Recognizing G Perms

[SRV88bluescuber]

I have been solving the cube using the 4-look LL method for the past month and have started learning the rest of the algorithms for the PLL besides the 7 or so I've already learned for the 2-look PLL. I was wondering how to get faster at recognizing the PLL algorithm, especially the G-Perms. Any tips/tricks you use that could help me get faster? Is it mostly seeing where the corners and edges have to move to and then executing the algorithm? I guess for now the 2-look is easier since I do the corners first, then the edges, so I was just wondering how recognizing corners and edges at the same time is done, especially fairly quickly. Any help would be great!

Current best avg of 12: 0:52
PB: 0:43
Cross avg: 7-10 secs
F2L: ~20 secs
LL: 15-20 secs

 

[yurivish]

This is what I use: http://www.cubewhiz.com/pll.html

Though these look the trickiest to recognize, they are actually quite simple. I first AUF to solve the 1x1x2 block. Then, I rotate the cube such that the two corners that share the same color on the same face are on the left side. Then, based on whether the block is at the back, front, far part of the right, or close part of the right, I know whether to apply #14, #15, #16, or #17, respectively.

 

[PCwizCube]

This video will teach you how to recognize the 4 G permutations easily.
http://www.youtube.com/watch?v=VSmCZL0mVcw

By the way, you breakdown doesn't really make much sense. Even if you do the maximum times, it still doesn't reach 52 seconds.
10 + 20 + 20 = 50 :confused:

Try using this timer - It will get you a more accurate breakdown of your times:
http://rubiks.tw/timer/timer.php

 

[blah]

This is what Macky has to say in case you're looking for an alternative:

Most cubers tell them apart by matching the 1x1x2 block with the first two layers and then looking at the non-U sticker of the edge that completes the 1x2x2 block on U that contains the 1x1x2 block. This color will either be opposite the one next to it or adjacent.

But personally, I don't like either way. And I have a very good and logical reason to use my way of recognizing I match the 1x1x2 block just like everyone else, then look at the color of the third sticker that's on the same face as the "two continuous/joined colors", depending on whether it's an opposite color or not (I don't care what color it is if it's not), and depending on whether it's a 2-1 pattern or a 1-2 pattern (the "2" is the joined block, and the "1" is the other sticker), I decide which algorithm to use. There's usually no lag time for recognition because I mostly use the AUF time to figure out which algorithm it is. Why I think it's the most logical way to recognize: you're already looking at the 1x1x2 block, so just look at the sticker next to it that's on the same face to determine which case it is! Why waste time looking at other stickers on other faces?

And I can see that you're still quite new to PLL, because you said something about seeing which pieces need to go where...

Is it mostly seeing where the corners and edges have to move to and then executing the algorithm?

No one EVER does that. It's too slow. Other than the U perms, every other PLL is recognized by blocks of continuous/joined colors. So yeah.

And also a bit of side advice, at your current time, don't worry about PLLs yet, work on F2L first. All the PLLs you need now are the U's and A's, and probably H, Z and any PLL that swaps 2 corners.

Edit: @Rubik's Cube Fan, I just checked out that video link you provided, and I'm gonna be frank, it's not really good. I really don't see why anyone would want to look at stickers on other faces just for recognition? Why can't you just look at one sticker, on the same face, which saves so much time? Plus, looking at the sticker on the same face means you know what you're gonne get even before AUF! If you're looking at stickers on another face, you've gotta identify, AUF, then identify again, think about the lag time.

 

[DavidWoner]

I really don't see why anyone would want to look at stickers on other faces just for recognition? Why can't you just look at one sticker, on the same face, which saves so much time? Plus, looking at the sticker on the same face means you know what you're gonne get even before AUF! If you're looking at stickers on another face, you've gotta identify, AUF, then identify again, think about the lag time.

because the R perms also have just a single 1x1x2 block, and in order to easily determine whether or not it is an R or a G you have to look at other faces. the side with two matching corners stickers is usually what people use to differentiate between G's and R's, and if you already looked at that side you may as well use it in your recognition. and despite what you say, recognizing G's this way doesn't require a second ID after AUF.

and even your method(which took a while for me to figure out what you were talking about, but it seems to work) requires a way to differentiate between G's and R's, because on both R's, the single sticker you talk about is not opposite colored. the matching corner side is the key difference, and is quite easy to follow.


@SRV88bluescuber: work on your cross and F2l. get sub-5 cross

 

[blah]

@Vault312: Oh yeah, the R perms, forgot about those, sorry, my bad. But I never seem to have a problem telling G's from R's even before AUF, gut feeling I guess, I dunno, maybe it's just experience...?

 

[nitrocan]

G has only 1 corner and 1 edge after AUF, R has 2 edge 2 corners.

 

[DavidWoner]

G has only 1 corner and 1 edge after AUF, R has 2 edge 2 corners.

well yeah, but we're trying to avoid another recognition phase after AUFing.

 

[nitrocan]

then lets say, G's are recognised with the two same colored corners put on the left and then depending on the 1x1x2 block. The block and the two corners are not connected. But on the R's, the 1x1x2 block and the corner pair share a corner.

 

[DavidWoner]

then lets say, G's are recognised with the two same colored corners put on the left and then depending on the 1x1x2 block. The block and the two corners are not connected. But on the R's, the 1x1x2 block and the corner pair share a corner.

precisely.(before i added the period and this part, my message was "too short" to be posted. can you believe that?)

 

[blah]

Yeap I believe that. Because I just tried it Anti-spam thingy I guess? But no one spams on this forum anyway...

 

[brunson]

I wrote a program to help me recognize PLLs from only seeing two sides of the cube. If you go here: http://brunson.com/drillsergeant, then click "use only PLLs, no OLLs" and "Drill" you'll get a view of the cube after a PLL and before AUF. It's helped me quite a bit, though there are several angles of some PLLs that I don't recognize without seeing a third side yet.

As far as the original poster's question, I learned to recognize G perms originally simply because they didn't look like anything else. G was the last set of PLL algs I learned, so I could recognize everything else. To tell the G's apart I would AUF until the 1x1x2 pair was oriented and examine it with the cube held so the block is either in URB and RU or ULB and LU, then look at the other two cubies on UB. If they have two stickers of the same color, then I start the alg with R'U'R or LUL', if not then I do the other alg.

All that being said, you only need tricks like that in the beginning. Try everyone's suggestions and find one trick that works best for you. As you get more and more experienced you'll find that you think about the actual trick less and less and start recognizing the more on intuition and "feel". When that happens you may find you intuition is wrong and you'll end up doing two PLLs, but over time it'll get better and better until you want to start doing drills to recognize them from any angle.

Stick with it and practice, practice, practice. Everything comes in due time.

 

[SRV88bluescuber]

Thanks

Thanks for the suggestions/comments. I see value in looking at the 1X1X2 block and checking the left side to see if the colors are the same for the top corner cubies or if they are different. This method along with the brunson program has definitely helped me get a better feel for which G alg to perform. Obviously I know I need to get my F2L a lot faster to get my times down but for now I'm trying to move to a 1-look Pll instead of a 2-look while still using a 2-look OLL. Thanks again guys!