2-Side PLL Recog +Hyperorientation= 1LLL Method

[jskyler91]

You numerated R U2 R' U' R U' R' wrong: 2 should be 3 and 3 should be 2.

Thanks!!! I will generate another one and replace it asap although I think you mean 3 and 4 are switched because a 2 is an edge. I did do the 3,4 corner wrong though

EDIT: Updated alg, let me know if you find others

 

[MWilson]

Thanks!!! I will generate another one and replace it asap although I think you mean 3 and 4 are switched because a 2 is an edge. I did do the 3,4 corner wrong though

EDIT: Updated alg, let me know if you find others

You didn't fix the 12 I mentioned when I posted the program. Also, you can use the program to generate the image for you, just click the cube graphic.

 

[MWilson]

Updated program to allow for practice recognizing the PLL. Now it's shift+click to save PNG, just click to randomize PLL.

Same link for DL: http://www.mediafire.com/?eaxofehqvc9w2w6

Edit: Updated with text field on left for rotation to have the color you want on top. For example, x will put green on top, x2 puts yellow on top, etc.

 

[IAmAPerson]

A bit late on this, I know, but I'm actually learning this to attempt to put it into real practice (competition-ready). However, I'm taking an easier approach by using VHLS to orient edges (it's surprisingly quick once you practice enough) and using Pure OCLL algs, so PLL recognition is almost just as quick as normal. What would you call a method that uses this? I think that "CFOP with VHLS, Pure OCLL, and 1LLL by 2-side PLL recog during OCLL recog" is a mouthful when someone asks you what method you use. I would say to call it SLL for "Skyler Last Layer," but SLL is a type of cancer. Anyway, I think if a solver put enough effort into this as to get them to solve as fast as normal OLL+PLL (which I'm going to make a serious attempt to do), the method would get attention.

I just think that there are so many good methods out there that don't get attention simply because people don't try hard enough to put it into normal practice.

 

[salova12]

A bit late on this, I know, but I'm actually learning this to attempt to put it into real practice (competition-ready). However, I'm taking an easier approach by using VHLS to orient edges (it's surprisingly quick once you practice enough) and using Pure OCLL algs, so PLL recognition is almost just as quick as normal. What would you call a method that uses this? I think that "CFOP with VHLS, Pure OCLL, and 1LLL by 2-side PLL recog during OCLL recog" is a mouthful when someone asks you what method you use. I would say to call it SLL for "Skyler Last Layer," but SLL is a type of cancer. Anyway, I think if a solver put enough effort into this as to get them to solve as fast as normal OLL+PLL (which I'm going to make a serious attempt to do), the method would get attention.

I just think that there are so many good methods out there that don't get attention simply because people don't try hard enough to put it into normal practice.

Hi man, How is your progress? Pure OCLL is so bad algs

 

[Humble Cuber]

Hi man, How is your progress? Pure OCLL is so bad algs

He's not gonna respond, he was last seen 2016

 

[PetraPine]

He's not gonna respond, he was last seen 2016

was that really necessary?

 

[Humble Cuber]

was that really necessary?

Idk I was just saying.

 

[Tao Yu]

I find it funny to look at the super convoluted 1LLL recognition systems that were proposed back in the day, considering the fact that nowadays, the best method is arguably "just look at the whole case like it's a painting".

 

[Humble Cuber]

I find it funny to look at the super convoluted 1LLL recognition systems that were proposed back in the day, considering the fact that nowadays, the best method is arguably "just look at the whole case like it's a painting".

Ikr

 

[Tao Yu]

Also I guess ROLL/JOLL is a method that pretty much achieves the goal of the method in the OP, and is actually considered useful at the top level. Which is interesting because I guess the advantage of ROLL/JOLL is that it doesn't try to be a perfect method. There are times where ROLL/JOLL is enough to fully predict PLL instantly, but where it doesn't it still gives you some useful information. This is much more practical than a method that can fully predict PLL 100% of the time, but requires a recognition pause.

I get the feeling that if more developments are made to CFOP, they'll probably be similar to this; the new methods or techniques would be "imperfect".

 

[salova12]

Also I guess ROLL/JOLL is a method that pretty much achieves the goal of the method in the OP, and is actually considered useful at the top level. Which is interesting because I guess the advantage of ROLL/JOLL is that it doesn't try to be a perfect method. There are times where ROLL/JOLL is enough to fully predict PLL instantly, but where it doesn't it still gives you some useful information. This is much more practical than a method that can fully predict PLL 100% of the time, but requires a recognition pause.

