How Muchh Did PLL's Speed you up??

[noobcuber]

I have learnt 12/21 PLL but am having trouble recognizing cases. After you got over this bumb how much did learning all the PLL's speed you up

 

[mrbiggs]

It's really, really hard to say because while I learned PLLs I was also practicing and improving the other aspects of my solving.

I doubt that it was more than 2-3 seconds from 2 step PLL to one step. On the other hand, when I was doing up to four step PLL for the first month or so, I'm sure I was much slower than that. Then again, at that point my execution times were awful anyway.

 

[bundat]

I was doing "2-Look" PLL for G and N perms well into sub-30.
Going 1-Look helped me by only 2-3 seconds.

But it wasn't really "2-Look", because in reality, I still only used 1-Look, and knew the 2 PLLs to use immediately and was able to use them in succession smoothly (e.g. I'd quickly do an E-perm as "T-perm,U2,T-perm,U2" without taking a 2nd look at the cube).

More like "a very long PLL composed of 2 PLLs".

 

[ThePizzaGuy92]

I was doing "2-Look" PLL for G and N perms well into sub-30.
Going 1-Look helped me by only 2-3 seconds.

But it wasn't really "2-Look", because in reality, I still only used 1-Look, and knew the 2 PLLs to use immediately and was able to use them in succession smoothly (e.g. I'd quickly do an E-perm as "T-perm,U2,T-perm,U2" without taking a 2nd look at the cube).

More like "a very long PLL composed of 2 PLLs".

I used to do T-perm, and then a T-perm on the other side, i found that a little faster

PS, I'm absolutely DISGUSTED that you typed "learnt"...

 

[info]

It's a good question, probably asked many times before. I haven't learned all of the PLLs yet (I no doubt will at some point); but, I suspect that knowing more than about half of them isn't likely to speed things up much at all, especially in the first months. Though mrbiggs clear answer doesn't surprise me, eventually learning them all would seem to be the natural path for any serious cuber. The question is timing, of course. Also, early on, I picked off quite a few OLLs (which we know are usually learned later), simply because they were so simple and were easily identifiable as high frequency patterns. At this point, it seems to me, that some of them are more quickly identified and can possibly save as much or more time than some of the PLLs. I am a noob too though with significantly unremarkable times at this point.

 

[noobcuber]

why are you disgusted that I typed Learnt and also can you explain to me in detail what you do as a 2 look but norm one look> Im not familiar with thes terms

 

[Mike Hughey]

why are you disgusted that I typed Learnt

I'm not sure why he would care that you typed "learnt"; I assume it would be the preferred spelling in British English today if it weren't for the influence of American English. Apparently "learned" is now the preferred spelling in British English, but that's certainly a recent development.

 

[blgentry]

As mrbiggs said, it's hard to answer this concretely because as I learned PLLs, I was improving my F2L and cross also.

I think the biggest benefit to learning the PLLs is that it keeps you from having to mess with the U face orientation and speeds *overall* recognition time. Let me give you an example:

Hold your cube with your *top* color on top and run this:

U R2 F R U R U' R' F' R U2 R' U2 R

That's an R permutation. If you know the R, you just run the alg. But what if you don't know it? Then you have to do a corner cycle and then an edge cycle. But which corners? Well the front two are made and the back two need to be swapped. Crap. That means I need to move the U face until only one corner matches. One U turn matches the UBR corner. Then you have to rotate the cube (y2 in my case) and run an anti-clockwise corner cycle. *Then* you need to see your edges, potentially move the cube again (I have to do another y2) and run an anti-clockwise edge cycle.

All of that having to fiddle with the U face and do multiple steps of recognition for the two look PLL slows you down. At least it slowed me down.

In a way, learning the PLLs takes out several steps of mental analysis, and reduces it to aligning a few pieces, recognizing the perm, and executing it.

Brian.

 

[AvGalen]

I agree with your general conclusion blgentry (save moves, but mostly save analysis and cube rotations/U-adjustments), but I don't agree with the execution.

Analysis1: If the 2 front corners are the same, that means 2 wrong (opposite) corners must be on the other side (or all 4 corners are correct)
so I would perform U2 and a clockwise corner-3-cycle (A-Perm). This simple concept of correct/opposite corners saves a lot of time.

 

[noobcuber]

oh lol but I aint American, I'm a Kiwi

 

[fanwuq]

it doesn't really help when you are pausing to look for F2L during the same solve. When I learned them, it did not help at all. It just turned into an excuse to get worse at F2L. (before, at one time, I can get F2L done always sub-20 and average about 16 on it; now, my LL is a lot faster, but F2L sucks (17-25). It helps eventually, but no significant speeding up immediantly. A PLL is like 2-3 seconds each alg. So it would only decrease your times by that much.

 

[ScottKidder]

It's really, really hard to say because while I learned PLLs I was also practicing and improving the other aspects of my solving.

I doubt that it was more than 2-3 seconds from 2 step PLL to one step. On the other hand, when I was doing up to four step PLL for the first month or so, I'm sure I was much slower than that. Then again, at that point my execution times were awful anyway.

I've just finished learning all PLL and I would say I've dropped 10 seconds but it's MOST definately not all from PLL. It's probably mostly from the practice I've put into my cross and f2l, but there is obviously some improvement that came from learning all of them. I've found that I'm now starting to recognize some cases (especially Y and T) without even AUF so I can execute and then AUF which saves cube rotation, it all comes with practice. I think what helped me with that was doing a PLL to practice execution, and then doing another one, and see what one came up next and recognizing it and doing it (if I knew it)

 

[CorwinShiu]

PLL cut about 4-5 seconds for me. I learned all my PLL's when I was still averaging around 40, and it helped. I didn't have crazy fast executions ;P.

I would say f2l is still the main factor in terms of speed. I would say I could average around 22 with a beginner last layer but with my f2l. It's around 4 seconds slower, but that's about how much time you save learning OLL & PLL. You should learn them, just don't grind through them expecting significatly faster times. In fact, your times should go up first because of recoginition. I suggest going slow for memorizing and focus more of f2l.

 

[bundat]

I just noticed your new sig... yesterday you knew 12/21 PLLs, now you know 17/21, showing that you learned 5 in a day.

I'm not underestimating your ability or anything, but you should practice each new PLL you're learning until it's REALLY embedded into muscle memory and you can execute it quickly and fluidly without even thinking about it.

I know some people who can manage to become fluid and comfortable with new algos in a matter of minutes, but I doubt someone who keeps making "how can I get faster" threads can do 5 in a day. So I suggest you truly MASTER each PLL you know first before moving on.

 

[linkmaster03]

I agree. I learn two a day, practicing about two hours daily. Each day I run practice 30 minutes for the new PLLs, then I do all the old PLLs I learned for 30 more minutes.

 

[Crzyazn]

PLLs are useless without good execution speed and strength in the other steps....

I would guess that full PLL cut about 5 seconds off my average, but with about half i managed to drop from 55 sec to 38ish in about a week