[Help Thread] Petrus Discussion and Help thread

[Noahaha]

I copied this from here so more people would see it and I could get more feedback.
I've got a bad edges video tutorial on the way.





​

I don't know why people have so much trouble with EO, but they do so I guess it's good that there's a tutorial. Also, no petrus tutorial needs to be that long as long as the intuition is explained and not just the moves.

 

[JLarsen]

Tutorial is up. Link in my sig. Let me know how I did.

 

[pi.cubed]

The alternative name for EJF2L is the original - MGLS. That's what it is. Just a little trick. And yea, didn't improve my times. I used it for like a year in about 1/3 of my solves. Some of the cases are just....ugh

Ah okay that makes sense.

Okay so this conversation is kinda over considering it was 2 weeks ago, but now I have another question.

What alternative advanced LL methods do you think are worth learning for Petrus? Well obviously OCLL followed by PLL can be pretty fast but I'm sure there is something faster. I believe Erik J uses COLL. Do you think that is worth learning and using instead of/as well as OCLL+PLL? What about ZZLL?

I just want some opinions from experienced Petrus users.
Thanks

BTW, I wasn't able to watch your bad edges video, but I read the text description and it was brilliant!

 

[Noahaha]

Ah okay that makes sense.

Okay so this conversation is kinda over considering it was 2 weeks ago, but now I have another question.

What alternative advanced LL methods do you think are worth learning for Petrus? Well obviously OCLL followed by PLL can be pretty fast but I'm sure there is something faster. I believe Erik J uses COLL. Do you think that is worth learning and using instead of/as well as OCLL+PLL? What about ZZLL?

I just want some opinions from experienced Petrus users.
Thanks

BTW, I wasn't able to watch your bad edges video, but I read the text description and it was brilliant!

If you have the motivation, learn COLL, although it's not really necessary. If you're into OH, you can't go wrong with CPLS+2GLL. I find WV, EJF2L and JTLE all pretty frivolous although I do use EJF2L sometimes for OH if I see a way to save a lot of moves. CLS is another option, but I personally prefer CPLS.

 

[cunningcuber]

Okay so this conversation is kinda over considering it was 2 weeks ago, but now I have another question.

What alternative advanced LL methods do you think are worth learning for Petrus?

well you can always try the ll system at Lars Petrus website. you have to memorize around 40 algs to solve any case(as long as you have the edges oriented) in around 15 moves and it's quite fingerfriendly

 

[oll+phase+sync]

What alternative advanced LL methods do you think are worth learning for Petrus? Well obviously OCLL followed by PLL can be pretty fast but I'm sure there is something faster. I believe Erik J uses COLL. Do you think that is worth learning and using instead of/as well as OCLL+PLL? What about ZZLL?

ZZLL: I used Phasing it is really easy and fast (and often comes for free) , but over 160 LL algs is quit a lot to learn. I don't know any (success)stories.

CLS/MGLS: just posted in another thread, that I beleave this could really speed you up. (the problem setting up a correct EJFS case during Step 2 is gone this way)

WinterVariation: I don't think its worth to know all of it but I use a few cases

COLL: My advice: learn all cases (to get good at recognition) , after you have done that use it in the cases you like and skipp it otherwise (I think ErikJ an David Cohen don't use COLL for Sune cases for exsample )

 

[5BLD]

BBF2L is pretty good too

 

[Escher]

ZZLL: I used Phasing it is really easy and fast (and often comes for free) , but over 160 LL algs is quit a lot to learn. I don't know any (success)stories.

This guy uses ZZLL.

 

[JLarsen]

I too am looking for a better LL or LS system to make EO actually worth doing. (Since you're solving edges in 2 steps for little benefit but 2 gen F2L finish)

A quick question about LS systems in Petrus. Is there any accommodation for which "slot" you decide to finish last? Like what if I ended with the unsolved slot on the back? Can I just rotate just like I do with the MGLS I and IM sets?

