' U' R2 U' R2 U2 R
E2lrx L' U2 R U L U' L' U R' U2 L
E2bbx (U) R' U' R2 U' R2 U2 R
E2brx (U) R U' R' U R U2 R'
E2xbr (U) R2 U' R U L2 D' R D' L2 U' R
E2xbf R' U L' U L' D2 L U2 R U L
E2xrr (U) R U2 R'
E2xrf (U2) R' U2 R U R U' R' U R ' U R2 U R'
E2lxr (U) R U2 R' U' R U2 R' U' R U' R'
E2lxf (U) R U2 R2 U' R U' R' U2 R
E2bxr R U L' U2 L U L' U R' L
E2bxf (U2) R U R' U R U' R U2 R' U' R U' R2

E3lbr R U R' U' R U R' U' R U' R'
E3lbf R U' R2 U' R U' R' U2 R
E3lrr (U') R2 U R' U R' U' R U R U2 R2
E3lrf (U) R U2 R' U R' U' R U' R' U2 R
E3bbr (U2) r U2 r' U2 R U' R'
E3bbf (U) R U' R' U R U' R' U R U2 R'
E3brr R U' R2 U2 R U R' U R
E3brf (U) F U2 L' U2 L F2 R' F R
SA= standard approach- alg above M' U M
M0xxx (U) R' U' F' U F2 R F' U F R' F' R

M1lxx SA
M1bxx L' U' R U B' M2 U2 R B' L' D x2- this alg needs fixing. It's by far the worst of the sets.
M1xbx (U) F R U2 R2 U' F' U F R F U2 F'
M1xrx F' U' F U R U2 R2 F R U F U' F'
M1xxr SA
M1xxf (U2) F' U' L D F' D' L D' L U L F

M2xbr F2 U2 L U F' L F2 L' U' L' F
M2xbf (U) F' L' U2 L U F U ' F2 L F L' F'
M2xrr SA
M2xrf (U2) F2 U' F2 U F2 U' F R U2 R2 F R (probably better to do this with a y')
M2lxr (U) L U2 F' U2 R' L F R F' L2
M2lxf (U2) F2 L2 F L' F U' R U2 R' F' L' F (although SA is probably faster)
M2bxr F' U2 L F U F' U' L2 U2 L U F (another y')
M2bxf R' F' L2 D F D2 F' D R L U' L F
M2lbx SA
M2lrx (U) F R' F2 U' F L2 F D F' L2 U R
M2bbx (U') F R U' R2 F R2 U2 R' U' F'
M2brx SA

M3lbr (U) R' F' L F' R' D2 L2 B R2 L
M3lbf SA
M3lrr F2 U F L' U R U' R' U2 L U F
M3lrf SA
M3bbr SA
M3bbf (U2) F' U F' L' U' L U F2 R' F U F' (another y')
M3brf (U) F R' F' R2 U2 R2 U' F' U F R
This step is basically identical to roux LSE although you always get an arrow 4 flip which is arguably the best case as it is easier to pair up either UF and UB or UR and UL while orienting the edges.
Lower alg count
Low movecount
Ergonomic (especially if some of the BLS algs get fixed)
Fairly good lookahead
Advanced and somewhat complex first step
Algs are pretty specific to the method.
TL;DR: sort of an advanced Roux

I would love to hear your thoughts, post any questions in the thread and I will do my best to answer them. I will also try to make a video soon with a walkthrough of first step.
 
I don't understand your CP+FB at all, or your BLS recognition. Please could you go into a bit more detail? Possibly an example solve would be good too. Text is fine for me.

also pls tell me how to do SB square+pair in 8 moves and I will love you forever
 
TDM said:
I don't understand your CP+FB at all, or your BLS recognition. Please could you go into a bit more detail? Possibly an example solve would be good too. Text is fine for me.

also pls tell me how to do SB square+pair in 8 moves and I will love you forever
Click to expand...
The CP step is essentially reducing the cube to 2-gen while inserting the final pair. On a 2-gen cube there are 3 distinct sets of 3 2 swaps which can occur (discounting mirrors or the order which the two swaps are in): (UBR/UFL,UFR/UBR and DBR/DFR), (UFL/DFR, UBL/DBR and URF/URB) and (UBR/UBL,UFL/DBR and UFR/DFR) the sets of colours would be (if white on top green in front) yellow,green,red with yellow,blue red, green red white with orange green white and white green orange with white blue red. My personal preference is the last set as I find it easiest to convert the swaps to although any of the sets can in theory be used- this would be difficult to do though it can help with the SB in some cases. The F/F' has the pieces on it to convert to this.

