[Help Thread] Big Cubes Discussion

[Bh13]

Would a method that combines the OLL parity and a 3x3 OLL algorithm, and a PLL Parity with a 3x3 PLL algorithm, so that you could do the LL in two steps max instead of four save any time, or would the length and akward moves in the algorithms be too time consuming to be of any advantage?
Also, I'm having a bit of trouble with pairing up 3 dedges on the 4x4. Does anyone have any suggestions on what I should do to fix this?

 

[ben1996123]

Would a method that combines the OLL parity and a 3x3 OLL algorithm, and a PLL Parity with a 3x3 PLL algorithm, so that you could do the LL in two steps max instead of four save any time, or would the length and akward moves in the algorithms be too time consuming to be of any advantage?
Also, I'm having a bit of trouble with pairing up 3 dedges on the 4x4. Does anyone have any suggestions on what I should do to fix this?

It would probably suck, u R U R' u'

 

[Christopher Mowla]

Would a method that combines the OLL parity and a 3x3 OLL algorithm, and a PLL Parity with a 3x3 PLL algorithm, so that you could do the LL in two steps max instead of four save any time, or would the length and akward moves in the algorithms be too time consuming to be of any advantage?
Also, I'm having a bit of trouble with pairing up 3 dedges on the 4x4. Does anyone have any suggestions on what I should do to fix this?

See
http://www.speedsolving.com/forum/showthread.php?22568-4x4-OLL-Parity-Idea
http://www.math.leidenuniv.nl/~mfung/speedcubing/algs/4x4x4/
http://www.speedsolving.com/forum/showthread.php?24658-1LOLL-even-parity

There are discussions about this there and example algorithms that you could use.

 

[Christopher Mowla]

For those who use direct solving methods for 4x4x4,

For the diagonal 2-cycle case in 2 opposite dedges (in the last layer), this is one of many algorithms I found in the past for this case. It's the shortest algorithm in 3 out of the 4 common big cube move metrics (I have found 2 other solutions of the same length: all 3 are listed in the wiki). The only move metric it's longer in than other algorithms is STM, and only 1 more.

Uw Lw' Uw' l' Uw Lw Fw' Lw2 Uw' l' Uw Lw' L' Fw Uw' (16 SQTM, 15 STM, 18 qtm, 17 ftm)


I'm not a serious speedcuber, but I executed this algorithm a couple of days ago, and I realized that it might actually flow most of the way through (only the Lw' L' at the end needs to be dealt with). (Give it a chance, because it is so different to perform than algorithms you are used to.)

, and I was wondering if you could try to finger trick it and see if it is faster than all of the commonly practiced algorithms:
l' S2 U2 l U2 l' U2 r U2 r' F2 l B2 r z2 (21 SQTM, 14 STM, 30 qtm, 22 ftm)
r2 B2 U2 r U2 r2 B2 r' U2 r2 U2 B2 r' B2 (25 SQTM, 14 STM, 34 qtm, 20 ftm)
l2 B2 U2 l U2 l2 B2 l U2 l2 U2 B2 l B2 (25 SQTM, 14 STM, 34 qtm, 20 ftm):
l' U2 l U2 l U2 r' U2 l U2 l' U2 F2 l2 F2 r (25 SQTM, 16 STM, 34 qtm, 24 ftm)
Lw' U2 l U2 l U2 r' U2 l U2 l' U2 M' U2 l2 U2 Lw (26 SQTM, 17 STM, 34 qtm, 24 ftm)

And, although I mentioned this somewhere before, this case can be handled with a very intuitive algorithm.

Spoiler

Think of Stefan's PLL parity conjugate algorithm Rw2 F2 U2 r2 U2 F2 Rw2,

[1] Instead of r2, do r :
Rw2 F2 U2 r U2 F2 Rw2

[2] Swap the U and F centers by using inner layer turns. That is, l' U2 l r U2 r' or r U2 r' l' U2 l. BUT, first do the move F2 and then undo it at the end of the algorithm to have the final product:
Rw2 F2 U2 r U2 F2 Rw2 F2 l' U2 l r U2 r' F2

Inverting the bold inner layer quarter turn will solve the mirror case.

 

[Zeotor]

When solving a 4x4x4 using K4, are the last two edge pairs always solved with a predetermined algorithm? If not, how are they solved?
Thanks.

 

[qqwref]

For those who use direct solving methods for 4x4x4,

For the diagonal 2-cycle case in 2 opposite dedges (in the last layer), this is one of many algorithms I found in the past for this case. It's the shortest algorithm in 3 out of the 4 common big cube move metrics (I have found 2 other solutions of the same length: all 3 are listed in the wiki). The only move metric it's longer in than other algorithms is STM, and only 1 more.