 

[Noahaha]

I too am looking for a better LL or LS system to make EO actually worth doing. (Since you're solving edges in 2 steps for little benefit but 2 gen F2L finish)

A quick question about LS systems in Petrus. Is there any accommodation for which "slot" you decide to finish last? Like what if I ended with the unsolved slot on the back? Can I just rotate just like I do with the MGLS I and IM sets?

If the edge is placed in it's correct spot, all edges are good even if you do a y rotation, but if the last edge is in the top layer, doing a y rotation will make it and the edge in the slot both bad. Since CLS is executed with the last edge in place, doing a y should be fine. For any other system though, you'll have to mirror your algs to L U.

 

[PandaCuber]

Can someone please write down algs for EO , for me please

Ive only figured out how to do 2 at a time. And how to recognize.

Petrus

 

[JLarsen]

Can someone please write down algs for EO , for me please

Ive only figured out how to do 2 at a time. And how to recognize.

Petrus

There is only one "alg" R' L F R L' and R' L F' R L'. Everything else is just a cancellation of your standard 3 move edge flips. For instance, you can flip two edges with R U' R', but you can flip 4 edges with R U' R' R' F' R. The Rs cancel. To do a 6 flip, do the 4 flip I mentioned first and cancel into the appropriate 2 flip.

Grats you just learned every case on Lars' site.

If you're a more visual person or you're just not getting it watch the second half of my bad edges tutorial.

Edit: I made some revisions to the text portion of my guide to better explain 4 flips here

 

[cunningcuber]

I have been thinking... How about ZBF2l? You can insert a corner/edge pair while orienting all edges on the top layer.so you place the bottom layer edget then you do Zblf2l and voila!You've done step 4a end edge orientation in a few moves.

 

[oll+phase+sync]

...You can insert a corner/edge pair while orienting all edges on the top layer.so you place the bottom layer edget then you do Zblf2l and voila!You've done step 4a end edge orientation in a few moves.

When I see an easy 4a block I always skip step 3 (orientation) and solve 4a first.

But I took a look at the ZBF2L cases with a connected CE pair in U : besides the few obvious cases (sledge hammer ...), most other algs look long (and ugly) - RUR' set is also not that great (yet things are even less obvious ) in my opinion.

 

[mariano.aquino]

I have been thinking... How about ZBF2l? You can insert a corner/edge pair while orienting all edges on the top layer.so you place the bottom layer edget then you do Zblf2l and voila!You've done step 4a end edge orientation in a few moves.

Bare in mind that you usually do ZBF2L while looking at the last 5 edges left to orient, but at the moment you start step 4, there's 6 edges to orient, and that ruins your recognition time/look ahead SEVERELY =S

 

[cunningcuber]

very good point you two.
P.S.:Sorry for taking so long to reply. I had a lot of studying to do

 

[mDiPalma]

what are reasonable movecounts for each step: 2x2x2, 2x2x3, eo, right block ?

 

[Noahaha]

what are reasonable movecounts for each step: 2x2x2, 2x2x3, eo, right block ?

2x2x2: 5-7
2x2x3: 6-8
EO (With 2 or 4 bad edges): 4-6
Right Block (not sure about this one because it's my worst step): 12-18 (obviously you can get a few 3-move cases and be a lot lower)

 

[F perm]

I think this page will help to fill in the Right Block step movecount, though it was originally for ZZ, so you have to scroll down to "Solving second 3x2x1 block"

I have no idea about the other steps though. Noahaha's stats look legit.

 

[Noahaha]

I think this page will help to fill in the Right Block step movecount, though it was originally for ZZ, so you have to scroll down to "Solving second 3x2x1 block"

I have no idea about the other steps though. Noahaha's stats look legit.

Those numbers are optimal and probably too low for a speedsolve by 5-7 moves, so I'm sticking with my guess. Matt uses ZZ mostly anyway, so he's most familiar with that step.