my direct solving method?

boo. http://rachmaninovian.webs.com/index.htm
help?

I just helped by moving this from beginners to how-to's, guides, etc

Thanks for posting after all that discussion earlier!

It looks kinda like a variation of the stadler method.

modified cage.

geez rachel why did you release it.

Neroflux said:

modified cage.

geez rachel why did you release it.

Click to expand...

dear grandson, I still love you, its okay.

mazei said:

It looks kinda like a variation of the stadler method.

Click to expand...

And stadler is a variation of my method that is a variation of Akimotos columns first. You may think K4 but K4 did not exist when I started to develop mine (spring 2006), it came onto the net a few monthts later so we must have developed during the same period. I would say also K4 is a variation of Akimotos method, he was first with puplishing this type of methods.

EDIT: After looking deep into the net I found it: http://rokumentai.akimoto3.com/rs12.htm

rachmaninovian I think is a bit of everything above and on top of that, Cage.

rachmaninovian sounds like a type of opening for chess.

And how are rachmaninovian and Neroflux related? I like to know if they really are grandson/grandmother.

er, rachmaninov is a composer-pianist's name, I was and still nuts about his piano playing/composing, therefore I used rachmaninovian as a username.

and...neroflux...argh long story
just for fun, sgcubers family tree

and the method...is called r4. woot. can't find a better name for it...any suggestions? of course for 5x5 it'll be r5...

another edit:TMOY has a very similar method. the different steps are step 5 and 6.

rachmaninovian said:

er, rachmaninov is a composer-pianist's name, I was and still nuts about his piano playing/composing, therefore I used rachmaninovian as a username.

and...neroflux...argh long story
just for fun, sgcubers family tree

and the method...is called r4. woot. can't find a better name for it...any suggestions? of course for 5x5 it'll be r5...

another edit:TMOY has a very similar method. the different steps are step 5 and 6.

Click to expand...

And I thought we had some weird off-topic topics on this forum

rachmaninovian said:

another edit:TMOY has a very similar method. the different steps are step 5 and 6.

Click to expand...

There are other small differences between our methods:
- (on 4^3 only) I do step 2 (corners) before step 1 (2 centers);
- I solve only 6 edges during step 3, and the 7th and 8th simultaneously during step 4;
- my step 7 is split in two: first I put all center pieces on the right pair of opposite sides, then I solve the pairs of opposite centers.
But I agree our methods are really very similar.

I really need help with the second step of r5: solving the 3x3 after solving the white and yellow centres. It is the slowest part of all!!! the F2L is excruciating slow...OLL/PLL is rather okay...but the F2L....

I see there is another difference between our methods
On 5^3 (and more generally on odd-sized cubes) during step 2 I still solve only corners. All middle edges are solved with the edge they belong to. (For the last 4 edges I first solve the wings, then for the middle edges I use my 3^3 corners first middle layer algs.)

TMOY said:

I see there is another difference between our methods
On 5^3 (and more generally on odd-sized cubes) during step 2 I still solve only corners. All middle edges are solved with the edge they belong to. (For the last 4 edges I first solve the wings, then for the middle edges I use my 3^3 corners first middle layer algs.)

Click to expand...

Then maybe the direct solving middle edges PLL's that I placed at this wiki page can come in handy.

Most are easy to find or already knowned but the Z-PLL is not that easy to find and much shorter than any other I know (most destroys some centres that you have to fix)

My last 4 edges are on the middle layer, not the top layer, sorry.
For "PLL parity", if you don't care about the centers you can go faster (Lw2 U2 l2 U2 Lw2).

rachmaninovian said:

I really need help with the second step of r5: solving the 3x3 after solving the white and yellow centres. It is the slowest part of all!!! the F2L is excruciating slow...OLL/PLL is rather okay...but the F2L....

Click to expand...

Switch to Roux, then you can use the empty middle layer to move edges without having to mind the centres.

Much of the slowness will still be there because it is harder to look ahead on a 5x5x5 than it is on a 3x3x3, it shouldn't but all the extra pices adds confusion.

Kenneth said:

rachmaninovian said:

I really need help with the second step of r5: solving the 3x3 after solving the white and yellow centres. It is the slowest part of all!!! the F2L is excruciating slow...OLL/PLL is rather okay...but the F2L....

Click to expand...

