ZZ Cubers

[JohnnyA]

Doing Fridrich F2L on a ZZ solve makes no sense. It will just misorient the last layer edges. The strength of ZZ I think is not so much the move count but the super fast F2L after eoline/cross, since it is 3-gen f2l. Also don't forget intuitive phasing of the LL leads to 130 (correct?) 1LLL cases.

 

[a small kitten]

I think he means cross, and then slot everything like Fridrich with RUL. I'm still doing that a little because it does eliminate awkward slotting cases that I hated. I think it's 167 cases for 1LLL.

"The main difference lies in so called "phasing". Just when finishing F2L, one must permute two opposite LL edges, usually UF and UB. It takes 2-4 moves on average and can be done on the fly (of course after proper training). Phasing style is individual to each speedcuber, as it depends on F2L techniques. However, you can see basic approach to this problem in Phasing section in Practical notes.

After ZZF2L comes the standard 1-look LL. However, fact of two opposite edges permutted correctly drastically decreases a number of possible LL cases. It's 80 distinct algorithms. When we count inversions and symmetries, it gives us 167 sequences. This is 2x that of Fridrich, but it's still a reasonable number that can be learned and recognized quickly"

 

[JohnnyA]

I think he means cross, and then slot everything like Fridrich with RUL. I'm still doing that a little because it does eliminate awkward slotting cases that I hated. I think it's 167 cases for 1LLL.

"The main difference lies in so called "phasing". Just when finishing F2L, one must permute two opposite LL edges, usually UF and UB. It takes 2-4 moves on average and can be done on the fly (of course after proper training). Phasing style is individual to each speedcuber, as it depends on F2L techniques. However, you can see basic approach to this problem in Phasing section in Practical notes.

After ZZF2L comes the standard 1-look LL. However, fact of two opposite edges permutted correctly drastically decreases a number of possible LL cases. It's 80 distinct algorithms. When we count inversions and symmetries, it gives us 167 sequences. This is 2x that of Fridrich, but it's still a reasonable number that can be learned and recognized quickly"

Ah well. I'm still a long way away from learning the 1LLL algs. I do EOCross too, and Fridrich style slots, but I thought he meant doing cross with EO and then doing regular Fridrich, i.e. with rotations.

Does anybody here use block building for F2L after eoline? Can you give examples of how you do this? (never used any blockbuilding xD)

 

[fanwuq]

I like ZZ, but there are 2 things that are keeping me away from it.

1. I like to see my blocks during inspections.
2. I like building my blocks without rotation or move restrictions.

Funny thing is that much of the time when I use Fridrich, I can build an extended cross, and use less than 2 rotations during the whole solve.
When I try ZZ, I can get the EO line in 5 seconds or so, that's not an issue, but for some odd reason, I'm tempted to rotate during my F2L and my lookahead for F2L is just ruined.
For ZZ's LL, I suggest to use MGLS and PLL. MGLS is 2gen and PLL is fast anyway.

 

[Stefan]

Does anyone have any concrete evidence that ZZ has a lower move count than Fridrich?

A while back I did some statistics suggesting the opposite, though I lost track of the thread afterwards:
http://www.speedsolving.com/forum/showthread.php?p=69583#post69583

 

[JohnnyA]

I thought it was that ZZ f2l with "3gen fridrich" took more moves, but 3gen blockbuilding of F2L took less?

I find lookahead difficult in ZZ, with the result that I try harder and actually get the same sort of lookahead. I also found that doing a lot of ZZ solves using all my effort in lookahead improved my normal lookahead by quite a lot, so I would suggest all cubers to at least learn ZZ to see if they can improve their main method too.

I think I will keep ZZ as a sub-method until after the French Open since I want to do as well as I can in that, but after the French Open I am going to concentrate on ZZ because I believe it is a better speedcubing method.

 

[Matthew]

Hello guys

I've been using ZZ method (precisely the ZZ-VH method - I do EOLine -> ZZF2L -> COLL -> EPLL) since december last year...

apart from zborowski, are there any pure (or mostly) ZZ users who are sub 18/17?

Yes

My records:

single: 10.53
avg12: 13.99
mean of 100: 15.78
mean of 1000: 16.33

If someone was not convinced to this method I think this can change his view about this method

Does anyone have any concrete evidence that ZZ has a lower move count than Fridrich? So far my experience tells me no, at least up to the completion of F2L, it definitely takes more moves than Fridrich.

Lower moves count is not everythig.. ZZ in my opinion and after my tests has a lower moves count but what is important - it has also a 1LLL, no cube-rotations and easier F2L

Now I'm working about full b variation..

 

[JohnnyA]

apart from zborowski, are there any pure (or mostly) ZZ users who are sub 18/17?

Yes

My records:

single: 10.53
avg12: 13.99
mean of 100: 15.78
mean of 1000: 16.33

If someone was not convinced to this method I think this can change his view about this method

Does anyone have any concrete evidence that ZZ has a lower move count than Fridrich? So far my experience tells me no, at least up to the completion of F2L, it definitely takes more moves than Fridrich.

Lower moves count is not everythig.. ZZ in my opinion and after my tests has a lower moves count but what is important - it has also a 1LLL, no cube-rotations and easier F2L

Now I'm working about full b variation..

Wow! A real ZZ user. Tell me do you blockbuild F2L?

 

[Matthew]

Wow! A real ZZ user. Tell me do you blockbuild F2L?

Yes - I use blockbulding - it is a bit more difficult but much more cost-effective...

