Big Cube BLD Discussion

Status
Not open for further replies.
3-style-corners is no problem, but a PLL-parity is: there you have to rotate one center (just count the U's of your parity-alg). You'll have to wait for solving a PLL-parity (and of course an OLL-parity) till the centers are solved (at best you would do this at the end of the solve - and cross your legs after memo to remember it inthe end!) because it swaps center-pieces on top. You can see it, if you do an "f" then the parity-alg and then "f'".

I memo/exe like you suggest it, but I'm no pro - pretty sure though that I saw a 6bld-solve of Cale Shoon where he exed corners first and PLL-parity last.
Oh yes parity! I actually figured it out how to deal it that. Solving the even number of corner letter pairs and then directly moving onto centers then wings. Then finally solve that odd corner target.
But I tend to use corner twist algs, which rotates the centers. How can I handle that?
 
I think that the really fast use kinda two comms to solve that.
But yeah: I know an alg for three twisted corners that rotates a center.

First idea: like with parity wait till the centers are solved (but can you keep that in memo?)
Second idea (of one who's avarage is above your 2nd solve time ...): I still use the alg of 1981, first I setup the twisted corners on the top-face, then do (R F' R' F) *2 for cw and (F' R F R')*2 for ccw (is it a "opposite" slege-hammer?). This is centersafe and surely good for two twisted corners, for three it takes a bit long - but they are rare.
And I'm very used to it so I don't want to change. When I started blindcubing the orientation was my worst thing. So I pre-orientated corners and edges first and then it was easier for me. I memoed the corner-orientation with numbers: 18 meant corner 1 cw, corner 8 ccw // 874 meant corners 4,7,8 ccw. Normally I took 2 or 3 corners together.
 
I think that the really fast use kinda two comms to solve that.
But yeah: I know an alg for three twisted corners that rotates a center.

First idea: like with parity wait till the centers are solved (but can you keep that in memo?)
Second idea (of one who's avarage is above your 2nd solve time ...): I still use the alg of 1981, first I setup the twisted corners on the top-face, then do (R F' R' F) *2 for cw and (F' R F R')*2 for ccw (is it a "opposite" slege-hammer?). This is centersafe and surely good for two twisted corners, for three it takes a bit long - but they are rare.
And I'm very used to it so I don't want to change. When I started blindcubing the orientation was my worst thing. So I pre-orientated corners and edges first and then it was easier for me. I memoed the corner-orientation with numbers: 18 meant corner 1 cw, corner 8 ccw // 874 meant corners 4,7,8 ccw. Normally I took 2 or 3 corners together.
I actually like the first idea. Solve all the regular corner targets and then after finishing the wings and centers go back to fix the twists/parity. Also I mainly use sune+lefty sune and mirrored sune+mirrored lefty sune and they don't rotate the centers and for most cases when there's 1 flipped corner either at the top or bottom, I set them up to sune twists. But if there two flipped, then I am using basic y perm to fix them as by doing 4 y-perms the top layer just rotates by 360° (I use this only in 4BLD and I know it's super inefficient) Also I couldn't understand the (R F' R' F) *2 and it's inverse notation. Could you please elaborate? Also which is a suitable buffer for x-centers? I currently use the Ubr as buffer.
 
4 y-perms? Couldn't you do once the inverse one?

(R F' R' F)*2 means that you do the things in brackets twice. Try this: R F' R' F R F' R' F U F' R F R' F' R F R' U' to get the idea for twisting.
Or this: R F' R' F R F' R' F U R F' R' F R F' R' F U R F' R' F R F' R' F U2
After 8 moves in the top-layer only the corner right-front is twisted and the U's are to bring in the next top-corner.
An hint: the first move (R or F') brings the "white" sticker up.
OK, if your buffer is in the bottom layer try the same idea with the sexy-move (R U R' U' twice - and the inverse): in the bottom layer only the corner right in front is twisted.
Or try this for top-face: R' D' R D R' D' R U R' D R D' R' D R U' (the last D of the first part cancels withe the start of the second part: only 16 moves).

Center-buffer: when I started 4bld I took Ubr as you, but I tried to find the system mostly on my own and can't tell, whether it is good. Using U2, maybe Ubl should be better for a right-hander because the inner-slice-moves are more often an r than an l. But maybe only in my system and I don't know whether this is really an advantage. But I think there are some threads here about it.
 
4 y-perms? Couldn't you do once the inverse one?

(R F' R' F)*2 means that you do the things in brackets twice. Try this: R F' R' F R F' R' F U F' R F R' F' R F R' U' to get the idea for twisting.
Or this: R F' R' F R F' R' F U R F' R' F R F' R' F U R F' R' F R F' R' F U2
After 8 moves in the top-layer only the corner right-front is twisted and the U's are to bring in the next top-corner.
An hint: the first move (R or F') brings the "white" sticker up.
OK, if your buffer is in the bottom layer try the same idea with the sexy-move (R U R' U' twice - and the inverse): in the bottom layer only the corner right in front is twisted.
Or try this for top-face: R' D' R D R' D' R U R' D R D' R' D R U' (the last D of the first part cancels withe the start of the second part: only 16 moves).

Center-buffer: when I started 4bld I took Ubr as you, but I tried to find the system mostly on my own and can't tell, whether it is good. Using U2, maybe Ubl should be better for a right-hander because the inner-slice-moves are more often an r than an l. But maybe only in my system and I don't know whether this is really an advantage. But I think there are some threads here about it.

