• 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!

Pooter

Member
Joined
Nov 8, 2021
Messages
2
Location
Thailand
I'm a new cuber and I've just started learning PLL and need some help with it. Whenever I do a Z perm half the time it goes to a H perm does anyone know why. Sorry if it's a simple fix.
 

Meisme

Member
Joined
Nov 22, 2020
Messages
13
I'm a new cuber and I've just started learning PLL and need some help with it. Whenever I do a Z perm half the time it goes to a H perm does anyone know why. Sorry if it's a simple fix.
You have to make sure it is in the right direction. Depending on the alg you use, you want to have the swapped edges in either the front left or front right.
 

tsmosher

Member
Joined
Aug 30, 2020
Messages
649
This time, I used the movecount coefficient calculator to do OLLs instead. (Note: these are not the OO OLLs)

OLL 1: R U' R2 D' r U' r' D R2 U R'
OLL 2: f U R U' R' S' U R U' R' F'
OLL 3: r' R2 U R' U r U2 r' U M'
OLL 4: R' F R F' U' S R' U' R U R S'
OLL 5: l' U2 L U L' U l
OLL 6: l U2 L' U' L U' l'
OLL 7: r U R' U R U2 r'
OLL 8: l' U' L U' L' U2 l
OLL 9: R U2 R' U' S' R U' R' S
OLL 10: f R f' r' U r U' r' U' r
OLL 11: r' R2 U R' U R U2 R' U M'
OLL 12: R r2 F' r U' r' F2 r U' M'
OLL 13: F U R U' R2 F' R U R U' R'
OLL 14: R' F R U R' F' R y' R U' R'
OLL 15: l' U' l L' U' L U l' U l
OLL 16: r U r' R U R' U' r U' r'
OLL 17: F R' F' R U S' R U' R' S
OLL 18: F S' R U' R' S R U2' R' U' F'
OLL 19: R' U2 F R U R' U' F2 U2 F R
OLL 20: S R' U' R U R U R U' R' S'
OLL 21: R U R' U R U' R' U R U2 R'
OLL 22: F U R U' R' S U R U' R' f'
OLL 23: R2 D R' U2 R D' R' U2 R'
OLL 24: F R B' R' F' R B R'
OLL 25: R2 D R' U R D' R' U' R'
OLL 26: r' F' r U' r' F2 r
OLL 27: R U R' U R U2 R'
OLL 28: R' F R S R' F' R S'
OLL 29: R F R' U R F' R2' F R U' R' F' R
OLL 30: S' R' U' R f R' U R U' F'
OLL 31: R' U' F U R U' R' F' R
OLL 32: S' r' F' r U f U F'
OLL 33: R' U' R U R f' U' f
OLL 34: F U F' U' R' F' r U M
OLL 35: R U2 R2 F R F' R U2 R'
OLL 36: r' F' r U' r' F r U r U' r' F
OLL 37: F R' F' R U R U' R'
OLL 38: F R U' R' S U' R U R' f
OLL 39: R B' R' U' R U B U' R'
OLL 40: R' F R U R' U' F' U R
OLL 41: F U R2 D R' U' R D' R2 F'
OLL 42: R' U' F2 u' R U R' D R2 B
OLL 43: R' U' F' U F R
OLL 44: F U R U' R' F'
OLL 45: f U R U' R' f'
OLL 46: F U F R' F' R U' F'
OLL 47: f' U' L' U L U' L' U L f /f' U' r' F r U' r' F r f
OLL 48: f U R U' R' U R U' R' f'
OLL 49: L F' L2 B L2 F L2 B' L
OLL 50: R' F R2 B' R2 F' R2 B R'
OLL 51: F U R U' R' U R U' R' F'
OLL 52: R U R' U R U' f R' f' R'
OLL 53: l' U' L U' L' U L U' L' U2 l
OLL 54: r U R' U R U' R' U R U2 r'
OLL 55: R' F R U R U' R2 F' R2 U' R' U R U R'
OLL 56: R U' R' U' S' R U' R' S U R U R'
OLL 57: R U' R' S' R U R' S
It'd be fun to see this done for PLL. Lately I've found a lot of new PLLs using MCC.
 
