2LOLL Probabilities

[toastman]

So, currently I solve using a 2-look OLL method, and the other night was thinking

"Why do I keep getting this sodding Pi OLL, which I hate the algorithm for, but never seem to get double-headlights, which I love the algorithm for (because it's sexy-move compliant)".

So I decided to check the probabilities of each 2LOLL case. Couldn't find them, so generated them (the hard way).

Anyway, for your amusement. Please let me know if you see any errors.

Edges (I won't go through the calculations)

Solved:		1/8		12.5%
Dot:		1/8		12.5%
Line:		2/8		25%
L:		4/8		50%

Corners
[cut working out. Let me know if you want it]

Solved			1/27		 3.7037%		
Sune			4/27		14.8148%
Anti-Sune		4/27		14.8148%
Bowtie			4/27		14.8148%
T			4/27		14.8148%
Headlights		4/27		14.8148%
Pi			4/27		14.8148%
H / Dub-Heads		2/27		 7.4074%

So, question answered. You're twice as likely to get Pi as you are Double-Headlights.

 

[maggot]

there is an easy solution to Pi OLL, even though it is "sexy move compliant" r U R' U' (sexy move) r' F R F'. . .

i had a similar situation happen to me. i hate the headlights/superman OLL, so i learned all the cases in which EO would give me the headlights OLL.

for example M U M' U2 M U M' U' <-- setup cube with this.

now, to use a 2LOLL system, you would EO, F U R U' R' F', and end up with the headlights OLL. . . so, i learned the 1 look OLL case for the FISH (thats what i call it) that way i wouldnt have to worry about getting a crappy OLL after the EO.

what i did to accomplish this is i picked out all the cases that have no symmetry (the L cases) and found all the OLL that would lead to headlights after EO. then i worked on the line cases, in which each OLL had to be performed 2 different ways, the first way as the OLL is performed, and then i would AUF with U2 and see if that would end up with the headlights. Those OLL took me a bit longer to figure out. Shortly after I learned all the OLL with EO headlights, i found that over 1/4 of my OLL had been learned. Of course, i knew the easy T cases, no corner dot cases, all corner dot case, etc. . . but learning full OLL from there was fairly easy, as all the remaining OLL felt like a subset of what i had already tackled. hopefully this experiment of yours will enable you to cruise through full OLL like I did!

 

[Anonymous]

If I understand correctly, Pi is actually what you're referring to, maggot, as headlights / superman.

Your finds are interesting, although I already had a feeling this was true, because there are fewer algs in the double Sune COLL than all the others.

maggot's advice seems logical, although I will point out that if you don't intend on learning full OLL for whatever reason, his approach might make you learn more algs than you need. I assume you use the 2-gen RU alg for Pi, right? How fast can you execute it?

 

[Lucas Garron]

So I decided to check the probabilities of each 2LOLL case. Couldn't find them, so generated them (the hard way).

Anyway, for your amusement. Please let me know if you see any errors.

Edges (I won't go through the calculations)

Why the hard way? The easy way is way better.

Also, that's a good point, probabilities should probably be in the wiki.

 

[toastman]

If I understand correctly, Pi is actually what you're referring to, maggot, as headlights / superman.

Your finds are interesting, although I already had a feeling this was true, because there are fewer algs in the double Sune COLL than all the others.

maggot's advice seems logical, although I will point out that if you don't intend on learning full OLL for whatever reason, his approach might make you learn more algs than you need. I assume you use the 2-gen RU alg for Pi, right? How fast can you execute it?

Thanks for your reply!
Yes, I intend to learn full OLL.

Currently, I'm doing the R U2' R2' U' R2 U' R2' U2 R alg for Pi, but I find I seem to need to "re-grip" a couple of times. Might just be a matter of practice. Might also try f (R U R' U') (f' F) (R U R' U') F', as it's similar to a case I already have memorized. Can't hurt to know both.

 

[Maniac]

Start with your thumb on the DR edge and your middle finger on the UR edge and preform the algorithm. You shouldn't need to regrip.
R U2 R2' U' R2 U' R2' U2' R

 

[da25centz]

Thanks for your reply!
Yes, I intend to learn full OLL.

Currently, I'm doing the R U2' R2' U' R2 U' R2' U2 R alg for Pi, but I find I seem to need to "re-grip" a couple of times. Might just be a matter of practice. Might also try f (R U R' U') (f' F) (R U R' U') F', as it's similar to a case I already have memorized. Can't hurt to know both.

i regrip several timed. I do R U2 and then i do R2 and while i do U' i bring my right hand back for the next r2

 

[FatBoyXPC]

R U2 R2 U' R2' U' R2 U2' R

I regrip after the U2. I do a double flick with my right index/middle fingers. The last U2 is just using my left index twice. Ideally I should learn to do a double flick w/my left/middle so that I don't have to have any regrips.

Using the F/f SM F'/f' algs back to back and with some AUFs, you can solve a few OLLs. Including Stingray, H, Pi as you found, etc.

 

[Kenneth]

Hold the 2x2x3 block with your left hand, thumb on F, pinky on D, next two on B and index free on top. Do R using your right hand and then U2' in one go using the LEFT! index (this is the 'gunnarfinger') Then R2 U' (all U using left index) R2 U' R2 U2' R.

No regripping, just extreamly fluid =)

Edit, I made a vid that shows how :

Sorry for the crappy lighting conditions, but it is still early morning here, no daylight :/

 

[peedu]

R U2 R2 U' R2' U' R2 U2' R

Another way is to turn R starting with right thumb on DRF, right index, middle and ring on URF, UR and URB stickers somewhere. After R turn push U' with thumb on FUR and pull another U' with left index on BLU. During that last U' you can regrip right hand depending on how you want to end the algorithm.


Peedu

 

[oll+phase+sync]

If you have a Last Layer case with 4 misoriented Corners ,

regardless wich direction the first corner pionts to, to form a doublesune to form the second corner has just one direction left , while for forming a Pi the second corner can point in one of two directions, that's why the probability is twice as high.

Some time ago I mesured my execution times for Bruno and Doublesune they where pretty much the same.

 

[Pyjam]

Each time (F SM F') leads to a bad position, try (U2 f SM SM f') instead.
Each time (f SM f') leads to a bad position, try (U2 F SM SM F') instead.

SM = Sexy Move = R U R' U'

 

[maggot]

I never responded to my 'pi' being the wrong case here.... I call this oll truck lmao... but w/e.
For this oll I use the typical, with regrips I would lockup all the time with the inaccurate R2s, but after doing like a sq1 with rocking back and forth kinda deal, this one is definately one of my fav oll to perform =)
Good luck with this case and for advice, don't use f sexy f' F sexy F' ... its so slow... its only fast for zamboni, and even then I hate it =(