 

[Escher]

matthew, im surprised that you havent made yourself more known - you may well be the fastest ZZ user in the world!

anyway, clearly ZZ can be extremely, extremely fast

- i thought that EOLine averaged 5 or 6 moves, whereas cross avgs 7/8, and the worst possible cross case was 10/11 moves, whereas the worst EOLine was 9 or 10? opposite cross doesnt make them even either.

- to those who get 8 or 10 misoriented edges, why not just do a y/y'?

- i'd practice doing 2 gen solves to get used to the 'awkward' f2l cases.

 

[yurivish]

I'm consistently getting sub-35 for OH now, with occasional ~45-second bad cases My Fridrich average is about 30, I imagine myself getting much faster once I learn COLL :rolleyes: Any opinions?

Edit: @yurivish: I just got what you meant by "just looked at your sig"! Umm, no, my err... what do you call that? "Proficiency"? My proficiency(?) in EO has nothing to do with BLD, I do M2 I just did a bit of homework for about an hour before I started ZZ, so I got the whole "intuitive feel" for EO, that's all. I'll post something up in this thread once I've done enough homework, it's probably gonna do beginners some good if they're lost at EO.

Keep at it with the left hand. I'm still not anywhere near ambidextrous, but I can comfortably do L U L' U'-type things now. Eventually I want to be completely comfortable with both hands.

COLL—I highly recommend it. I was skeptical, as well as slow, when I first started COLL. But now I use it for all the cases except for Sune/Antisune.

I use all M-slice-based edge PLL's, so the last part of a solve is always quick and fun.

 

[Matthew]

- i thought that EOLine averaged 5 or 6 moves, whereas cross avgs 7/8, and the worst possible cross case was 10/11 moves, whereas the worst EOLine was 9 or 10? opposite cross doesnt make them even either.

0 moves : 1 case
1 : 9
2 : 91
3 : 851
4 : 6831
5 : 41703
6 : 130239
7 : 88683
8 : 1927
9 : 1

Statistic made by Grzegorz Luczyna

9-move case is when we have a 12 misoriented edges and the DF and DB edges are in their places..

- to those who get 8 or 10 misoriented edges, why not just do a y/y'?

This not solve completely our problem (edges orientation) - in some cases this can be helpful but in some.. this can disturb.. and we have only 15s preinspection

 

[bamman1108]

The numbers of moves for EOLine and the cross are about the same.

 

[Escher]

The numbers of moves for EOLine and the cross are about the same.

if you read the link given earlier in the thread, and go to the EOLine section, youll see that 48% of the time, it takes 6 moves, 32% of the time it takes 7 and and 15% of the time it takes 5.

then if you go to vandenbergs site -cubezone.be - the statistical analysis of the cross shows that only 8% of crosses are 5 moves, 30% are 6 moves, 48% are 7, and 11% are 8.

so if we are going to be accurate (which we should be) EOLine is, on average, one move less than cross, and more likely to be fewer moves than 7.

using opposite colour neutrality for cross and f2l vs standard EOLine, you can see that it becomes a bit more even, but EOLine still wins out.

using colour neutrality on cross only, cross wins. i would imagine that colour neutrality on EOLine makes things significantly better for EOLine.

 

[Stefan]

if you read the link given earlier in the thread, and go to the EOLine section, youll see that 48% of the time, it takes 6 moves, 32% of the time it takes 7 and and 15% of the time it takes 5.

then if you go to vandenbergs site -cubezone.be - the statistical analysis of the cross shows that only 8% of crosses are 5 moves, 30% are 6 moves, 48% are 7, and 11% are 8.

so if we are going to be accurate (which we should be) EOLine is, on average, one move less than cross, and more likely to be fewer moves than 7.

Just preserving this so you can't hide your shame...
I won't explain it just yet, rather I give you the chance to rectify your attempt to deceive or honest utter failure, whatever it is.

 

[Escher]

lol, okay.

what i really meant was...

EOLine takes less moves than cross. what i said wasnt being accurate, it was generalising from two tables.

if i was really going to be accurate, i would put up both of the tables comparing the two in QUARTER turn metric, without colour neutrality or opposite cross or whatever. you could draw your own conclusions.

am i ok now?

 

[Stefan]

No you're still utterly wrong.

 

[Escher]

damn...
well. i like being wrong...
erm...
i honestly can say i have no idea.
tell me!

 

[Stefan]

Please do show us those two tables for comparison.

 

[Escher]

Quarter turn metric

Depth # cases Distribution
0 1 0.00%
1 10 0.01%
2 73 0.04%
3 500 0.26%
4 3,078 1.62%
5 15,528 8.17%
6 57,180 30.08%
7 91,654 48.22%
8 2,1849 11.49%
9 207 0.11%
Average: 6.59 moves
is the first, from cubezone.be, on the cross, with quarter turn metric, with the cumulative percentages omitted...


from

http://www.emsee.110mb.com/Speedcubing/ZZ speedcubing system.html#id4

no. of moves for EOLine
0 : 1
1 : 9
2 : 91
3 : 851
4 : 6831
5 : 41703
6 : 130239
7 : 88683
8 : 1927
9 : 1

In percentages...
0 : 0.00037%
1 : 0.00329%
2 : 0.03366%
3 : 0.31479%
4 : 2.52686%
5 : 15.42636%
6 : 48.17671%
7 : 32.80473%
8 : 0.71282%
9 : 0.00037%

sorry about the first table, can't seem to get the formatting right.