I get the feeling that if more developments are made to CFOP, they'll probably be similar to this; the new methods or techniques would be "imperfect".

I'm interested in PLL prediction. As far as I know, by using roll and joll it is not possible to predict exact pll. You can predict it gonna be one of the 2/2/2/3/4/8. To know what pll you will get you should do zbll/1lll recognition which I find really hard. This method is something different, but probably too hard as well

 

[Tao Yu]

I'm interested in PLL prediction. As far as I know, by using roll and joll it is not possible to predict exact pll. You can predict it gonna be one of the 2/2/2/3/4/8. To know what pll you will get you should do zbll/1lll recognition which I find really hard. This method is something different, but probably too hard as well

I think George Scholey or Jayden McNeill who told me at some point they were able to sometimes fully predict PLL using ROLL/JOLL. I think it may have been that they used some extra tricks on top of ROLL/JOLL though.

If you're interested in PLL prediction I think you need to be realistic about how much benefit that can really provide. PLL can be recognized very quickly already, so I would say even at best you could probably only shave off like 0.3-0.5 seconds of time by predicting PLL. Now that's not insignificant, but means that if you spend too much effort trying to predict PLL, you may actually spoil the advantage of knowing your PLL in advance. For example, it's pretty clear that the method described by IAmAPerson is slower than simply doing OLL/PLL because it adds a lot of moves (Pure OCLL algs are quite long, VHLS is inefficient). Even if it saves 0.5 recognition (which is extremely optimistic, I suspect it might even add time) the extra moves probably add about 1-1.5 seconds.

Again, this is why I think ROLL/JOLL was the approach that was adopted by top cubers. It is more practical to use a method that gives partial information for a small amount of effort, than a method that gives perfect information for a large amount of effort. The reason for this is simply that the information of knowing what PLL you are going to get is not guaranteed to save a lot of time.

 

[salova12]

I think George Scholey or Jayden McNeill who told me at some point they were able to sometimes fully predict PLL using ROLL/JOLL. I think it may have been that they used some extra tricks on top of ROLL/JOLL though.

If you're interested in PLL prediction I think you need to be realistic about how much benefit that can really provide. PLL can be recognized very quickly already, so I would say even at best you could probably only shave off like 0.3-0.5 seconds of time by predicting PLL. Now that's not insignificant, but means that if you spend too much effort trying to predict PLL, you may actually spoil the advantage of knowing your PLL in advance. For example, it's pretty clear that the method described by IAmAPerson is slower than simply doing OLL/PLL because it adds a lot of moves (Pure OCLL algs are quite long, VHLS is inefficient). Even if it saves 0.5 recognition (which is extremely optimistic, I suspect it might even add time) the extra moves probably add about 1-1.5 seconds.

Again, this is why I think ROLL/JOLL was the approach that was adopted by top cubers. It is more practical to use a method that gives partial information for a small amount of effort, than a method that gives perfect information for a large amount of effort. The reason for this is simply that the information of knowing what PLL you are going to get is not guaranteed to save a lot of time.

I didnt mean the Pure OLL thing. It is definetly not worth it, it is a lot slower actualy. I was talking about hyperorientation, but this is too hard as well, maybe for some easy caseses it's may be good. I thougt of combine ROLL and hyperopientation, but proboably it do nothing

 

[Tao Yu]

I didnt mean the Pure OLL thing. It is definetly not worth it, it is a lot slower actualy. I was talking about hyperorientation, but this is too hard as well, maybe for some easy caseses it's may be good. I thougt of combine ROLL and hyperopientation, but proboably it do nothing

I think hyperorientation is similarly flawed. I don't think it saves any recognition time as it simply moves the point where you do the recognition to an earlier time, where it's harder to do. The only way it's supposed to save time is through being able to transition from OLL to PLL instantly, but I think that's far from enough to make up for the harder recognition of PLL during OLL.

I haven't talked to Jay or George about this in a while, but I would guess that the way they are able to fully predict PLL is based on recognizing ROLL+JOLL and then noticing some other pattern about the case, such as some blocks or opposite colors. Maybe this could be thought of as a form of ROLL and Hyperorientation, but I think you would be right in saying that it is probably only good for easy cases (again, I think it's risky to invest too much effort into PLL prediction).