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.

I've been on and off trying to get better at DR. Some days I try to ONLY do DR and I either feel like I get lucky or I waste time I could have spent doing block building. I'm curious if the FMCers here feel that DR really is the best method or if it just has a big following and top solvers would be getting those numbers either way. There is no question that the past few years have shown a huge growth in FMC.

I highly, highly doubt top-level solvers would be averaging the same that they do without and kind of domino reduction as they do when they're doing only domino reduction in 100% of their solves.

There has been so many advancements related to domino reduction, like better knowledge of getting into HTR, more edge cycle algorithms for insertions, and better filtering of good DR's (immediately finding something else if the DR is too long or there are bad corners), just to name a few.

The best people in the Fewest Moves Facebook group literally globally average 22-23 moves. I actually don't quite understand it myself really, because my global average is around 26-27 moves, which is pretty far off of that, but I guess the top-level solvers just have had more deliberate practice with FMC (I know I could get down to a 25.0 global average if I really tried, but I'm not doing FMC much these days as I'm focusing on other things).

If you really want to test it, grab what you think are the best FMC solvers, and make them do a non-DR ao12 and a full DR ao12 and see which is better. Although this is a bit flawed, since the best FMCers don't practice non-DR techniques as much, likely because they think of pure blockbuilding as just a waste of their precious one hour of time. Doing nothing but farming and trying tons and tons of DR's is also more consistent since you're almost guaranteed to get a lucky one, but if you only try a few DR's and also look for blockbuilding solutions as well, you'll have less certainty you'll get a solution of, for example, 25 moves or less within the hour (if you're top-level).

If you think you're bad at DR, just keep doing it. If you want to also do blockbuilding solutions, that's fine too, and having a more open mindset for different kinds of solutions is probably healthier. But until you can get a sub-30 ao12 or so doing only domino solutions comfortably, then it's best to figure out why you think you're wasting time doing DR instead of blockbuilding solutions. Are you trying to use 14 move DR's? Are you trying to use DR's that have 10 move optimal corners? If you are getting really short DR's with good corners plenty often, then maybe change your approach once you're in DR. Perhaps immediately try doing HTR on normal and inverse scramble right when you've got to DR or something. I'm not sure.

Top-level FMCers use a varying degree of domino reduction in their solves. Some of them literally only try domino reduction and that's what 100% of their solutions are, some of them will pursue a blockbuilding-based solution if there's an easy couple squares, couple pairs, and EO in like six moves or something, some are more experienced and practiced with blockbuilding-based solutions so they spend more deliberate seeking them out, and some of them try crazy more rare FMC techniques and get really good solves with them, like building HTR blocks, doing edges first, partial DR, corner orientation, or blockino.

In summary, yes, DR-based solutions have seen a huge growth in FMC the past few years, and no, top-level FMCers would not average anywhere near what they do now without the developments of DR.

I've been on and off trying to get better at DR. Some days I try to ONLY do DR and I either feel like I get lucky or I waste time I could have spent doing block building. I'm curious if the FMCers here feel that DR really is the best method or if it just has a big following and top solvers would be getting those numbers either way. There is no question that the past few years have shown a huge growth in FMC.

tl;dr DR is a good method, maybe the best, but maybe it is not for everyone.

It is undeniable that many people are getting extremely good using mostly or only DR. But this is not the case for me. I don't think I have a problem with DR itself, I grasped the method quite well, but it's the mindset of "spamming dozens of EOs and DRs until you get lucky" that does not work for me.

Here is my story with DR:
I have not practiced DR seriously until mid- or late-2019. Previously I would only use DR occasionally if I saw a good start. In my official 18 I literally got a 6 moves DR by chance while doing EO + blocks.
Around September 2019 I started doing DR-only attempts. It was rough at the beginning, and I could rarely finish a solve within one hour, but that was expected. But soon it payed out and in 3 months or so I was back at the same level that I was before DR (25-26). At that point I was very motivated and I kept going.
But at the beginning of 2020 things changed. Or rather, they did not: after another 4 months I was stuck at the same level, just a little less consistent than I was before DR; I got more very good solves (22-23) but also more bad solves (28+). That was very demoralizing, and most importantly I did not (and do not) enjoy doing FMC with DR only as much as I enjoyed doing it the "old way".
So In the last year I have slowed down with my FMC practice - not entirely out of frustration, but because I do enjoy other events too . Lately I have been trying to get back to old style FMC with some DR (after all, "old style" is about trying everything you know), but without much success (in the last 8 attempts or so I got something like 7 DNFs).

