New ZZ Variant: ZZ - Profane Koala (ZZ-D with 4 algs)

[StachuK1992]

Very cool, but I think the process if we simplify the process like this:

Set up the same way (DBR and DFR in DBR, UBL).
Read the corner the the right of the yellow sticker, for UFL, UFR, and DBR/UBR (doesn't matter which, just keep it consistent)

For example,
Setup: R U2 R'
Stickers: ogr or ogb

Setup: tperm R U2 R'
Stickers: bgr or bgo

ogr = grb = rbo = bog
etc.

It's pretty easy to 'see' the DFR, LUF and BUR stickers without looking.
Really the DFR-corner sticker's obvious to see and you should only have to 'look around' for one of the other stickers on average.


I'm unfortunately still not satisfied with this method.

 

[Joël]

Very cool, but I think the process if we simplify the process like this:

...

I'm unfortunately still not satisfied with this method.

I agree, it's possible to recognize this by looking at just three stickers. You do have to identify which stickers, this is guided by the top face stickers.

I am quite satisfied with just knowing this method. I am not sure if I will train myself to do this, but I am really excited about learning how to go from 2x2x3 block into 2-gen easily. I am pretty sure this can become a reflex. For OH, the 2gen ending just feels so comfortable.

I have been trying EO-line, and I like it a lot, but I've also been playing around with this order of doing things:

1. Cross
2. 2 adjacent pairs
3. Petrus style edge orientations/edge control
4. Profane Koala to get into 2gen
5. Solve 2gen.

Obviously for step 5 it would be recommended to learn the 85 2GLL cases, if you are serious about this.

Con's: 2 Adjacent pairs instead of opposite ones needs planning. Sometimes you just have free pair, but you can't do anything with it.
Pro's: Easier inspection. Planning 2 pairs ahead forces better look ahead.

 

[Joël]

Very cool, but I think the process if we simplify the process like this:

Set up the same way (DBR and DFR in DBR, UBL).
Read the corner the the right of the yellow sticker, for UFL, UFR, and DBR/UBR (doesn't matter which, just keep it consistent)

For example,
Setup: R U2 R'
Stickers: ogr or ogb

Setup: tperm R U2 R'
Stickers: bgr or bgo

ogr = grb = rbo = bog
etc.

It's pretty easy to 'see' the DFR, LUF and BUR stickers without looking.
Really the DFR-corner sticker's obvious to see and you should only have to 'look around' for one of the other stickers on average.

I'm unfortunately still not satisfied with this method.

So yeah, to elaborate more on how to recognise by looking at three stickers, here is some more:

We read the stickers that are clockwise from the top layer sticker from the corners ULF URF DRB, and find a 'color code'.

When DBR is at DBR and DBF is at ULB, we are looking for the order: GRBO.

If we have no opposites next to each in the 'color code', it's either in good order or reverse order. So GRB would be good, but RBG would be bad. In the bad case, fix it with D R' U' R D'.

If there are two opposites in sequence, that means 2 corners would have to switch to restore the order. For example ORB. Orange and Red are opposites. We then check the order of the RB part. RB is good, so the Orange (ULF) and Green (URB) have to switch: L R U' R U L'

If we get the colors ORG, we see O and R are opposite, RG are not in the right order. So in that case URF and DRF have to switch to restore the order.

(Of course when DRB is at ULB and DRF is at DRB, the order you are looking for is reversed)

Etc etc.

It seems that a system like this could also be used with Roux btw. Since M doesn't effect CP, once we have CP done we can do M R and U moves without destroying CP.

1. First block
2. Profane Koala into 2 gen for corners
3. Solve however you like without breaking up the first block.

 

[StachuK1992]

[stuff]

Just a quick note while I have time - one can on-the-fly choose to check the clockwise OR counter-clockwise sticker relative to the U-layer color.
With this knowledge, I think only 2/9 of the time you'll have to do any 'thinking' when recognizing.

I plan to do further analysis tonight and make a short video demonstrating this method of recognition.

This might be nearing something acceptable for me, but I'm still hopeful that there's something better out there.


By the way, using it with Roux would be pretty neat. Do first block, PK, 1x2x2, CLS/WV, then do step 4 as usual.

Also, WV can be dramatically reduced for Roux; just some thoughts, need to explore further.

 

[Joël]

Just a quick note while I have time - one can on-the-fly choose to check the clockwise OR counter-clockwise sticker relative to the U-layer color.
With this knowledge, I think only 2/9 of the time you'll have to do any 'thinking' when recognizing.

