# Speed FMC Trial Competition

#### Mike Hughey

##### Super Moderator
Staff member
I would like to run a couple of trial speed FMC test competitions to see what scoring method makes the most sense. We've had two different proposals that seem to have some support. For this first one, I am going to use the logarithmic scoring method. However, at the end, I am going to list the winners by both the linear and the logarithmic scoring methods, just to compare. But the winner of this round will be declared based on the logarithmic scoring method.

So that we have time to run a couple before the end of the year, this will be a very short competition. I hope some people might participate since it is the holidays, but we'll see.

The scoring formula used for this round will be:
score = moves * ln (time in seconds) (where ln is natural logarithm)

And we will have a 10 minute time limit. Anything past 10 minutes is a DNF.

I also have wondered - if this is really intended to be a fairly fast event, it seems like it would be in the spirit of the Weekly Competition to make this an average of 5 instead of a mean of 3? For this round, I am going to provide 5 scrambles. If this proves to be an unpopular idea, I will scale back to 3 on the next trial competition. 3 would be easier for me to implement than 5, since it would then be the same as regular FMC in the form and in the timer. But I wouldn't like to start with mean of 3 this year and then have to switch to average of 5 next year, so I'd like to play with this now to see what happens. If no one participates with all 5 scrambles, I'll assume it was a bad idea and go with mean of 3.

The rules:
Same as normal FMC - you can use up to 3 solved cubes, stickers, and pen and paper. Maximum time of 10 minutes. For each attempt, time the solve from the time you reveal the scramble. You must have the full solution written down on paper before stopping the timer. Then submit here for each solve your solution as well as the time taken to perform the solve.

Scrambles for round 1:
1. R' U' F R2 F' L2 B R2 D2 F' L2 R2 B2 L' R2 F' U' B' L2 D2 U B2 U' F' R' U' F
2. R' U' F D2 R2 D2 F2 D L2 B2 F2 R' U L F D B2 F L D R' B2 F2 R' U' F
3. R' U' F R2 D' L2 D2 B2 U' R2 D' B2 U2 R F R U' L2 U L B U' F2 R' U' F
4. R' U' F L2 R2 D L2 D U2 F2 D2 U' B' F2 D R' D2 L D' B2 D' F R D R' U' F
5. R' U' F D2 U2 L2 U2 L R2 D2 F2 R B F2 U B2 F' L D2 U' F R F' R' U' F

I apologize for the shortness of these rounds, but I would like to do 2 rounds in time for next year's first competition. Therefore, this competition will close just before midnight GMT on December 26, 2019. So that's only 3 days. So hurry if you'd like to compete! #### Mike Hughey

##### Super Moderator
Staff member
Please also feel free to discuss formats/rules in this thread. I will sort through and collect results when the first round is complete.

One thing worth discussing is this: I would like to go with a linear scoring method for the second round. What should k be?
score = moves + k * (time in seconds)

I played a little with k = 1/60, but to me it seems like it would almost always be worth it at that value of k to use all 10 minutes, and that's a little silly.

#### Mike Hughey

##### Super Moderator
Staff member

Solve 1: 9:33.30, 50 moves (score: 317.57)
scramble: R' U' F R2 F' L2 B R2 D2 F' L2 R2 B2 L' R2 F' U' B' L2 D2 U B2 U' F' R' U' F
solution: F D L2 F L F' U D' R2 D F R' D' R D R' B' R' B2 R D2 R D2 R' D2 R D2 R' F' R F2 L' F2 R' F2 L F' R' L' D2 L D L' D2 R D' L D' B R

2x2x2: F D L2 F L F' U F
2x2x3: F' D' R2 D F R' with premove R
3x cross: D' R D R' B' R' B2 R with premove B
4th pair: D2 R D2 R' D2 R D2 R'
OLL: F' R F2 L' F2 R' F2 L F'
PLL: R' L' D2 L D L' D2 R D' L D'

Solve 2: 9:54.50 (ouch - scary!), 54 moves (score: 344.94)
scramble: R' U' F D2 R2 D2 F2 D L2 B2 F2 R' U L F D B2 F L D R' B2 F2 R' U' F
solution: R' F2 R F' L2 D L' D' L2 D' B2 D B L2 F' D2 F L' D2 L' B D' B' D U2 D R' F' R F U F' U D' B L' F L2 B' L F' U' D' L2 U D L2 D' B D2 F' L' F L

2x2x2: R' F2 R F' with premoves F' L' F L
2x2x3: L2 D L' D' L2 D' B2 D B
3x cross: L2 F' D2 F L' D2
4th pair: L' B D' B' D with premoves D' B D2
OLL: U2 D R' F' R F U F' U D'
PLL: B L' F L2 B' L F' U' D' L2 U D L2