With regards to the SB I'll admit I was slightly optimistic as I was going off 6 for square+ 2 for pair. Maybe more realistic would be 10-12 in total but I do mention rotations and reflctions being one way to get to that efficiency for BLS. To counter balance this I was also slightly pessimistic about FB+CP.

The way that I described each of the cases for BLS is basically indicating how each of the corners is oriented. For example, do a R U2 R' U'. This would be noted as E2xrr. Tbh you only have to pay attention to the last 3 letters as they tell you the orientation of the corners. You also recognise whether there is one misoriented edge or 3. It's similar to dot OLL recognition but you recognise a slightly different set of pieces.

I hope this helps a little bit.


Sorry about the lack of example solve but it's late so I'll do one in the morning because I can't deal with notation right now (and everyone know how bad I am with that...)

There's a sort of example solve in the example solve thread although that doesn't do FB+CP as it already had permuted corners (which incidentally makes FB much more horrible).
 
Last edited:
Ok, example step 1 solve with explanation (probably won't be particularly efficient as I'm not combining steps so it's easier to understand (and cause im tired as funk...)

Scramble: D2 B2 R2 F2 D B2 F2 U L2 D2 F2 R U2 F D L F R' F' D' R

x'// just making FB more fingertrickable (note, I will be predicting the positions of the corner that will be in FRD and BLD
M2 D'// First square as per normal
R U R' U' R// pair
U2// pair in place for solving CP
R' (F' U F) U' R just a standard 2 swap so all pieces are set up.
(B)// Finished CP
 
Last edited:
shadowslice e said:
Ok, example step 1 solve with explanation (probably won't be particularly efficient as I'm not combining steps so it's easier to understand (and cause im tired as funk...)

Scramble: D2 B2 R2 F2 D B2 F2 U L2 D2 F2 R U2 F D L F R' F' D' R

x'// just making FB more fingertrickable (note, I will be predicting the positions of the corner that will be in FRD and BLD
M2 D'// First square as per normal
R U R' U' R// pair
U2// pair in place for solving CP
R' (F' U F) U' R just a standard 2 swap so all pieces are set up.
(B)// Finished CP
Click to expand...

was wondering why the solution wasn't working
 
Oh, just a sidenote (for the moment, what I'm about to describe is promising so may end up being in the final finished method): you can actually use Pseudopair insertion in a way similar to PCMS except the pairs are off by a D or D'. This is useful as you can insert 3 pseudopairs, do one short thing to the corners and do a D/D' to finish off the whole first two steps (so have a guess what I'm going to call the method :p). It seem like this method could be only 15-18 moves for F2B (though not the final pair) which, for obvious reasons, would be insane and reduce this to a 3-step method. Atm I'm trying to find ways to simplify the process but I find it is looking like a real contender for the new way to start the method (and give it even greater efficiency!) :)

Updates shortly.
 
This is a somewhat cool concept, but honestly I preferred your original version. CP-Roux honestly isn't very useful, if at all. The main advantage of the original Briggs Method was the awesome 2-gen F2L/LL. Without anything like that, this method just become a Roux variant that solves CMLL while inserting the last piece of the SB with a more complex FB. Don't know if that's a good trade off. Personally, I don't think L6E has the potential to be any faster than maybe 2-3 seconds. I know GuRoux got a 2.9x average of 12/100? for L6E, but Feliks Zemdegs got similar times for 2-gen scrambles. So really, I think your original version has potential to be faster and it relies less on intuition, which is faster IMO.
 
crafto22 said:
This is a somewhat cool concept, but honestly I preferred your original version. CP-Roux honestly isn't very useful, if at all. The main advantage of the original Briggs Method was the awesome 2-gen F2L/LL. Without anything like that, this method just become a Roux variant that solves CMLL while inserting the last piece of the SB with a more complex FB. Don't know if that's a good trade off. Personally, I don't think L6E has the potential to be any faster than maybe 2-3 seconds. I know GuRoux got a 2.9x average of 12/100? for L6E, but Feliks Zemdegs got similar times for 2-gen scrambles. So really, I think your original version has potential to be faster and it relies less on intuition, which is faster IMO.
Click to expand...
One of the reasons I think this is better is because it is much more efficient than the previous method; especially the first step. It also relies on intuition as much as roux does and no more than the previous method did. In addition it also has fewer algs than my previous method did and the recognition for that is easier. Lastly, the LSE will likely be faster than GuRoux's 2.9 (assuming the same skill level) as it requires 3-5 less turns on average. Also, you maintain the 2-gen from the original method for the most part when doing SB. You could do the original method+ BLS I guess if you want as that will help recognition.
 