I plan to do further analysis tonight and make a short video demonstrating this method of recognition.

This might be nearing something acceptable for me, but I'm still hopeful that there's something better out there.


By the way, using it with Roux would be pretty neat. Do first block, PK, 1x2x2, CLS/WV, then do step 4 as usual.

Also, WV can be dramatically reduced for Roux; just some thoughts, need to explore further.

I've been practicing this today, and I already notice an improvement in how fast I can recognize it. It's still slow, but like I said before I am sure with practice, it becomes a reflex. It is kind of a mind****, because the order you are looking for is different depending on the situation. For example when you have the 2x2x3 block on the right, and mirror all of this, the 'order' that is 'correct' is also mirrored (=reversed). A similar thing happens on when you want to be x2 neutral.

I still have to look into the case where you place the two bottom layer edges next to UL.

Another random thought: An alternative for L R U' R U L' would be L U' R2 U L', which does the same thing. Not very significant, but I thought I'd share it.

 

[TDM]

Another random thought: An alternative for L R U' R U L would be L U' R2 U L, which does the same thing. Not very significant, but I thought I'd share it.

Fixed.
Also, that's just the alg used in ZZ-porky v1 (which is exactly the same as Profane Koala but with different (and fewer and shorter and better for OH) algs) (the moves are inverted, but it's basically the same). So many people are thinking more about ZZ-d after this than after Porky v1 and v2. Why?

Also, my CP Recog:
1- Put the two corners in place and see which one is which as you do this, as usual.
2- For recognising the CP on the U face, see where the UFR and UFL corners are in relation to each other, and same with UBR and UBL.
2a (if UBL and UBR (or UFL and UFR) are adjacent to each other - this should happen in most solves)- If one pair is swapped, then it's an adjacent swap of those two. If neither are swapped, corners are correctly permuted. If both pairs are swapped, it's a diagonal swap.
2b (if UBL and UBR (or UFL and UFR) are opposite)- Do the same thing as before, except there'll always be one pair swapped. Here's how you can do it (although I don't usually do this because I know where the corners should go): put the two corners that go on UR/UL on UR/UL. Look at either of the front two corners, let's say UFR. If it is permuted then it's an adjacent swap for UFL and UBL. If it isn't, it's a swap of UFR and UBR.
I don't ever look at the same position when recognising (e.g. I mentioned UFR in the last example), but it can help to AUF and then do this to begin with.
Some other stuff you may know if you know a bit of COLL: if two adjacent corners have yellow on the same face (like U OCLL) and opposite colours on U, they are permuted correctly, and if they are the same colour they are swapped. Also, if the two adjacent corners with yellow on opposite faces (like the T COLL), they are permuted if their colour is the same and swapped if the colours are opposite. For both cases, if they are adjacent colours (e.g. red and blue), then you know there's an adjacent swap somewhere on the U face.
It looks difficult at first, but if you do a lot of solves with CP in (I do), it becomes really easy to recognise CP. I don't actually practise OH that much (I use a variant of ZZ with CP in for OH only), but I still don't take very long at all recognising CP.

 

[Joël]

Fixed.
Also, that's just the alg used in ZZ-porky v1 (which is exactly the same as Profane Koala but with different (and fewer and shorter and better for OH) algs) (the moves are inverted, but it's basically the same). So many people are thinking more about ZZ-d after this than after Porky v1 and v2. Why?

I my case, it's because I didn't cube/learn new stuff for a while, and now stumbled upon this thread before learning about the existence of Porky.

 

[a small kitten]

Profane Koala was conceived early 2011.

People are probably posting on this topic because it appears to be the most recent topic relating to ZZ-d.

 

[TDM]

I my case, it's because I didn't cube/learn new stuff for a while, and now stumbled upon this thread before learning about the existence of Porky.

True, but porky v2 was posted before this. Thinking about it, it's probably because v1 didn't have its own thread so less people saw it and v2 has too many algs/cases.

Profane Koala was conceived early 2011.

People are probably posting on this topic because it appears to be the most recent topic relating to ZZ-d.

I never said porky thought of his ideas before you did; I'm saying he posted them earlier but no one really noticed, and even after v2 it's still this thread that many people are posting on.

 

[a small kitten]

Is this a joke?

I'm not exactly up to date on the multitudes of ZZ variations, but isn't this just ZZ-porky?

It's just ZZ-porky with different (and more) algs.