Joined
Aug 30, 2020
Messages
1,899
Location
On a long train journey, Smashin' PBs one a stop
YouTube
Visit Channel
I was comparing S/AS COLL/EPLL approach vs S/AS OCLL/PLL approach(by MCC because movecount isn't enough) and here's what I found:

Yes, OCLL/PLL wins by ~4 movecount coefficient points but this would probably be the case for all the COLLs so the statement "S/AS COLL isn't worth learning but the rest of COLL is" isn't so true imo.
 

tsmosher

Member
Joined
Aug 30, 2020
Messages
649
I was comparing S/AS COLL/EPLL approach vs S/AS OCLL/PLL approach(by MCC because movecount isn't enough) and here's what I found:

Yes, OCLL/PLL wins by ~4 movecount coefficient points but this would probably be the case for all the COLLs so the statement "S/AS COLL isn't worth learning but the rest of COLL is" isn't so true imo.
A similar comparison of JTLE/PLL vs. CDRLL/L5EP would be cool.

I'm also curious how expensive it would be if you're down to L4C/L5E (e.g., after DCAL)-- to just insert the D edge, OCLL, PLL vs. the above approaches.

My only critique: The algs completely ignore AUFs. There should be some initial AUF from the recognition angle or some approximation of initial AUF (e.g., 0.75) in there to make the comparison true. (I think this would skew the results more in favor of OCLL though.)
 
Joined
Aug 30, 2020
Messages
1,899
Location
On a long train journey, Smashin' PBs one a stop
YouTube
Visit Channel
The algs completely ignore AUFs. There should be some initial AUF from the recognition angle or some approximation of initial AUF (e.g., 0.75) in there to make the comparison true. (I think this would skew the results more in favor of OCLL though.)
yeah with AUF average MCCs would be:

COLL-EPLL: 27.9875
OCLL-PLL: ~24.75

it kinda skewed the results in favour of COLL lol
 

xyzzy

Member
Joined
Dec 24, 2015
Messages
2,465
I was comparing S/AS COLL/EPLL approach vs S/AS OCLL/PLL approach(by MCC because movecount isn't enough) and here's what I found:

Yes, OCLL/PLL wins by ~4 movecount coefficient points but this would probably be the case for all the COLLs so the statement "S/AS COLL isn't worth learning but the rest of COLL is" isn't so true imo.
You need to weigh the cases by probability. The two U perms are 1/3 each, Z perm is 1/6, H perm is 1/12, and EPLL skip is 1/12. (Same for the PLLs, and don't forget PLL skip there either.)

Also:
>17.4 S' U2 L' U2 L U2 L F' L' f
13 R' U2 R U F R' U R U' F'

>17.2 M2 U M2 U M' U2 M2 U2 M'
The "modern" Z perms have lower MCC, e.g. 16.2 M U M2 U M2 U M U2 M2. (And they're also faster in practice.)

>20 R' U R' U' R D' R' D R' U D' R2 U' R2 D R2
RUD V perm bad, RUf V perm cool: 17.5 R' U R U' R' f' U' R U2 R' U' R U' R' f R
 

kubesolver

Premium Member
Joined
Nov 15, 2019
Messages
332
You need to weigh the cases by probability. The two U perms are 1/3 each, Z perm is 1/6, H perm is 1/12, and EPLL skip is 1/12. (Same for the PLLs, and don't forget PLL skip there either.)

Also:
>17.4 S' U2 L' U2 L U2 L F' L' f
13 R' U2 R U F R' U R U' F'

>17.2 M2 U M2 U M' U2 M2 U2 M'
The "modern" Z perms have lower MCC, e.g. 16.2 M U M2 U M2 U M U2 M2. (And they're also faster in practice.)

>20 R' U R' U' R D' R' D R' U D' R2 U' R2 D R2
RUD V perm bad, RUf V perm cool: 17.5 R' U R U' R' f' U' R U2 R' U' R U' R' f R
Isn't it fairer to compare two methods by looking at move optimal algs only?

We can probably assume that all methods have similar optimization potential. Otherwise we're biased towards already popular over optimized methods
 

xyzzy

Member
Joined
Dec 24, 2015
Messages
2,465
Isn't it fairer to compare two methods by looking at move optimal algs only?
I don't know if it's truly fairer, but that is what I've been advocating:
[FTM-optimal move count] is a good (but imperfect) proxy for execution time with speed-optimised algs. Execution time is highly subjective, but optimal move counts aren't.

One of the uses of MCC is that it's an objective metric, so it's still possible to find "objectively optimal" algs for this specific metric, while also being closer to actual human execution time compared to the classical face/slice turn metrics. I guess one problem here is that the algs used in the method comparison weren't chosen to minimise MCC, which would indeed make comparing by MCC unfair.
 
Top