I've been on and off trying to get better at DR. Some days I try to ONLY do DR and I either feel like I get lucky or I waste time I could have spent doing block building. I'm curious if the FMCers here feel that DR really is the best method or if it just has a big following and top solvers would be getting those numbers either way. There is no question that the past few years have shown a huge growth in FMC.

100% DR is good enough for low 21 global averages (possibly sub 21 too). If your not having success with DR there are probably some areas that you are not good at but could easily improve. Some tips: (1) Do DR 100% of time. (2) Do HTR 100% of time (or close to). (3) Make sure you can find all sub 5 EOs on a scramble (With the possible exception of some of the more complicated NISS EOs). (4) Check heaps of EOs. (5) Switch after DR-4C4E, DR-4C2E for 5 or less moves. Switch after DR-3C2E for 6 or less moves. (6) Make sure you are comfortable with solving these DR-4C4E cases:

R, B2 L, F2 D2 R, B2 U L, L2 D R, F2 D R, R2 U R, R2 F2 R, R2 U2 R, F2 B2 L, L2 U' D R, B2 U2 D2 R, B2 U' D2 R, F2 U2 D' L, R2 F2 U2 R, F2 U D R, L2 B2 D R, B2 R2 U L, R2 F2 U R, F2 L2 D R, L2 F2 U L, B2 L2 D L, R2 B2 D L, F2 R2 U R, R2 U2 B2 L, F2 R2 D2 R, R2 D' F2 R, F2 U' F2 R, L2 D' F2 L, B2 U' F2 L, L2 U' B2 R, B2 D' F2 L, L2 U B2 R, B2 D B2 R, R2 L2 D' R, F2 U2 F2 R, R' D2 F2 R, R' U2 F2 R.

(7) Learn the 7 move or less corner cases here: https://drive.google.com/file/d/1lYQnaF1VuiR7EDfi1DUcIJM0FYC2B3mo/view?usp=sharing.
(8) On the majority of scrambles I only try finding finishes for DRs where DR length + quarter turns in corner solution < 14. In your case you might want to replace 14 with 15 or 16. (9) HTR triggers: R, R U2 R, R U2 L, R U2 F2 R, R U2 D2 L, R U2 F2 U2 R. (8) Some HTR edge algorithms: R2 F2 R2 U2 R2 F2 R2 U2 (and its 15 other variants), R2 U2 F2 R2 F2 U2 R2 F2 (and its 15 other variants), R2 F2 R2 D2 F2 R2 F2 D2 (and its 15 other variants), R2 U2 R2 U2 R2 U2 (and its variants), L2 R2 U2 L2 R2 D2 (and its variants including the 8 moves double turn algorithms which solve this case).

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

D F2 // Pseudo 1x2x3 (2/2)
R B' R2 // Pseudo 2x2x3 (3/5)
L' // F2L-1 missing 1 corner (1/6)
F D U2 F D' F' U2 L D' L' D F2 // All but 3 corners (12/18)
U' R // Undo premoves (2/20)

Spoiler: Example 2

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

(L’ U2) // Square (2/2)
B2 F2 U’ // Two 2x2x1s + Edge (3/5)
(D R’ B R2 D2 R’) // 2x2x2 (6/11)
(L’) F2 // F2L-1 (2/13)
D2 L D’ L’ D L D L’ // L3C (6/19)

Skeleton: B2 F2 U' F2 D2 L D’ L’ D L D R D2 R2 B’ R D’ U2 L

Spoiler: Example 3

R' U' F U2 F2 D U L2 B2 F2 U' B2 D' F' U2 R' F2 D' B' U' F' L' D2 U R' U' F

U // pair (1/1)
(F' D2) // 2x2x1 (2/3)
(L' F2) // pseudo 2x2x2 (2/5)
(R2 B' R B) // pseudo 2x2x3 (4/9)
F' // correct pseudoness (1/10)
* R2 D2 // F2L-1 (2/12)
B R' B' // 3e3c (2/14)

* F B' D F' D' F' B R F R' // insert edges (7/21)

Skeleton: U B' D F' D' F' B R F R D2 B R' B2 R' B R2 F2 L D2 F

Spoiler: Example 4

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

(L2 D B2 L) // EO (4/4)
D2 L2 F2 D L2 B // DR (6/10) – The first L2 is inserted to give better continuation
D' U2 R2 U2 D2 B2 R2 // 3c (7/17)