Switch to Roux, then you can use the empty middle layer to move edges without having to mind the centres.

Much of the slowness will still be there because it is harder to look ahead on a 5x5x5 than it is on a 3x3x3, it shouldn't but all the extra pices adds confusion.

Click to expand...

hehe, I tried that long ago, it's still pretty hopeless I guess....I'll just have to train brute force. maybe, I'll give the corner's first 3x3 method another try...

Kenneth said:

mazei said:

It looks kinda like a variation of the stadler method.

Click to expand...

And stadler is a variation of my method that is a variation of Akimotos columns first. You may think K4 but K4 did not exist when I started to develop mine (spring 2006), it came onto the net a few monthts later so we must have developed during the same period. I would say also K4 is a variation of Akimotos method, he was first with puplishing this type of methods.

EDIT: After looking deep into the net I found it: http://rokumentai.akimoto3.com/rs12.htm

rachmaninovian I think is a bit of everything above and on top of that, Cage.

Click to expand...

Here are a couple of Gilles Roux's ideas posted on the Yahoo group:

2004:
The method I found looks like this: 1) L and R sides. 2) M-edges. 3)
M-centers. It needs many slice moves and I'm still slow with it
(2:10). It requires about 140-150 moves.

2005:
My method is different, it looks a bit like my 3x3x3 technique:
- 1x3x4 block on a side
- 1x3x4 block on the opposite side
- U corners
- 12 remaining edges
- last 4 centers
I can't reach good times (about 1:50 on average), perhaps because I
rarely practice.

Thom uses Gilles' 3x3x3 method so K4 may have been influenced by that 2005 post. It seems like Gilles came up with those steps himself basing them on his 3x3x3 method. Though he sometimes asked Akimoto and others questions about 4x4 move counts, he probably wasn't influenced by Akimoto's method.

Yet another direct solving structure portrayed as a new method.

This method/structure is not new, it has been available since 1997 at http://www.helm.lu/cube/solutions/revenge/index.htm where it is simply named "Rubik's Revenge Solution".

The K4 site was made because I belived it to be a very efficient and unknown direct solving structure for 4x4x4, and at the time direct solving pages for speedcubing didn't really exist.

Regarding your email...

Hi Thom,
May I ask for your permission to use your html code temporarily for your k4 method for my own method?
I will come up with a new interface after I've finished learning html coding.

Click to expand...

Would you have sent that email had Quadresence not prompted you? I belive the correct thing to do would have been to ask me *before* you went ahead and did it. I see that since this event, you started using all your own pictures and changed the credits page a bit. However, I am still not credited for anything.

r' U2 l F2 l' F2 r2 U2 r U2 r' U2 F2 r2 F2, aka OLL parity

Click to expand...

I also noticed that you didn't actually read my site while copying it...

Kenneth, Athefre; K4 was developed as a cage alternative in 2004/2005, I didn't even know columns at the time.

Sorry for everything.

I did say that it is an advancement of Denny Dedmore's solution, which is the same one at helm.lu. Just to let you check: http://www.alchemistmatt.com/cube/revenge/index.htm

I never said that it was a new method. I can't really call it modified cage, because I just did not feel right to call it modified cage, after a chat with per.
The reason why I released my version of it is because some of my friends wanted to learn a more advanced solution to Denny's method. If you had noticed, I did post a topic on centres last solving on this forum very long ago. Again, I did emphasize that it is not my method.

I've just put the website...offline for the moment. Sorry for being rude, Thom. I'm working on a new interface. I was using your html as a model. Sorry for being ignorant about such stuff. I sincerely apologize for my mistakes and ignorance.

about the "OLL" parity...I did see you mention it as a 2-cycle but I got used to calling it OLL parity. It just feels weird for me. =P In fact I have meticulously read everything in your site.

Once again, I am very very sorry for what has happened. After all, as human and finite beings, we are bound to make mistakes. Perhaps the mistake I made was large and unforgivable. Please do forgive me for my mistakes. In any case, I know I am wrong, and I know that I owe you an apology. I'm just really really really really sorry.

The concept of cage methods is definitely not new. I was already using a similar method for my very first 4^3 solves in 1982 (Don't remember the details, sorry. But at this time I was not aiming for speed, I was just trying to solve it). It's simply the most natural extension to big cubes of the 3^3 corners first methods..