So many people are thinking more about ZZ-d after this than after Porky v1 and v2. Why?

I got a strong impression that you were implying that Profane Koala was a derivative of ZZ-porky (judging from your above question and the quoted statements from others). I know you didn't actually state anything about who came up with what first.

I feel the answer to your "why" question is simple. These specific threads about minority methods seem to die if nothing radically awesome gets concretely developed. There are very few people that care about ZZ enough to discuss. The set of people that care about experimental extensions of ZZ is also very small. Naturally, a new thread popping up will revitalize the discussion for the people that lost track of older threads. Once these threads pop up, it's easily imaginable that people will read the thread and comment directly on the same thread. It's not like people will migrate back to older threads.

Obviously, I will feel momentarily miffed if I perceive that people are saying or implying that the thread's idea is derivative and/or a joke. Ultimately, though, I don't care that much about who thought of these ZZ-d variations first. It's more interesting to me that people keep revisiting the idea. Every time someone posts a new thread, the people that care return to discuss. At the same time, the thread attracts new people that missed old threads (like Joël).

 

[TDM]

I got a strong impression that you were implying that Profane Koala was a derivative of ZZ-porky (judging from your above question and the quoted statements from others). I know you didn't actually state anything about who came up with what first.

I'm not. When I said "It's just ZZ-porky with different (and more) algs", I was trying to say that they are the same thing, even if developed separately from each other. I was answering their question. Also, I knew asking "Why" was bad as I couldn't really word it right. I wasn't trying to say the ZZ-porky's should've had more people posting about them, I was asking why it is that it's this thread that people are interested in. And I answered myself last post.

I feel the answer to your "why" question is simple. These specific threads about minority methods seem to die if nothing radically awesome gets concretely developed. There are very few people that care about ZZ enough to discuss. The set of people that care about experimental extensions of ZZ is also very small. Naturally, a new thread popping up will revitalize the discussion for the people that lost track of older threads. Once these threads pop up, it's easily imaginable that people will read the thread and comment directly on the same thread. It's not like people will migrate back to older threads.

It's just porky v2 was posted first, then this shortly after, and this was the thread that more people saw. Why was it that nothing radically awesome was developed in the other thread but it was in this? And, as I said before, I answered myself when I said "Thinking about it, it's probably because v1 didn't have its own thread so less people saw it and v2 has too many algs/cases".

Obviously, I will feel momentarily miffed if I perceive that people are saying or implying that the thread's idea is derivative and/or a joke. Ultimately, though, I don't care that much about who thought of these ZZ-d variations first. It's more interesting to me that people keep revisiting the idea. Every time someone posts a new thread, the people that care return to discuss. At the same time, the thread attracts new people that missed old threads (like Joël).

I think the reason why porky asked if this was a joke was because of the similarity to his old method and because you have said it is "new". This is like my "why?" question: you meant it as new to the public, but when reading it it looks like you're saying it's a new idea that no one has thought of before.
Also, a question for Joël: why did you see this thread and not see the porky v2 thread?

 

[StachuK1992]

I'm not. When I said "It's just ZZ-porky with different (and more) algs", I was trying to say that they are the same thing, even if developed separately from each other. I was answering their question. Also, I knew asking "Why" was bad as I couldn't really word it right. I wasn't trying to say the ZZ-porky's should've had more people posting about them, I was asking why it is that it's this thread that people are interested in. And I answered myself last post.

It's just porky v2 was posted first, then this shortly after, and this was the thread that more people saw. Why was it that nothing radically awesome was developed in the other thread but it was in this? And, as I said before, I answered myself when I said "Thinking about it, it's probably because v1 didn't have its own thread so less people saw it and v2 has too many algs/cases".

I think the reason why porky asked if this was a joke was because of the similarity to his old method and because you have said it is "new". This is like my "why?" question: you meant it as new to the public, but when reading it it looks like you're saying it's a new idea that no one has thought of before.
Also, a question for Joël: why did you see this thread and not see the porky v2 thread?

The reason Porky's thread didn't do well is because he didn't present his method well.

His post makes me sleepy. That's nothing against his cubing skills; he's just not someone that can captivate my interest at all.

Further, people know Phil and I. We've both been active on the forums for years, and have made various contributions.
No offense to Porky, but I don't think he bears the same reputability on the forums, so it's likely that our efforts have some different weight to them because of this.