Skeleton: D2 L2 F2 D L2 B D' U2 R2 U2 D2 B2 R2 L' B2 D' L2

Spoiler: Example 5

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

R B D' U2 L' // EO + pseudo 2x2x2 (5/5)
D B2 D' B D R2 // 2x2x3 (6/11)
D B D' B2 D' B2 // 2x2x3 + 2 squares (6/17)
D' B D' B2 U // 3c (5/22)

Spoiler: Example 6

Scramble: R' U' F U L2 U R2 U' R2 D' B2 U R D' B2 U' R U2 F R2 U' F' U2 R' U' F

L D F U' F' // 2x2x2 (5/5)
(B R F R F') // 2x2x3 (5/10)
(B2 R' D B' D') // pseudo F2L-1 (5/15)
R B U R' U' R // 3c (6/21)

Skeleton: L D F U' F' R B U R' U' R D B D' R B2 F R' F' R' B'

Spoiler: Example 7

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

(L B2 D' R') // EO (4/4)
D B U // 1x2x3 (3/7)
(L2 F U2 F2) // 2x2x3 + 1x2x2 + 1x1x3 (4/11)
(B) // fluke DR (1/12)
(U' L2 D' L2 D U) // 3c (6-2/16)

Skeleton: D B D' L2 D L2 U B' F2 U2 F' L2 R D B2 L'

Spoiler: Example 8

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

(R2 D' F2 R') // EO (4/4)
(D B2) // 1x2x3 (2/6)
(F2 D F' + U2 B U') // 2x2x3 (6/12)
(F' D B D B' F) // 3c (6/18)

Skeleton: F' B D' B' D' F U B' U2 F D' F2 B2 D' R F2 D R2

Spoiler: Bonus: 4 corners left

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

(L F) // square (2/2)
(U' L' U') // EO (3/5)
F R2 F L' F' // 2 more squares (5/10)
L' B2 L B2 // another square + pair (4/14)
L D2 F' R' // 4c (4/18)

Skeleton: F R2 F L' F' L' B2 L B2 L D2 F' R' U L U F' L'

Spoiler: Bonus: 5 corners left

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

(L R' F) // EO (3/3)
U2 L' // DR-3E1C (2/5)
(R2 F2 D2 L D' L') // DR (6/11)
(U' L2 U D B2 D') // 5C (6/17)

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

You can check whether you found the optimal insertion with Insertion Finder.

I got a 31
Scramble: R' U' F B2 D' F2 D L2 D2 U F2 U B2 R2 B L' B R2 F L2 R D2 B D R' U' F
(D2 L U R2 F’)//pseudo 2x2x2 (5/5)
F’//2x2x2 (1/6)
(L2 B’ U2)//2x2x3 (3/9)
F L’ B2 L F’//F2l-1 (5/14)
U B’ U’ L’ B L B’ U B2 U’ B’//AB4C (11/25)
Skeleton: L’ B2 L F’ U B’ * U’ L’ B L B’ U B2 U’ B’ U2 B L2 F R2 U' L' D2
Solution: L’ B2 L F’ U B’ L’ D’ L R U’ L’ U R’ U’ D B L B’ U B2 U’ B’ U2 B L2 F R2 U' L' D2
tips on how this could improve would be appreciated

Blockbuilding and NISS are plenty enough for sub 30. DR is probably necessary now to be consistently world class, but blockbuilding and NISS can definitely take you easily below 30 global average if you get good at them.

First successful mean in the weekly comp. Any critiques?

R' U' F U' F L D' L B2 U B R B R2 F' R2 D2 F2 B L U' R' U' F

Spoiler

D B F2 L' B' // 2x2x2
L2 D' F2 D2 F' //2x2x3
F' D' F D L' D F L' F' //F2L-1
D B' L B L' D2 B' L B L' //AB2C3C3E
//# (D F2) B L' B' L2 D' F2 D2 F2 D' F D L' D F L' F' D B' L B L' D' B' L2 * (F U' F) U F2 L' B L'
//#B' D F2 D' B D F2 D' (8-4)
//*D' F U' F' D F U F' (8-5)

//FINAL: B' D F2 D' B D B L' B' L2 D' F2 D2 F2 D' F D L' D F L' F' D B' L B L' D' B' L2 D' F U' F' D F2 U F2 L' B L'
//41 moves

R' U' F D L' B' L' U' L2 U' D2 L' D' L2 D' B2 D' L2 B2 U B2 L2 F' R' U' F

Spoiler

D R' L F2 L' F D F U F' D' //2x2x3
U' R B R' //EO
U2 B2 U L U' L' R' U' R U' R' U' R U //AB3C3E
//# D R' L F2 L' F D F U F' U' D' R B R' * U2 B2 U L U' R' L' U' R U' R' U' R U
//#B U B' D B U' B' D' (8-2)
//*R' F2 D2 B2 L' D2 F2 U2 (8-3)

