[Help Thread] Petrus Discussion and Help thread

[5BLD]

Thanks for the tips Cyrus.
Yes, RU always seems to be so long... But of course turn speed will make up for it.
And I'll look into EJF2L.

 

[oll+phase+sync]

For EO I use LRMFU too, regarding EJF2L do you use y' + mirrored alg for the mirrored cases or do you think learning a different alg without a y is worth it? (e.g. invers)

 

[Cyrus C.]

For EO I use LRMFU too, regarding EJF2L do you use y' + mirrored alg for the mirrored cases or do you think learning a different alg without a y is worth it? (e.g. invers)

I just went ahead and learnt all of the cases. Eight algs isn't too bad.

Cubersmith, maybe post some 2x2x3 expansion examples?

 

[JLarsen]

I just went ahead and learnt all of the cases. Eight algs isn't too bad.

Cubersmith, maybe post some 2x2x3 expansion examples?

16 including (reflections?) The other ones that we're not counting.

For 2x2x3 I found that slow solves and just figuring out cases worked best for me. The main thing I focused on was trying to solve 2x2x3 from all of the ergonomic angles, for instance holding the 2x2 block in all three orientations while it's in the bottom left half. Even anywhere in the bottom if you are good with both hands. I've got a slight bias to RU because of Step 4 =P. Just make sure that EO isn't some absurd rotation away when you're finished expanding and you'll be all set.

 

[Cyrus C.]

16 including (reflections?) The other ones that we're not counting.

Learning the reflections would be eight though, and that's what he was asking about.

 

[JLarsen]

Learning the reflections would be eight though, and that's what he was asking about.

Do you and others really find reflections are that easy to learn that you don't even count them as separate algorithms? I feel like when it comes down to it it's still a completely different alg that may not work as well for you as the other.

 

[jms_gears1]

Do you and others really find reflections are that easy to learn that you don't even count them as separate algorithms? I feel like when it comes down to it it's still a completely different alg that may not work as well for you as the other.

Generally mirroring an alg is fairly easy. But generally when I mirror RU algs to LU, I do a z so that LU=UR

 

[Cubenovice]

I found that sometimes I get a better flow with lefty versions compared to my right hand versions.
Especially the U-perm R U' R U R U ... The mirror L' U L' U' L' U'... goes soo much more fluent.

This weekend I also found out that I can actually use my thumb for the F move in left J-perm.
I've never managed the F' with my thumb in righty algs

 

[Ranzha]

I'm gonna be trying out Petrus to help with case recognition and piece recognition in general. =3
So like, yesterday I posted an example solve in the NA Speedcubers group on fb and on my wall, with no responses. Perhaps I should've consulted this thread first xD
Feedback would be greatly appreciated =D

Scramble: F' R' F2 R2 D2 U R2 U' L2 R' D' U' L2 F L B2 D2 U2 F B2 U' R' U D' L2
2x2x2: L2 U2 M2' F' L x2 [5]
2x2x3: U R2 U F U2 F2 [6/11]
3rd pair: R U2 R' [4/15]
4th pair + EO: M' (R') U R U' M (Yes, I know I should've EO'd first; shoot me) [5/21]
COLL: U' L' U2 L2' U L2' U L2' U2 L' [10/31]
EPLL: M' U M2' U M2' U M' U2 M2' U [10/41]

~Ranzha

 

[mariano.aquino]

I'm gonna be trying out Petrus to help with case recognition and piece recognition in general. =3
(...)
Feedback would be greatly appreciated =D

Scramble: F' R' F2 R2 D2 U R2 U' L2 R' D' U' L2 F L B2 D2 U2 F B2 U' R' U D' L2

~Ranzha

Welcome! a couple of ideas i came up with while solving:
2x2x2: L2 U2 M2' F' L
then:
x' U2 x' u' R u y
EO: r U' r' R U' R' y'
F2L: R' U2 R' U' R' U2 R'
COLL: (U2) R' (F R U' R' U' R U R' F R U R' U' R' F R F') R
EPLL: R2' U' FB' R2 BF' U' R2 (U')

Wierd 2lll just for fun xD
cp: x R'UR'D2RU'R'D2R2 x'
co: (URU'R') x2 L' (RUR'U') x2 L

Enjoy, cya!

 

[Cyrus C.]

Do you and others really find reflections are that easy to learn that you don't even count them as separate algorithms? I feel like when it comes down to it it's still a completely different alg that may not work as well for you as the other.

Reread the last couple posts of that discussion. I don't think you understand what I was saying (maybe I'm misunderstanding you though). He asked whether it was worth learning new algorithms for the reflections, or if we just used mirrors. I said you might as well learn new algorithms, because it's only eight more, and there are some better algorithms for doing the mirrors, along with not having to do a rotation.