Solve 3: 8:23.40 (not quite as scary), 50 moves (score: 311.07)
scramble: R' U' F R2 D' L2 D2 B2 U' R2 D' B2 U2 R F R U' L2 U L B U' F2 R' U' F
solution: D2 U B U F U2 L U' L2 U L' U' L B2 L' D' B2 D L' B L B U B U' D L B L' B' L B L' B' D' R B2 R' B2 R D' R' B' R B R D R2 B' U2

2x2x2: D2 U B U F
pseudo 2x2x3: U2 L U' L2 U L' U'
3x cross: L B2 L' D' B2 D with premove U2
4th pair: L' B L B U B U'
OLL: D L B L' B' L B L' B' D'
PLL: R B2 R' B2 R D' R' B' R B R D R2 B'

Solve 4: 9:47.77 (this is just crazy - I have to stop doing that!), 39 moves! (score: 248.68)
scramble: R' U' F L2 R2 D L2 D U2 F2 D2 U' B' F2 D R' D2 L D' B2 D' F R D R' U' F
solution: U2 F' D R2 L D L' F' D F D2 R F2 R' F2 R F R' D R F R' F' D' R' L' F2 L F L' F2 R F' L F R' L2 F2 L'

2x2x2: U2 with premoves L2 F2 L'
2x2x3: F' D R2 with premove R'
3x cross: L D L' F' D F D2
4th pair: R F2 R' F2 R F R'
OLL: D R F R' F' D'
PLL: R' L' F2 L F L' F2 R F' L F

Solve 5: 5:34.68 (was tempted to go for corner insertion, then decided not to), 37 moves! (score: 215.09)
scramble: R' U' F D2 U2 L2 U2 L R2 D2 F2 R B F2 U B2 F' L D2 U' F R F' R' U' F
solution: B L B' U B' U2 F' D2 F2 D' F2 R D R D2 R2 D2 R F' R' F B R D R' D' B' R' B R' F2 R B' R' F2 R2 F2

2x2x2: B L B' U B' U2
2x2x3: F' D2 F2 D' F2
3x cross: R D R D2 R2
4th pair: D2 R F' R' F with premove F2
AB3C: B R D R' D' B'
PLL: R' B R' F2 R B' R' F2 R2
Very lucky! Would probably have been a decent regular FMC solve for me with insertion; I would have been tempted to quit after finding this!

Ao5: 292.44

Last edited:
• fun at the joy

#### fun at the joy

##### Member
I went over the time limit.
Scramble: R' U' F D2 R2 D2 F2 D L2 B2 F2 R' U L F D B2 F L D R' B2 F2 R' U' F

Solution:
z2 U' R' F D L' D U R' // Cross
L' U L2 U' L' // 1st Pair
F' U' F U' L' U L // 2nd Pair
U' R U' R' U R' U' R // 3rd Pair
U2 R U R' U' R U2 R' // 4th Pair
y L' U' L U' L' U L U L F' L' F // OLL
U' Lw' U R' D2 R U' R' D2 R2 // PLL

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

Solution:
x2 R' F2 L D // Cross
L' U' L y R' U R2 U R' // 1st Pair
U R' F R F' // 2nd Pair
y' R U' R' // 3rd Pair
U2 L' U L U' L F' L' F // 4th Pair
y R U2 R2 U' R2 U' R2 U2 R // OLL
y F2 U' L R' F2 L' R U' F2 U' // PLL

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

Solution:
y B' D R D R2 U' F D // Cross
y' U R' U' R // 1st Pair
y' z' D' L' U2 L D // 2nd Pair
y' U2 R U' R' U R U R' // 3rd Pair
R U' R' U2 F' U F U F' U F // 4th Pair
y2 F R' F' R U R U' R' // OLL
y L U R U2 L' U L U2 L' R' // PLL

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

Solution:
y x2 F R' U' L2 F' D' // Cross
R U' R' L U' L' // 1st Pair
U2 R' U R y U' L U' L2 U L // 2nd Pair
F U2 F' // 3rd Pair
U2 R U R2 F R F' // 4th Pair
y Lw' U' L U R U' Rw' F // OLL
R2 F2 R U2 R U2 R' F R U R' U' R' F R2 U2 // PLL

Final Solution: L B' D' F2 L' U' B D' B' F D' F' D2 B' D B D' L D' L2 D L B D2 B' D2 R D R2 B R B' F' R' B R F R' B' R F2 R2 F D2 F D2 F' R F D F' D' F' R F2 D2
awful solutions and lol time consistency

Last edited:

#### WoowyBaby

##### Member
Yes!! I'll give this a shot! I'm so excited to have Speed FMC in Weekly Comps! Also, I support logarithmic scoring and average of 5.