//FINAL: B U B' D B U' B' R' L F2 L' F D F U F' U' D' R B R2 F2 D2 B2 L' D2 F2 B2 U L U' R' L' U' R U' R' U' R U

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

Spoiler

z2 // orientation
U L2 B U2 L' R2 F // 2x2x2
R B2 U R' U2 B // 2x2x3
R U F' U F //EO
U R2 U' R U2 R2 U2 R //DR
//D R2 B D2 R' L2 F L B2 D L' D2 B L D F' D F D L2 D' L % D2 L2 D2 L
//% D B' L2 F' B D2 F D' (8-1)
//# D R2 B D2 R' L2 F L B2 D L' D2 B L D F' D F D L2 D' L D * B' L2 B F' D2 F D L2 D2 L
//#R' U R D R' U' R D' (8-3)
//*R' B' L2 B R B' L2 B (8-6)
//Final: R' U R D R' U' R' B D2 R' L2 F L B2 D L' D2 B L D F' D F D L2 D' L D R' B' L2 B R F' D2 F D L2 D2 L
//40 moves

After your fourth move, notice how you can add any D move in between. Adding that move would orient any one of the other edges in the bottom layer. Maybe it wouldn't be that great, but you can switch that move to any other D move, to try a different solution.

that was a nine move f2l-1 and I found a seven move one. L D' L2 D F L' F'. Basically just pair up the pieces a different way, so it only takes 3 moves to get it in the right spot.

That was a really long 2x2x2, and what I would have done was notice that R' would solve the white-orange cross edge relative to the white-green edge. So all you have to do is solve white-orange-green corner to make a square, which can be done with F' D, and then when you insert white-orange, that makes a square. Unfortunately, the orange-green edge is in a bad spot to make a 2x2x2, but what you could is F2 U, which makes a pseudo block. So the whole thing would be F' D R' F2 U.

Also I noticed that you never switched to the inverse before f2l-1. Basically every time you try something on the normal scramble, unless it's very easy, always switch to the inverse to try stuff there.

I mean here are my solutions for the weekly comp, so ig u can use them to see what else you could have done. But can someone also point out any things I could have improved here(I know that there are quite a few, especially in the second solve)
Scrambles:
1. R' U' F U' F L D' L B2 U B R B R2 F' R2 D2 F2 B L U' R' U' F
2. R' U' F D L' B' L' U' L2 U' D2 L' D' L2 D' B2 D' L2 B2 U B2 L2 F' R' U' F
3. R' U' F D2 F2 L D' B R' F2 B2 D' U2 R2 L2 B2 L2 U2 B L2 B R2 B2 R' U' F

Spoiler: Solve 1(34)

(F2 B L2 B’) D' R//pseudo 2x2x2
(R)//2x2x2
(U F)//pseudo 2x2x3
F'//2x2x3
(L2 U' L' U)//f2l-1
U L2 U’ F’ L F L//AB2C2C1T
Skeleton: D’ R ^ F’ U L2 U’ F’ L F L # U' L U @ L2 F' U' R' B L2 B' F2
//#=L D L' U' L D' L' U
//@=F R F' L2 F R' F' L2
//^=R F' L F R' F' L' F
Final Solution: D' R2 F' L F R' F' L' U L2 U' F' L F L2 D L' U' L U D' F R F' L2 F R' F2 U' R' B L2 F2 B'

Spoiler: Solve 2(41)

L D F’ B’ D’//pseudo 2x2x2
(D')//2x2x2
(U' R U2)//pseudo 2x2x3
U'//2x2x3
R F R F L F’ R F L’//f2l-1
(R’ U R’ F R2 F’ U’ R U)//AB2C3C2E
Skeleton: L # D F’ @ B’ D’ U’ ^ R F R F L F’ R F L’ U' R' U F R2 F' R U' R U' R' U D
//#=L U' L' D L U L' D'
//@=D F' D' B' D F D' B
//^=D2 L2 B' L' B L' D2 R F' R'
Final Solution: L2 U' L' D L U L' F' D F' D' B' D F U' L2 B' L' B L' D2 R2 F L F' R F L' U' R' U F R2 F' R U' R U' R' U D

Spoiler: Solve 3(36)