Oh thanks! I guess I will switch to Ubl as the comms with this buffer is better for right handed solving.
 
Also do top 4BLDers use audio centers? Center targets seems to be a lot for me.
From what I understand, top BLD solvers use audio for the last set of centers, so the only set for 4BLD and + or X centers for 5BLD. 20ish targets is a lot, and you can get almost equally fast using visual/rooms or whatever you normally use. Maybe try memoing half the centers with audio and slowly work up to the full set.
 
From what I understand, top BLD solvers use audio for the last set of centers, so the only set for 4BLD and + or X centers for 5BLD. 20ish targets is a lot, and you can get almost equally fast using visual/rooms or whatever you normally use. Maybe try memoing half the centers with audio and slowly work up to the full set.
That's what I am doing now. For the x centers, the first 4 letter pair goes to a room, and then for the remaining I do audio. Full audio is currently impossible to me, but will try pushing it.
 
Sorry for the bump, but has anyone compiled a list of oblique comms. I have been following bestsiteever tables for 4 years now, and have not seen oblique comms being added up to the list.

Is anyone working on it?

I know it is pretty straightforward, after you learn U2 interchange method, with which you can do any center 3-cycle. Also the t-center algs can be extended out for obliques, just by doing different slice moves.

But it would be cool if there is a separate algsheet tab dedicated for obliques.
For example, it will help in pasting reconstructions (of oblique comms) of bigBLD attempt quite easily.
 
Sorry for the bump, but has anyone compiled a list of oblique comms. I have been following bestsiteever tables for 4 years now, and have not seen oblique comms being added up to the list.

Is anyone working on it?

I know it is pretty straightforward, after you learn U2 interchange method, with which you can do any center 3-cycle. Also the t-center algs can be extended out for obliques, just by doing different slice moves.

But it would be cool if there is a separate algsheet tab dedicated for obliques.
For example, it will help in pasting reconstructions (of oblique comms) of bigBLD attempt quite easily.
Yes, I believe both Graham’s and DRL’s doc have oblique tabs. I think DRL’s is pretty easy to configure for lettering scheme too. Let me know if you can’t find it, I can go digging through the ones I copied because I know I have one.
 
Yes, I believe both Graham’s and DRL’s doc have oblique tabs. I think DRL’s is pretty easy to configure for lettering scheme too. Let me know if you can’t find it, I can go digging through the ones I copied because I know I have one.
I could not find the link for oblique targets. It would be helpful if you could link it up.
 
Hello @bellarzios,

The paper pen technique has proved super useful to me while I started out with bigBLD in 2014.

I even made a video on how I practiced sighted execution and writing down the memo for a while, before I got my first 4BLD and 5BLD success.

 
I realize this is bumping a very old thread, but years ago I would have put this observation here.

I made an interesting mistake in a 5BLD solve today. I executed two adjacent images in the wrong order.

So, for example, I had the cycle:
(A BC DE FG)

And the next two commutators I executed were:
(A DE BC)

I realized this mistake before executing the third image, and was able to undo what I had done and execute again in the proper order.

During this same solve I also had a recall delay of one centers image. I moved on past that image and used my “pick-up” cycling approach of adding one final cycle at the end with floating buffer to account for the image that I purposefully skipped earlier, once I was later able to recall it.

Well, this got me thinking that there must be a pickup cycling approach for swapping two images.

Mapping it out made it clear that there is an approach to use that does not involve backtracking the incorrect execution.

Start with a scramble memo of:
(A BC DE FG)

Execute, incorrectly, (A DE BC)

The scrambled state itself would start out looking like:
BCDEFGA

Note that solving this state has memo: (A BC DE FG)

So from the starting state above, I execute (A DE) and end up with a scrambled state of:
FCDBEGA

Now execute, still incorrectly to the original memo, (A BC)

The scrambled state becomes:
DFCBEGA

To solve this state I would memo:
(A DB FG)

And this shows already our pickup cycle would be: (A DB)

Now go back to the original memo of:
(A BC DE FG)

If I execute with the images BC and DE swapped, I don’t have to undo those two cycles, then redo them in the proper order. After completing the second incorrect cycle, I can fix my error by executing: (A DB) which is the first letter from the second image followed by the first letter from the first image.

Executing this one pickup cycle is equivalent to undoing the two cycles and then redoing them again in the correct order!

The way to remember this is to orient yourself around the “correct” memo order.

If I know that I executed:
(A DE BC) when I actually was supposed to do (A BC DE) then I can complete the incorrect execution of (A DE BC) and follow by:
([Buffer] -> [First letter of 2nd image in correct memo order] -> [First letter of 1st image in correct memo order])

Note also that you can anchor off your “wrong” memo and pickup cycle with:
([Buffer] -> [First letter of 1st image in “wrong” memo order] -> [First letter of 2nd image in “wrong” memo order])

And now I can save myself three cycles by doing one pickup cycle to fix the mistake rather than four cycles to fix the mistake.

I’m fairly out of the loop on current BLD theory, so this may already be a well known trick used. If so, I’m happy to also discover it as a technique to use! If this is new, hopefully it will help someone!
 
Status
Not open for further replies.
Back
Top