shadowslice e said:
One of the reasons I think this is better is because it is much more efficient than the previous method; especially the first step. It also relies on intuition as much as roux does and no more than the previous method did. In addition it also has fewer algs than my previous method did and the recognition for that is easier. Lastly, the LSE will likely be faster than GuRoux's 2.9 (assuming the same skill level) as it requires 3-5 less turns on average. Also, you maintain the 2-gen from the original method for the most part when doing SB. You could do the original method+ BLS I guess if you want as that will help recognition.
Click to expand...

Nah, BLSing the original method wouldn't be good because it would still provide a 1LLL, which is already supplied through 2GLL.
 
This method looks pretty cool. I've pretty much figured out how you do the 3 cycle of the corners to fix cp but I ran into a problem. There are some patterns that seem to match the reference picture on jaap's site but aren't in 2gen. Could you explain how you can tell a diag swap from a two gen state. If you just did a y perm the shape be X I which is a valid state? I assume you have some way of detecting which way the pieces are swapped and not just based on the pair exciting over those two locations.
 
gyroninja said:
This method looks pretty cool. I've pretty much figured out how you do the 3 cycle of the corners to fix cp but I ran into a problem. There are some patterns that seem to match the reference picture on jaap's site but aren't in 2gen. Could you explain how you can tell a diag swap from a two gen state. If you just did a y perm the shape be X I which is a valid state? I assume you have some way of detecting which way the pieces are swapped and not just based on the pair exciting over those two locations.
Click to expand...

Yeah, I had this problem initially. Essentially it comes down to the direction which the base pair (the ones which would be in DRF and DRB when they are solved are). Their orientation in the swap they represents dictates the orientation of the rest of the pieces.

Needless to say, it can be very difficult to do this with a 3-cycle sometimes (which is one of the advantages of the F/F' and D/D' style). It may just be easier in some situations to just do a standard 2-swap on the base pair than to work out the new cycle.


Also, I will post a full summary of the D/D' style (or the D insertion method :p) later when I have the time (but I have class soon so I don't have time to type it all out).

For now, however, it is essentially a pseudo pair PCMS/Roux thing. While doing it you insert pseudo pairs which are off by a D or D' apart from the pair in DBR which is the corner you will take a not of and track it's opposite corner and see where it is after the pairs have been inserted as well as looking ahead for the pair that would be in DFR when solved. You then do a two swap to make sure that the set of patterns is met (this is normally~4-5 moves long). It is similar to CPLS except that you don't have to insert the pair and it is also much easier to influence the swap you get and hence the location of the DFR corner so you can get an easy pair.
 
Last edited:
shadowslice e said:
One of the reasons I think this is better is because it is much more efficient than the previous method; especially the first step. It also relies on intuition as much as roux does and no more than the previous method did. In addition it also has fewer algs than my previous method did and the recognition for that is easier. Lastly, the LSE will likely be faster than GuRoux's 2.9 (assuming the same skill level) as it requires 3-5 less turns on average. Also, you maintain the 2-gen from the original method for the most part when doing SB. You could do the original method+ BLS I guess if you want as that will help recognition.
Click to expand...

guroux's LSE is like 2.3 which is average for someone our speed. Mine is 2.5 cuz im bad and blox are life. Alex on the other hand can probably sub 2 consistently.
 
You must log in or register to reply here.

Similar threads

R
I aim to average 15 seconds in 3x3x3.
Replies
8
Views
325
risekake
R
Filipe Teixeira
OO way to control a 3x3x3 rubiks cube using a gamepad?
Replies
8
Views
330
Filipe Teixeira
Filipe Teixeira
Lord_Oscar
[NR] Dwyane Ramos 4.54 3x3x3 single
Replies
4
Views
246
Puzzle Maniac
Puzzle Maniac
F
[AsR] [AsR][WR2] | Ruihang Xu (许瑞航) | 3x3x3 | 3.24 single |
Replies
5
Views
444
abunickabhi
abunickabhi
S
3x3x3
Replies
4
Views
338
Salvation3in1
S
Back
Top