To your actual question though, I find LU reflections easy to learn. Executing them as y', and then RU is difficult for me, though. I think you could get a handle of it with some practice though. It's probably useful if you're learning a large set of algorithms.

I'm gonna be trying out Petrus to help with case recognition and piece recognition in general. =3
So like, yesterday I posted an example solve in the NA Speedcubers group on fb and on my wall, with no responses. Perhaps I should've consulted this thread first xD
Feedback would be greatly appreciated =D

Scramble: F' R' F2 R2 D2 U R2 U' L2 R' D' U' L2 F L B2 D2 U2 F B2 U' R' U D' L2
2x2x2: L2 U2 M2' F' L x2 [5]

For the 2x2x2 I would have done R2 D2 B' L. Notice How the R2 builds the 1x2x2, as well as places the Blue-Yellow edge.

2x2x3: U R2 U F U2 F2 [6/11]

Yep, that's how I would have solved that.

3rd pair: R U2 R' [4/15]
4th pair + EO: M' (R') U R U' M (Yes, I know I should've EO'd first; shoot me) [5/21]

I would have solved this with:
y U L F' L'
y' R U R U R2 U R U R U' R'

But, your way of doing it looks fine. You may have gotten lucky with your EO though, you could have been stuck with a pretty bad case if you had been unlucky. I'd do EO first if you want to stay constant. Also, doing EO first gives you the nice RU gen extension.

~Ranzha

-Cyrus

 

[JLarsen]

Reread the last couple posts of that discussion. I don't think you understand what I was saying (maybe I'm misunderstanding you though). He asked whether it was worth learning new algorithms for the reflections, or if we just used mirrors. I said you might as well learn new algorithms, because it's only eight more, and there are some better algorithms for doing the mirrors, along with not having to do a rotation.

To your actual question though, I find LU reflections easy to learn. Executing them as y', and then RU is difficult for me, though. I think you could get a handle of it with some practice though. It's probably useful if you're learning a large set of algorithms.
-Cyrus

Ah. That makes sense. I did read the posts a little out of context. I personally use the y' RU algs.

When you say "RU is difficult for me" I'm assuming you mean it's difficult for you to translate an original RU alg to one with a y? As in trying to figure one out by knowing the other without treating it as a completely different alg? I've never bothered doing reflections of things. When I learned the subsets of MGLS I didn't even realize they were reflections. I just treated them as separate algs and learned them. On a relevant note a lot of the algs are incredibly similar. I've always hated learning RU alg sets because it's so hard to remember them all. I've forgotten all of the super freakin long ones over the course of my (2 year?) hiatus. They don't really seem to make that much of a difference to me time wise so I didn't bother relearning them. They're kind of tricky to get used to an incorporate for me. IMO they just complicate the solve too much to be worth it. They are really freakin cool though...hehehe

 

[Cyrus C.]

Ah. That makes sense. I did read the posts a little out of context. I personally use the y' RU algs.

When you say "RU is difficult for me" I'm assuming you mean it's difficult for you to translate an original RU alg to one with a y? As in trying to figure one out by knowing the other without treating it as a completely different alg? I've never bothered doing reflections of things. When I learned the subsets of MGLS I didn't even realize they were reflections. I just treated them as separate algs and learned them. On a relevant note a lot of the algs are incredibly similar. I've always hated learning RU alg sets because it's so hard to remember them all. I've forgotten all of the super freakin long ones over the course of my (2 year?) hiatus. They don't really seem to make that much of a difference to me time wise so I didn't bother relearning them. They're kind of tricky to get used to an incorporate for me. IMO they just complicate the solve too much to be worth it. They are really freakin cool though...hehehe

Yes, that's what I meant. I don't have trouble mirroring over to LU, I just have trouble executing it with the y and as RU.

RU algs are definitely terrible to memorize. Other subsets (RUD, RUL, etc) are so much easier to memorize. To be honest, I don't think RU has much of an advantage over RUD speed-wise either.

 

[Ranzha]

But, your way of doing it looks fine. You may have gotten lucky with your EO though, you could have been stuck with a pretty bad case if you had been unlucky. I'd do EO first if you want to stay constant. Also, doing EO first gives you the nice RU gen extension.

I knew the case would result in solved EO, so I did it =)
Perhaps another?

L2 U B2 R L' U' D2 F2 B2 D' F' U' L' F R2 U' R2 B U' D B2 F' R' F2 L'

2x2x2: x U' L U F2 x' [4]
2x2x3: y' R' U2 R' U' R [5/9]
Rouxblocks + DB + EO: L' U2 L U Rw U R' U' M U' M' U M [13/22]
L5E: U2 M' U2 M' U M' U2 M U M2 Dw2 [11/33]
L2C: R U2 R' U' R U' R' L' U2 L U L' U L [14/47]

Um, I improvised. xD
But was the beginning satisfactory? xD

-Cyrus

~Ranzha

 

[oll+phase+sync]

Yes, that's what I meant. I don't have trouble mirroring over to LU, I just have trouble executing it with the y and as RU.