F’ D R’ F2 U//pseudo 2x2x2
(U’)//2x2x2
B L D2 L' D2//pseudo 2x2x3
(D2)//2x2x3
L’ D L D’ B2//f2l-1
L’ B L2 B’ L’ B L B’ L2 U’ L’ U L’//AB5C
Skeleton: F’ D R’ F2 U B L D2 L’ D2 L’ D L D’ B2 L’ # B L2 B’ @ L’ B L B’ L2 U’ L’ U L’ D2 U
//#=L' F' L B L' F L B'
//@=D' L' B L B' D B L' B' L
Final Solution: F' D R' F2 U B L D2 L' D2 L' D L D' B2 L2 F' L B L' F L' B' D' L' B L B' D L2 U' L' U L' U D2

Hey, I just got a really lucky scramble (27 moves, even though I average 48.917 moves per scramble) in Speed FMC last week and was wondering if there was a way to do it better and if there was a way to get more sub-30 move solves. Currently, the steps I go through are similar to Petrus, except at the end:

2x2x2 block //pretty simple, just 3 edges and a corner around 3 centers
2x2x3 block //adding on another 2 edges and a corner around a fourth center
Adjacent 2x2x1 block //adding on yet another 2 edges and a third corner around a fifth center
Strategic F2L-4 insertion //solving a fourth corner and an eighth edge to make a 3x3x2 block similarly to CFOP, but inserting it strategically to
//better the edges on the OLL case
OLL //Orientation of the Last Layer
PLL //Permutation of the Last Layer
Please let me know any suggestions you might have on how I can improve.

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

Spoiler: My Solution

Solution: U2 L2 B’ U F’ L U’ F’ R’ U F’ U’ R’ F R’ F’ R U’ R’ U B’ D’ F D’ F’ D2 B

Breakdown:
U2 L2 B’ U F’ L U’ //2x2x3 Block (7/7)
F’ R’ U F’ U’ //Adjacent 2x2x1 Block (5/12)
R’ F R’ F’ R U’ R’ U //F2L (8/20)
B’ D’ F D’ F’ D2 B //LL (PLL skip!) (7/27)
27 moves (FMC PB!) + 3.39 minutes = 30.39

So if you see any improvements to that specific solution I could've made, that would be amazing. Additionally, If you know how I can do better on average (because I'm really bad at FMC overall and my best solution besides this is a 42), then please let me know. Thanks!

Hey, I just got a really lucky scramble (27 moves, even though I average 48.917 moves per scramble) in Speed FMC last week and was wondering if there was a way to do it better and if there was a way to get more sub-30 move solves. Currently, the steps I go through are similar to Petrus, except at the end:

2x2x2 block //pretty simple, just 3 edges and a corner around 3 centers
2x2x3 block //adding on another 2 edges and a corner around a fourth center
Adjacent 2x2x1 block //adding on yet another 2 edges and a third corner around a fifth center
Strategic F2L-4 insertion //solving a fourth corner and an eighth edge to make a 3x3x2 block similarly to CFOP, but inserting it strategically to
//better the edges on the OLL case
OLL //Orientation of the Last Layer
PLL //Permutation of the Last Layer
Please let me know any suggestions you might have on how I can improve.

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

Spoiler: My Solution

Solution: U2 L2 B’ U F’ L U’ F’ R’ U F’ U’ R’ F R’ F’ R U’ R’ U B’ D’ F D’ F’ D2 B

Breakdown:
U2 L2 B’ U F’ L U’ //2x2x3 Block (7/7)
F’ R’ U F’ U’ //Adjacent 2x2x1 Block (5/12)
R’ F R’ F’ R U’ R’ U //F2L (8/20)
B’ D’ F D’ F’ D2 B //LL (PLL skip!) (7/27)
27 moves (FMC PB!) + 3.39 minutes = 30.39

So if you see any improvements to that specific solution I could've made, that would be amazing. Additionally, If you know how I can do better on average (because I'm really bad at FMC overall and my best solution besides this is a 42), then please let me know. Thanks!

The two best things you can do to improve your solves are NISS and insertions. NISS is too complicated to explain in a post, look up a tutorial on YouTube. Insertions revolve around basically solving most of the pieces (usually by solving all edges after you reach a cubestate like in step three of what you laid out, leaving up to 5 corners), then solving those pieces with commutators "inserted" into the middle of the solve wherever they cancel the most moves. This also can be learned in full on YouTube

The two best things you can do to improve your solves are NISS and insertions. NISS is too complicated to explain in a post, look up a tutorial on YouTube. Insertions revolve around basically solving most of the pieces (usually by solving all edges after you reach a cubestate like in step three of what you laid out, leaving up to 5 corners), then solving those pieces with commutators "inserted" into the middle of the solve wherever they cancel the most moves. This also can be learned in full on YouTube