[Help Thread] Petrus Discussion and Help thread

[sud_darkhell]

Except that Petrus 5+6 is just inefficient COLL, so perhaps it's better to learn that.

I for one think that a system with 11 algs (Petrus 5+6) is much more efficient than your usual COLL (42 algs). Although the former has slightly more average number of turns, I find it much more easier to execute.

 

[Noahaha]

Petrus Home Thread

I for one think that a system with 11 algs (Petrus 5+6) is much more efficient than your usual COLL (42 algs). Although the former has slightly more average number of turns, I find it much more easier to execute.

It's efficient in terms of alg number, but when most people say efficient they're talking about moves. I personally think that it's like this:

Petrus 5+6<OLL+PLL<COLL+EPLL

 

[Ooi Yan Qing]

Doing Petrus in a different way

I have an idea to use petrus in a different way. Usually we'll build the 2x2x2 block and expand it, but i found that the step is not very efficient for me.
Making the 2x2x2 block usually will use some B, F, and D moves and X, Y, Z as well. This will actually slow you down in execution.
If we look into Roux, we will see that one of the thing that make it fast is because minimum B, F, D moves and zero X, Y, Z moves is used in the whole solve.
Ok, let's go into how i solve in petrus.

First step: Solve a 1x2x3 block at the L slice (Roux First step)
2nd Step: Expand the 1x2x3 block into 2x2x3 (Just have to solve 2 pieces, and only r, R, U moves is needed)
3rd Step: Edges orientation (Original Petrus Step)
4th Step: The rest 1x2x3
5th Step: OLL and maybe OLLCP as well
Last Step: PLL

Hopefully you guys have some feedback, cause i don't know if this is the better way to solve in petrus

 

[aznanimedude]

this is OBLBL for the 3x3x3 isn't it lol?

 

[mDiPalma]

ya but if you can orient the other edges during the 3x2x2 expansion step, then this might be a viable method. ill try it some.

 

[Ooi Yan Qing]

yes you can influence the edges orientation during the expansion step, just using the way you orient edges in roux (M' U M case) to insert an edge into the MFD or r U' r' to preserve oriented edges

 

[oll+phase+sync]

If I see an easy 3x2x1 at the beginning I alwas start my petrus solves with this block, though I don't care about edges orientation at all, while placing the D edges.

This is a collection of my thoughts, I had about this idea. (I found it to difficult to use in speedsolves, and focused on pseudo blocks instead)

- not seeing DBM edge is just anoying
- iterativ orientation is smuch harder now, because M'UM and rUr' orient 4 edges at atime while M'U2M orients none
- there are nine edges to scan before starting the step (quite a lot)

at least I believe you must know the number and position of misoriented edges because while placing the first D-edge you should make sure
a) 4 edges keep missoriented (or all edges become oriented)
b) if 4 edges are missoriented the final D-edge should be one of them.

 

[chimchu]

thanks guys i was trying to decide between petrus and roux

 

[Noahaha]

Petrus method...

thanks guys i was trying to decide between petrus and roux

You should probably know that at this point the fastest Roux users are much faster than the fastest Petrus users (if there are any). If you're just doing it based on "potential speed" then Roux is the obvious choice, but Petrus is more fun and stuff so it's a good choice as well. It's best to learn all four main methods (CFOP, Roux, ZZ and Petrus) and then decide which one you like best.

 

[chimchu]

yea, i figured. i really dont like fridrich though. i share petrus idea that it seems stupid to build everything then break it up. but it is "the fastest", i just dont want to learn the algs right now.

 

[mark49152]

i share petrus idea that it seems stupid to build everything then break it up.

I have wondered about the validity of this quote. In the abstract it makes sense, but practically isn't that just the way the cube works? A commutator breaks then restores, regardless of method. Does Petrus LL not break and restore a load of stuff? Is there any truth in the suggestion that avoiding disruption to already-solved pieces leads to a faster or more efficient solution?

 

[Noahaha]

Petrus Home Thread

I have wondered about the validity of this quote. In the abstract it makes sense, but practically isn't that just the way the cube works? A commutator breaks then restores, regardless of method. Does Petrus LL not break and restore a load of stuff? Is there any truth in the suggestion that avoiding disruption to already-solved pieces leads to a faster or more efficient solution?

That quote refers to Petrus F2L vs layer by layer, which is what most other people were using back in the 80s. So it is accurate, but only in relation to beginner's method.

 

[mark49152]

That quote refers to Petrus F2L vs layer by layer, which is what most other people were using back in the 80s. So it is accurate, but only in relation to beginner's method.

Makes sense; so I guess it could also be considered accurate if applied to CFOP versus LBL F2L?

 

[Noahaha]

Petrus Home Thread

Makes sense; so I guess it could also be considered accurate if applied to CFOP versus LBL F2L?

Yes. That is a feature that all fast methods share.

 

[mark49152]

Yes. That is a feature that all fast methods share.

I'm only just getting interested in Petrus. Looks to me like the main difference between these methods is whether you do EO before (ZZ), at the mid-point (Petrus) or after (CFOP) the F2L phase. Yes, block-building is a characteristic of Petrus and strictly also of ZZ, but I assume solving by slot is a legitimate way to build those blocks if the cuber really is more comfortable with that, and although frowned upon, it doesn't make the method something else.

I also read some articles on partial edge control, and my question is, does anybody use this kind of technique to solve EO fully and intuitively during F2L, rather than one of those three prescribed separate steps? Is it realistic to aim to always finish F2L with edges oriented just by a sequence of careful choices of slot-insertion or block-building moves, without a dedicated step? Or is that just too slow?

 

[aznanimedude]

Re: Petrus Home Thread

What you're taking about is zbls where you finish f2l and orient last layer edges so you get a last layer cross. and not many if not no one uses it because of the huge movecount

 

[mark49152]

What you're taking about is zbls where you finish f2l and orient last layer edges so you get a last layer cross. and not many if not no one uses it because of the huge movecount

No I'm talking about doing EO intuitively over the last 2-3 slots or while block-building. In practice the last slot might be solved by one of the easy ZBLS cases but my question is more about whether edge control can be used earlier than that to make sure that you do indeed get a really easy (preferably intuitive) ZBLS to finish.

 

[mDiPalma]

You can see a stepwise process very much like that in Michał Pleskowicz' reconstructed OH solves. From what I can tell, his usual approach is xcross, f2l with smart eo insertions, coll/epll/zbll.

See Brest's reconstructions of averages like this one.

 

[mark49152]

Thanks mDiPalma, I will check that out.

 

[oll+phase+sync]

I have wondered about the validity of this quote. In the abstract it makes sense, but practically isn't that just the way the cube works? A commutator breaks then restores, regardless of method. Does Petrus LL not break and restore a load of stuff? Is there any truth in the suggestion that avoiding disruption to already-solved pieces leads to a faster or more efficient solution?

Makes sense; so I guess it could also be considered accurate if applied to CFOP versus LBL F2L?

On Petrus Website it is stated, that not breaking parts of the cube, that are already build, is one idea wich inspired Petrus to make up the methode as it is.

-But for sure there is no proof that this principle yields a better movecount in every case.
-the EO step itself contradicts this principle

CFOP F2L contradicts this principle even more.

But Roux is very close to this principle! I never looked at Roux that way before.