Anyway, let's talk about recognition. How is everyone doing?
I can recognize using my most recent system in about 1s. That's not too shabby. With practice, I could see a .6s recog.

 

[Joël]

Also, a question for Joël: why did you see this thread and not see the porky v2 thread?

I guess it's a coincidence. Thread was not on top, maybe not even on the first page, or maybe the title didn't interest me.

One thing that was attractive about this thread was the easy-to-understand video. EO-Line, 2x2x3... How do you get into 2gen..? Magic trick! I was intrigued by it.

I've been taking a look at the porky v2 thread, and I recognise the funny cycles (5cycles) and diagrams. When I worked out the cases for having the two corners at DRB and BLU, I started drawing diagrams just like that!

I agree with Phil Yu that it's a good thing that there are multiple threads where different approaches to similar steps are being discussed.

... How is everyone doing? ...

I am doing fine. I don't have a lot of time to work on this, because I work full time. But I've been practicing it slowly, and I it's getting better.

Some random thoughts:

1) You can also choose to look at 3 corners in this order: BRU ULF URF. This is better for TH, because you your thumb won't block the vision on the stickers you need to see. The BRU has two visable stickers, and that's enough. Judging by the number of stickers you'd have to look at, I think this is equivalent to recognizing a COLL case. Actually... Now that I think about it, it's more equivalent to recognizing what type of PLL you have (diagonal, adjacent or solved) by looking at two sides. Everybody already agrees that this is not "too hard to apply" in speedcubing.

2) I had another random thought today. A solver could choose to put the UL and UR corners in DF and DB positions during EO line, and finish with an LSE ending. Not sure if it is that much better than just ending with U, H or Z. I dont think it's that good, but I am just putting it out there.

3) I've also been thinking about other ways to recognize whether the 'order' is correct. When the DRF corner is at DRB and DRB is at BLU, the order kind of makes sense, because when I look from ULF to URF, if it's 2gen it will follow the same order as when I look at the F and R centers, which is also looking from left to right (relating this to the centers could also be helpful for those who want to be CN).

Now when the DRB corner is at DRB, this doesn't really work for me. One thing I just thought about, is that in this case, you can try to see (=imagine) what would happen if you would twist the ULF, URF and DRF corners in such a way that all top face stickers would come to the F face. If you imagine doing that, the stickers at ULF and URF would have to match, and the RFU and RDF stickers have to match too. I haven't tried this a lot yet, but I think this way of thinking about it could make it easier for me.

Edit:
(Similarly, when DRF is at DRB, you would have to imagine rotating the ULF, URF and DRF corners so that the top face stickers are on the L and R face. In that case ULF and URF would have to match, and FRU and FRD would have to match).

Edit2:
Yes! This seems to be much easier, it becomes exactly like determining the type of PLL when OLL is done.

Edit3:
Holy cow, it's much better this way. It's weird how thinking about something in a slightly different way can make all the difference in the world. I am making fewer mistakes now, and it's much faster.

 

[TDM]

ty both for giving good answers to my questions.

Anyway, let's talk about recognition. How is everyone doing?
I can recognize using my most recent system in about 1s. That's not too shabby. With practice, I could see a .6s recog.

Well I usually feel like I can recognise in under a second, but with me when I feel like I'm doing something fast it's usually slow.
My lookahead sucks, but has anyone tried to track a couple of corners during the end of the first F2L half? If you could follow at least three you could effectively have 0s recognition, but that's going to be difficult - with me even 1 is impossible, and that's when doing OH so I'm turning slower than 2H. But my lookahead is very bad and I don't know much about lookahead; I'm not too sure about whether 3 corners is too many for other people.

 

[porkynator]

I'm sorry if my "Is this a joke?" comment seemed rude, but...

I thought this was a joke considering that in these days there are many people posting new threads about new methods, and I thought Phil was aware that this had already been proposed.

...I was honestly thinking it was a possibility.

The reason Porky's thread didn't do well is because he didn't present his method well.

His post makes me sleepy. That's nothing against his cubing skills; he's just not someone that can captivate my interest at all.

I never thought about it, but it's true: I didn't care at all about making my post(s) interesting. I should work on that if I ever get to make a porky v3 approach to ZZ-d (as well as giving it a cool name)

Further, people know Phil and I. We've both been active on the forums for years, and have made various contributions.
No offense to Porky, but I don't think he bears the same reputability on the forums, so it's likely that our efforts have some different weight to them because of this.