#1 - 43 in 9:46 (274pt)
U D' R'
B' U D R2 B'
D2 L2 D L2 D L2
B D' F' D B' D' F L2
D R2 B2
R2 D' R' D R' B2 L U' L'
R2 D'
#2 - 33 in 9:13 (208pt)
R L F
U2 L
D' R2 B2 U' L2 U
R D2 R U' R' D2 R2 B2 F D2 F' U2 F D2 F
D' B2 L2 U2 L2 U2 L2
#3 - EDIT: DNF solution didn't work lmao

#4 - 57 in 3:28 (304pt)
D2 B' R D L F' L'
U' R U2 R2 U' R
F U2 F' L U L'
R U R' U F' U' F
U2 F' L F L2 U2 L
U' F U F' U F U' F'
L F' L' B L F L'
B U R L' B2 R' L U B2
#5 - 27 in 9:27 (171pt)
B'
L' U2 F2 L2 F'
U R U D' R
U D2 L'
B2 L B L' F2 L B' L'
U2 D F2 R2 D

(Log scoring ofc,)
Final Average of 5- 262.0

This is really fun and stressful! You have to be so very quick and make decisions to not explore things because it takes time, you just gotta go! Solves 1 and 3 I was panicking a ton at the end due to time and I was going to DNF if I tried to find the best insertion. I'm really happy to have this event in the Weekly Comp!

Last edited:
• Mike Hughey

#### Kit Clement

Whew. Found a fat sune skip really early, barely found it again in time to write it all on normal.

F L2 U R2 D' F //222+pair
(D2 L D' L') // F2L-2
(D B D' B' D' B' L B L') //F2L-1
(D2 R F' R' F D' F D' F') //Fat sune
(B R F' R F R2 B' D2) //finish

Final: F L2 U R2 D' F D2 B R2 F' R' F R' B' F D F' D F' R F R' D2 L B' L' B D B D B' D' L D L' D2

Pretty garbage, ended up just doing a setup to Tperm on the last.

F D B' //221
L' U L' D L' D2 F //223
R' B R2 B' //+square
F' R2 F D R' D' R' D2 B' D' B //LL
D2 L2 D' z R2 Uw R2 Uw' R2 y' R2 Uw' R2 Uw R2 U' F U2 //finish

This is where I realize that doing solves on inverse really puts a time crunch on you.

U2 F' U B D R U //222
F' L2 F' D F L' D //223
F2 L F2 L F' //3C+2T
F L F' R' F L' F' R //5C
L U' R' U L' U' R U' //3C
U' L' U' R U L U' R' //finish

Final: R U L' U' R' U L U2 R' U L U' R U L' R' F L F' R F L2 F2 L' F2 D' L F' D' F L2 F U' R' D' B' U' F U2

Tried to do some quick un-stickered insertions, but ran out of time. Note that there is a move in here that I did cancel in my final written solution.

L2 R2 F L' F' U2 //222
D2 L' D L F2 //223
F L F' //F2L-1-edge
R D' L2 D R' //F2L-1
B' L2 B L B' L' B L' D L' D' L //3C
D2 R D' L2 D R' D' L2 D' //finish

Finally tried to go for an earlier finish to see if that actually helps you in scoring significantly. Had time to do 3c, but was curious to calculate stuff after to see if attempting to save moves was worth it. And same as last time, I had done the cancel appropriately in my final solution.

U2 D' B2 D //221
R U R //222
L F' D2 F L2 D' //F2L-1-edge
F U B U' F' //F2L-1
B' L B L' //5c
L' D' R D L D' R' D //3c
B' R F' R' B R F R' //finish

Final ao5: 242.75

This was really fun! It's nice to see that the natural log function does seem to give the right tradeoff to mostly discourage inserting - I'd have to do 3 moves better within 3 minutes on solve 5 to get a better score. Although I do agree with the critiques from the old thread that this scoring method is rather opaque, and I don't really know that tradeoff in the moment very well.

I'm actually wondering if a better way to address time would be to do an additive method like proposed in the earlier thread, but weight each of the seconds differently. That is, your final score would be:

Move Count - Time Remaining Score, where

Time Remaining Score = (1/120)*(unused seconds in the last minute) + (1/100)*(unused seconds in the second to last minute) + (1/90)*(unused seconds in the third to last minute) ... etc.