RU algs are definitely terrible to memorize. Other subsets (RUD, RUL, etc) are so much easier to memorize. To be honest, I don't think RU has much of an advantage over RUD speed-wise either.

A little statistics of mine:

2 cases: idividual algs (9move case and 1-corner-to-flip case )

1 case: anti-reflect ( the original alg starts with y' already) R U2 R' U R U' R' U R U R'

1 case: optimal-reflect ( I believe the reflected version is the very best y' R' U R U R' U' R U R' U2 R)

4 cases: normal reflects


I know for this 4 cases there are better algs,
but reflection not only makes it easier to learn algs, the real simplyfication I feel with it is recognition (case to alg mapping to be precise)

 

[Jaycee]

I learned Petrus today! Though I haven't done a timed solve yet.

I'm still not the best blockbuilder, but I'm okay I guess as I've known how to use Roux for a while.

Here's an example solve of mine :

F2 D' L' U' L B F' L B U' D' R' D2 B L F2 D2 B U' L F2 D' F2 U2 R

U' L D' U' L' U' L'
R F' U2 L' U' L
F' U2 F2 R2 F' R' F' U' F
R2 U' R' U R U' R2' U R U' R' U R
U2 R U2 R' U' R U' R'
U' R' U' R y R2 u R' U R U' R u' R2 U2


No way my blocks would be that good in a speedsolve. I'll start learning COLL once I finish OLL (Hey, I'm still using CFOP as my main! At least now I have something to motivate me...) What do you Petrus pros think needs the most work?

Thanks,

~Jaycee

 

[Cyrus C.]

Your blocks look fine. Just practise using them in speedsolves I guess.

How well does executing bad edges on R and U work for you? I'd think doing a y and doing them on R and U would help recognition in a speedsolve.

 

[Jaycee]

As a matter of fact, the bad edges are the worst part of my speedsolves, because even though my blocks aren't very efficient, I can execute them fairly quickly. I pause for ~ 10 seconds to count the bad edges and position them correctly for fixing (I only know 2 ways to fix. F' U' F for 2 bad edges and F' U2 F2 R2 F' for 4.)

Average of 12: 43.03
1. 47.37 B2 U B' R B R F2 U' D' F' L D2 F2 D' R2 B' U D2 F' B2 U2 B D2 R2 F
2. (51.85) R2 U' F2 U D' F D L' U' D' L R D B2 D L' R U' D2 F D' F B' L2 F
3. 39.41 B D B2 L B D' B2 D' L B2 F L' R B2 U R D2 F2 U' D' R' B D L U
4. 40.83 F2 L2 B D' R F' L' F B' U F' R' B D2 B' F' L R' U' R2 B' U F' D' B
5. 46.35 L2 D' R U' F2 U' F' U2 F L R' U' F' L2 F R B L B' R U' D B R' L
6. 39.33 D' F2 B2 U2 D B' R D F B' D2 F2 B2 D2 F L' F2 R' B' F2 L F' D' F' L2
7. 29.67 R2 F' R D' R L F2 L' F2 U2 D2 B' L2 D L R' F R2 U' F2 L' R2 B2 F' R
8. 44.57 L R' U' B' U' D R2 B L2 B R D2 U' B2 L U' B2 D' R L' U2 R' L' F2 U'
9. (29.53) R D R D' F L' B' F' D2 F' L2 B D2 F' D R' U' F L R' D L2 D B F
10. 42.12 L' R' F2 R2 D B' L' D' U R B D2 F U' F2 B2 D' R F2 R' L2 D' B2 U' B'
11. 50.81 U2 B2 U' R2 F2 D F2 L2 B U L2 U2 F L' U D2 F U2 L' R' D' L' F B2 L'
12. 49.87 B2 L R' B L2 F' U R F' U' B' U' R2 B' F U' F2 R L' B2 U' F' D' L D

I just KNOW I can be sub-40 if I can improve bad edge recognition.

 

[Cyrus C.]

That is *not* how you should be doing bad edges. First of all, try doing a y rotation and putting the unsolved faces on F and U, I find this easier for recognition.

When I do bad edges, I only do 2 edges at a time. Just look for 2 unoriented edges, then orient them. While you're orienting those, you can look for the next pair. Use the triggers R U R', R U' R', L' U' L, L' U L', R U2 R', R' F2 R, L F' L, R F R', and other variations. Understand how these moves actually work. Fixing bad edges will become intuitive, and you won't have to do much setting up at all.

If you want to add a 4 edge trick, try M U M' instead.

 

[Jaycee]

Hmmm... I'll do another Ao12 and see what I get out of it, then come back here. Thanks for the tips!

~Jaycee