I totally agree. When I read something on the forums I always try to forget about who has written it and to care only about what they have written, but it's just impossible. I'm not sure if this is a good or a bad thing.

[...]
it's one of the cases when different people discover the same thing in different moments
[...]

I thought this would have put an end to this "debate" (or whatever you want to call it).

Sorry for the OT, but I wanted to make it clear that I'm not like "I posted it first!", but more like "Cool, if people are still thinking about this method it means that it wasn't that bad after all!" (as you can see here, I was a little skeptical about it).

 

[StachuK1992]

You should be skeptical about it.
This is really neat, but I still think it's horrible. I'm nowhere near satisfied with this method until a better recognition method is found.
This is far too clumsy!

 

[Joël]

In order to make this a bit more understandable/readable, I've made some diagrams for the cases when the bottom layer corners are at DRB and ULB

Scenario 1
DRB corner is at DRB
DRF corner is at ULB

Recognition: Imagine rotating the corners ULF URF and RDB so that the top face sticker will come to the F face. Determine the relationship of the ULF URF sticker pair, and the RFU RFD sticker pair. For both pairs the relation can be "match", "opposite" or "adjacent", so we have 2x3=6 patterns. We can then figure out what needs to be done in order to get into 2GEN. In the diagram, this is shown by an arrow indicating that 2 pieces need to swap. But we will not actually swap those pieces. Instead, we do an algorithm that brings the corners into a 2GEN state.

The diagrams will show the orientations with the top face color already on F. This is just for clarity. In a real solve, the solver will need to imagine rotating the corners to that orientation, or find a different way to determine which stickers to compare.

Sticker pairs to compare when top face stickers are rotated to F:
ULF URF
RFU RFD

1.1) 2GEN

ULF and URF match
RFU and RFD match
ALG: None, it's 2GEN already!



1.2) ULF <-> URF

ULF and URF opposite
RFU and RFD adjacent
ALG: D' R U' R' D


1.3) FUR <-> FDR

ULF and URF adjacent
RFU and RFD opposite
ALG: D R' U R D'

1.4) RDF <-> RUB

ULF and URF match
RFU and RFD adjacent
ALG: D' R U2 R' D

1.5) ULF <-> URB

ULF and URF adjacent
RFU and RDB match
ALG: L U' R2 U L'


1.6) FLU <-> FDR

ULF and URF opposite
RFU and RFD opposite
ALG: D R' U' R D'



Scenario 2
DRB corner is at ULB
DRF corner is at DRB

Recognition: Imagine rotating the corners ULF URF and RDB so that the top face stickers will come to the L and R faces. Determine the relationship of the ULF URF sticker pair, and the FUR FDR sticker pair. For both pairs the relation can be "match", "opposite" or "adjacent", so we have 2x3=6 patterns. We can then figure out what needs to be done in order to get into 2GEN. In the diagram, this is shown by an arrow indicating that 2 pieces need to swap. But we will not actually swap those pieces. Instead, we do an algorithm that brings the corners into a 2GEN state.

The diagrams will show the orientations with the top face stickers already on L and R. This is just for clarity. In a real solve, the solver will need to imagine rotating the corners to that orientation, or find a different way to determine which stickers to compare.

Sticker pairs to compare when top face stickers are rotated to L and R:
ULF and URF
FUR and FDR

2.1) 2GEN

ULF and URF match
FUR and FDR match
ALG: None, it's 2GEN already!

2.2) ULF <-> URF

ULF and URF opposite
FUR and FDR adjacent
ALG: D' R U' R' D

2.3) FUR <-> FDR

ULF and URF adjacent
FUR and FDR opposite
ALG: D R' U R D'

2.4) RDF <-> RUB

ULF and URF match
FUR and FDR adjacent
ALG: D' R U2 R' D

2.5) ULF <-> URB

ULF and URF adjacent
FUR and FDR match
ALG: L U' R2 U L'

2.6) FLU <-> FDR

ULF and URF opposite
FUR and FDR opposite
ALG: D R' U' R D'

 

[Joël]

You should be skeptical about it.
This is really neat, but I still think it's horrible. I'm nowhere near satisfied with this method until a better recognition method is found.
This is far too clumsy!

This kind of reminds me of a quote from Ron van Bruchem about what he thought when he first read about F2L a looooong long time ago. "Too hard to apply", that's what he thought of it.

 

[JackJ]

Not sure if it's been mentioned but wouldn't it make more sense to learn the COLL cases? Surely 15+ move 2gen algs aren't faster.