I'm not convinced about what the right strategy is here, nor have I tested what the appropriate coefficients should be, but my thought is that weighting the earlier seconds more might promote people to stop when they have a decent solution since they will be giving away their most valuable seconds. But on the other hand, it could encourage people to keep going since they have already passed on the very most valuable seconds, the initial ones. Maybe this stream of consciousness should just be more evidence we should do this purely linearly, since it's easiest to understand in this manner and is the middle ground to this balance.

• WoowyBaby and Mike Hughey

#### Mike Hughey

##### Super Moderator
Staff member
My gut feeling after trying this myself and seeing the other attempts so far: Average of 5 is really the right format for this. Any disagreement?

#### Cubinwitdapizza

##### Member
Can you explain how the scoring works a little more? I’m kinda confused...

#### WoowyBaby

##### Member
My gut feeling after trying this myself and seeing the other attempts so far: Average of 5 is really the right format for this. Any disagreement?
I sure don’t disagree, ao5 is the way to go! The reason FMC has mo3 is because it’s very time consuming, but this you could do all 5 in under an hour. Also, I know DNF’s will be VERY common in this event, so having a buffer is nice, it does sort of ruin your mean but it’s better than everyone getting DNF means all the time. Just my thoughts.
Can you explain how the scoring works a little more? I’m kinda confused...
I know it can be a bit confusing eek
Currently, and hopefully forever, score is calculated by ([number of moves] multiplied by the natural log of [seconds used]). Lowest score wins.
Other methods include ([seconds used] times (some constant k) plus [number of moves]),or perhaps ([number of moves] times [used minutes]), or even ([number of moves] plus the log of [seconds leftover]).

Last edited:

#### Cubinwitdapizza

##### Member
I sure don’t disagree, ao5 is the way to go! The reason FMC has mo3 is because it’s very time consuming, but this you could do all 5 in under an hour. Also, I know DNF’s will be VERY common in this event, so having a buffer is nice, it does sort of ruin your mean but it’s better than everyone getting DNF means all the time. Just my thoughts.

I know it can be a bit confusing eek
Currently, and hopefully forever, score is calculated by [number of moves] multiplied by natural log of [seconds used]. Lowest score wins.
Other methods include [seconds used] times (some constant k) plus [number of moves], or perhaps [number of moves] times [used minutes], or even [number of moves] plus log[seconds leftover].
Still confused lol. So multiply the number of moves by the amount of seconds? Cuz I tested that on one of mikes calculations and it didn’t appear to be the same as what he put.

#### WoowyBaby

##### Member
Still confused lol. So multiply the number of moves by the amount of seconds? Cuz I tested that on one of mikes calculations and it didn’t appear to be the same as what he put.
Not quite, okay, so you take the number of seconds used first, let’s use Mike’s first solve as an example. He used 9:33.30.
9:33.30 = 573.30 seconds
Now, get the natural log of 573.3, which you can do by typing “natural log of 573.3” into google.
In(573.30) = 6.3514...
Now, take the number of moves in your solution and multiply it by this number. Mike got 50 moves for his first solve.
50 * 6.3514... = 317.57 points

Hope this helps!

• GenTheThief and Mike Hughey

#### qT Tp

##### Member
F R D' B2 R2 F' R' //223-1EDGE
U' F2 L' F' L U //F2L-1 IN 13
D' L' D L D' B D' B' //F2L
F D L D' (L' F') // 2 CORNERS TWIST
(F L) F' L' F L F' R F L' F' L F L' F' R'

Really good start, 13 move f2l-1. Messing around with different OLL's for a minute or 2 till I found this 19 move LL
Thought I had a 49 in 9:38.66. turns out I made a mistake in scrambling
B' L' D R' U B2 //222
R' F' R F' R' D U' R' U R' //F2L-1 -1EDGE
F' R' L D' L' //EO
F D2 F D' F D'L' (F' L) //F2L
(L' F) F R F R' F (L) //OLL
(L') D' L D L B' L2 D L D L' D' L B//T-PERM
U2 L2 B' U L' U2 R' F' R F //2X2X3
L B L' //EO
B' D2 B //F2L-1 IN 16
D' B D' B' D B D' B' //F2L IN 24
Z2 L F R' F' L' F R F' //OLL
R2 D' R' L F2 R L' U2 D R2 U2 //U-PERM

Found some nice pairs starting with U2. build them out to a 2x2x3
fairly standard petrus style solve.
B2 R B//221
L' U' F2 U2 R U' L U' L2 U2 //223
L2 B U B U' B' //F2L-1 (L')
L' B' L' B2 L //F2L + OLL-SKIP
D' B2 D B2 D' R D B D' B' D' R' D2 B'//R-PERM

AO5: 269.07

I really enjoyed Speed FMC. The 10 minute time limit combined with the AO5 format makes this a really nice event to do when you don't have 3 hours to dedicate to regular FMC.
My strategy for Speed FMC was to always go for a block building/petrus style. DR is not viable in this amount of time for me and also tends to finish with edge/corner cycles which are difficult to find insertions for within the time limit.

