• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

Pineapples or no pineapples?


  • Total voters
    137

xyzzy

Member
Joined
Dec 24, 2015
Messages
2,877
Hawaiian Kociemba was originally created as an elaborate April Fool's joke by TheCubicle.com. In a video titled, "Michael Humuhumunukunukuapua'a's International Debut", a man was shown in Hawaii solving a 3x3 consistently under 5 seconds using a new method called Hawaiian Kociemba. Brayden Mossey, known online on the SpeedSolving forums as wir3sandfir3s, looked at reconstructions of some of the solves in the video and attempted to decipher his method. What he created was the current Hawaiian Kociemba method, except that originally the missing edge in the EOArrow was DR.
Just to add on a bit: the method itself wasn't named by Cubicle or Michael, and both it and its name are completely a creation of our forum denizens.

(For the record, I think "Hawaiian Kociemba" is a stupid name, and have thought that for the past four years. The biggest problem is that the name reflects neither the origins of the method nor the actual meat of it—the solves by "Michael" (actually done by Rowe) were completely set up scrambles that were fundamentally based on CFOP, and the method has little to do with the Kociemba algorithm per se. And if you were going to name a thing after someone (someone who even visits these forums occasionally!), it seems like a pretty good idea to at least ask them if they mind having their name attached to it, and I don't think that was ever done.)
 
Joined
Jun 8, 2019
Messages
757
Location
A Pineapple Orbiting Neptune
YouTube
Visit Channel
I like it! Couple notes/questions

1.) What's HKCLL?
2.) HK+ is just Yau except you don't solve 4th cross edge
3.) A couple times there is bad spacing(specifically on the About page)
4.) Saying "thanks to ProStar for writing this page" may look better down in the corner or something
1: HKCLL is just CMLL algs that preserve DF

2: I know HK+ is basically Yau, which is why I embedded the wiki page, which I edited to say that

3: spacing is bad on mobile, should I make it auto-align on both sides?

4: will do
 

brododragon

Member
Joined
Dec 9, 2019
Messages
2,274
Location
Here
Which would be DF. An L5E edge gets oriented in the process, but it doesn't matter which one
Ohhh I thought LL edges were oriented.
it seems like a pretty good idea to at least ask them if they mind having their name attached to it, and I don't think that was ever done.)
Good thing it's not their real name.
 
Last edited:
Joined
Sep 10, 2019
Messages
1,542
Update: I have started learning HKOLL, I know most of the 3 EO cases. I have realized that most of my pair solutions are sucky when edges aren’t oriented or are in DF, so I will watch Roux videos. I may end up maining the MS 3x3, and I plan to have my website fully completed with HKPLL and all within a month
But edges should always be oriented in HK
 
Joined
Jun 8, 2019
Messages
757
Location
A Pineapple Orbiting Neptune
YouTube
Visit Channel
I just discovered this in the original proposal
This will probably be THE MOST complex variation but can be VERY fast and significantly improve recognition. You will want to plan it in inspection. During the explanation, pretend the E slice is divided in half - left and right.
1. Solve a 2x2x3 block (preferably on the left) while leaving 2 edges disoriented. Be sure to track these edges.
2. Now it gets pretty tricky, and reliant on look ahead. Pre mute the bottom corner on the right, they do not need to be oriented. Pair the front one with one of the disoriented edges.
3. Figure out which adjacent corners needs to be swapped. Put them on the left, while making sure the other disoriented edge is at UR or UL. If corners need to swapped diagonally, do this step but use a conjugate to move the corners adjacent to each other.
4. Do either F' U F (if the edge is at UR) or F' U' F (if the edge is at UL) to reduce the cube to an entirely 2-gen state. Yes, all R U moves the entire way (if you want, I wouldn't recommend it as some 2-gem EHKPLLs are pretty long). Solve the other 2 pairs on the left (cube is in F2L-1 state) and use 1/7 Algs to orient corners whilst keeping permutation, then 1/7 EHKPLLs.
This sounds like a lot, but it can be executed pretty quickly. The reason for doing all this is when you finish F2L-1, you will only have to orient corners (only use 2 gen algorithms) and the corners will be solved. You will then be left with a EHKPLL, which there are only 7 of and are pretty easy to recognize and execute. This makes the last layer only use 14 algorithms, beating ZZ-R, DRASTICALLY improving recognition, and takes the cake for fewest number of algs for 2LLL.
IDK where he got 7 L5EP from, as there are 16 of them, but I just thought that this was interesting. Doesn't seem feasible to track and plan though. Also, I have learned all of the 3 EO cases, need to driil the algs to work on recollection, execution, and recognition
 

brododragon

Member
Joined
Dec 9, 2019
Messages
2,274
Location
Here
I just discovered this in the original proposal
This will probably be THE MOST complex variation but can be VERY fast and significantly improve recognition. You will want to plan it in inspection. During the explanation, pretend the E slice is divided in half - left and right.
1. Solve a 2x2x3 block (preferably on the left) while leaving 2 edges disoriented. Be sure to track these edges.
2. Now it gets pretty tricky, and reliant on look ahead. Pre mute the bottom corner on the right, they do not need to be oriented. Pair the front one with one of the disoriented edges.
3. Figure out which adjacent corners needs to be swapped. Put them on the left, while making sure the other disoriented edge is at UR or UL. If corners need to swapped diagonally, do this step but use a conjugate to move the corners adjacent to each other.
4. Do either F' U F (if the edge is at UR) or F' U' F (if the edge is at UL) to reduce the cube to an entirely 2-gen state. Yes, all R U moves the entire way (if you want, I wouldn't recommend it as some 2-gem EHKPLLs are pretty long). Solve the other 2 pairs on the left (cube is in F2L-1 state) and use 1/7 Algs to orient corners whilst keeping permutation, then 1/7 EHKPLLs.
This sounds like a lot, but it can be executed pretty quickly. The reason for doing all this is when you finish F2L-1, you will only have to orient corners (only use 2 gen algorithms) and the corners will be solved. You will then be left with a EHKPLL, which there are only 7 of and are pretty easy to recognize and execute. This makes the last layer only use 14 algorithms, beating ZZ-R, DRASTICALLY improving recognition, and takes the cake for fewest number of algs for 2LLL.
IDK where he got 7 L5EP from, as there are 16 of them, but I just thought that this was interesting. Doesn't seem feasible to track and plan though. Also, I have learned all of the 3 EO cases, need to driil the algs to work on recollection, execution, and recognition
Who proposed it?
 
Top