#### ZZ'er

##### Member
Not quite, okay, so you take the number of seconds used first, let’s use Mike’s first solve as an example. He used 9:33.30.
9:33.30 = 573.30 seconds
Now, get the natural log of 573.3, which you can do by typing “natural log of 573.3” into google.
In(573.30) = 6.3514...
Now, take the number of moves in your solution and multiply it by this number. Mike got 50 moves for his first solve.
50 * 6.3514... = 317.57 points

Hope this helps!
So the longer you take the better?

#### TheLegend12

##### Member
So the longer you take the better?
I think the goal is to get the lowest score.

#### ZZ'er

##### Member
Ok that makes more sense, thanks.

#### irontwig

##### Member
1: DNF
Spent too much time trying to get
F L2 U R2 D' F2 D B D B' F' //Pseudo F2L-1 corner
to work.

2: 37 in 7:24.87 (219.52 points)
R D' R' U B2 L F' B D B' F' R2 B2 R' L' B L' B' U' F U2 F' U' F U2 F' R U2 R D L' B2 L D' R2 U (36)

Spent too much time trying to get a nice pseudo start with
R D' R' U B2 //Two squares
At 6 minutes I panicked and quickly did
L F' B D B' F' R2 B2 R' //2x2x3+pairs
L' B L' B' //F2L-1
U' F U2 F' U' F U2 F' //F2L
R U2 R D L' B2 L D' R2 U //Lucky ZBLL

3. DNF, looks like I miscrambled :/

4. DNF, real waste of the nice start
U2 L2 B' U L' U2 R' F' R F D2 //2x2x3+Square

5. 38 in 7:58.98 (234.52 points)
D' B2 D R U R L B' D2 B2 D' R' U2 F U F2 L F L2 U2 L2 U' L' U L' U' D' B D' F2 D B' D' F2 D2 L' R B2 (38)

D' B2 D R U R //2x2x2
L B' D2 B2 D' //2x2x3
(B2) //F2L with one twisted corner and one flipped edge
R' U2 F U F2 L F L2 U // F2L+EO
U L2 U' L' U L' U' //OLL
D' B D' F2 D B' D' F2 D2 L' //PLL

This was stressful, though I would probably done a bit better if had more than one cube at hand (not at home at the moment).

#### TipsterTrickster

##### Member
U D' R' B' D U2 R2 D' B2 D2 B L2 D B D B2 L2 B D F' D F B D' B' D2 F B D' B' D F' D2
L R F R D' R U2 R2 L2 U' L' U L' B2 D2 U R' U2 R U L' U' L U L' U2 L D2 L2 Uw' L U' L U L' Uw L2 F U' F'
ran out of time
D2 L2 B' U2 R2 B' L B' U R' B2 D R' D R D' B D' B' L' D2 L B D R D' R' D R D' R2 B' R D R D' R' D2
F2 L' F L B' L' F' L B D' L B L' D' L' D L B' D2 F D' F' D R' D R D2 F' L2 F L' F' B' U' R2 D' R'
Ao5: 230.38

Last edited:

#### Pyjam

I admit, I haven't tried, but I have doubts about the formula.
If someone solves the cube with his left hand while writing down the solution with his right hand, without barely any attempt to optimize, he may very well solve in 50 moves and 50 seconds, for a score of 195.6. Better than almost every score posted until now.
But I might be wrong, I haven't tried.

Also, it's disappointing if the HTM metric is used because it disqualifies the Roux method which is one of the best method (if not the best) for speed FMC. But the world is ruled by CFOP users! (Just kidding)

• GenTheThief

#### ImmolatedMarmoset

##### Member
I admit, I haven't tried, but I have doubts about the formula.
If someone solves the cube with his left hand while writing down the solution with his right hand, without barely any attempt to optimize, he may very well solve in 50 moves and 50 seconds, for a score of 195.6. Better than almost every score posted until now.
But I might be wrong, I haven't tried.

Also, it's disappointing if the HTM metric is used because it disqualifies the Roux method which is one of the best method (if not the best) for speed FMC.
I use Roux, and I disagree with the second point. That's exactly how regular FMC works in competition, and so why shouldn't this work that way as well? HTM just seems cleaner from an FMC point of view.