Uw Lw' Uw' l' Uw Lw Fw' Lw2 Uw' l' Uw Lw' L' Fw Uw' (16 SQTM, 15 STM, 18 qtm, 17 ftm)

Oh wow o_0

I'm going to mirror it though:
Uw' Rw Uw r Uw' Rw' Fw Rw2 Uw r Uw' Rw R Fw' Uw

 

[PeelingStickers]

During 3-2-2-2-3 I commonly get a case during edge pairing where the edge I'm trying to find is already paired up with my "buffer edge", to eradicate this, I use an algorithm to solve them both however it ruins the flow of the solve and also requires fixing to get back on track, furthermore when I do get this case it usually happens more than once due to ruining the pairing. Any way to avoid this?

Without this I can get around 1:20-1:40 accounting for DP, however with it I end up with horrible times sometimes well over 2:00

 

[ben1996123]

They don't have to be, but to minimize the number of algorithms used, I believe Thom emphasizes to use one alg for every last two dedges case. However, to fix them in 2 algs, you can just execute a 3-cycle algorithm that affects only those two dedges to give you an alternate case that you are familiar with. You may also use a 2 dedge flip 3x3x3 alg to transform an ugly case into a familiar one (just depending on the case you have and depending on what specific cases you have memorized algorithms for).

It's not the fastest for speedsolving, but the last 2 edges cases are really all the same, they can all be solved with simple setup moves, a commutator and undoing the setup moves.

eg. this or this

 

[A Leman]

Are there any sources for solving the last 2 centers on a 7x7?

I have checked this entire thread, multiple SS searches, looked for WR reconstructions(which I understand why I can't find), etc, but I have stuck out. I try to pair with lines and also use 5x5 algs last two center algs, but the obliques are very difficult for me and I end up using a commutator or two to finish them off. As I said before, I am having trouble finding resources for this except for beginner methods that are quite bad.

What kind of strategy are the very fast cubers using?

 

[ben1996123]

Are there any sources for solving the last 2 centers on a 7x7?

I have checked this entire thread, multiple SS searches, looked for WR reconstructions(which I understand why I can't find), etc, but I have stuck out. I try to pair with lines and also use 5x5 algs last two center algs, but the obliques are very difficult for me and I end up using a commutator or two to finish them off. As I said before, I am having trouble finding resources for this except for beginner methods that are quite bad.

What kind of strategy are the very fast cubers using?

you usually do have to use a commutator on 7x7, and I think most people pair with lines too

 

[kcl]

you usually do have to use a commutator on 7x7, and I think most people pair with lines too

Sometimes, you can use blocks like on 5X5 can make two lines at once.

 

[SpicyOranges]

How do I improve at 7x7? I am stuck at around 9 minutes and it is SO SLOW...

 

[antoineccantin]

Are there any sources for solving the last 2 centers on a 7x7?

I have checked this entire thread, multiple SS searches, looked for WR reconstructions(which I understand why I can't find), etc, but I have stuck out. I try to pair with lines and also use 5x5 algs last two center algs, but the obliques are very difficult for me and I end up using a commutator or two to finish them off. As I said before, I am having trouble finding resources for this except for beginner methods that are quite bad.

What kind of strategy are the very fast cubers using?

I used to have lots of trouble with this as well. I found this tutorial very useful: http://www.youtube.com/watch?v=Xz4ei39dI3E

 

[Ross The Boss]

before i can contribute th othi thread, i must know, what are lines?

 

[JianhanC]

before i can contribute th othi thread, i must know, what are lines?

1x5 center bars

 

[Qthulu]

I'm thinking about buying a Shengshou 8x8 or 9x9, but I might have heard that the turning quality on the 9x9 is better and that the 8x8 has a catchy feeling. So, can anyone confirm this? I don't want to spend that much money for a single cube with these kind of problems. Also, are any of both worth their money regards of their solving time? Or is it just "too much" time you spend on solving this cube? (I solved a 7x7 from my friend twice, because I don't own one yet, and I needed about 1 hour for both solves)

By the way: < Insert obligatory first post welcome message here >

 

[ben1996123]

I'm thinking about buying a Shengshou 8x8 or 9x9, but I might have heard that the turning quality on the 9x9 is better and that the 8x8 has a catchy feeling. So, can anyone confirm this? I don't want to spend that much money for a single cube with these kind of problems. Also, are any of both worth their money regards of their solving time? Or is it just "too much" time you spend on solving this cube? (I solved a 7x7 from my friend twice, because I don't own one yet, and I needed about 1 hour for both solves)

By the way: < Insert obligatory first post welcome message here >

8x8 turns good but it isnt very smooth out of the box, but the 9x9 is apparently really good

How do I improve at 7x7? I am stuck at around 9 minutes and it is SO SLOW...

practise

 

[Dene]

How do I improve at 7x7? I am stuck at around 9 minutes and it is SO SLOW...

It's hard to give any better advice than "practise" when you don't provide any real information about your solves. For example, how long do you take for each step of centres, edges and 3x3? How do you approach building centres? What is your edges method?

If you really want to get better at big cubes I'd recommend committing yourself to doing an average of 5 on 7x7 every day. You'll get down to sub7 in no time, then you can start doing bigger averages to push yourself down further.

 

[illuminatiNPS]

Is 7x7 algorithms the same as 5x5??

Hello all,

Very new to this site. I just bought my first 7x7 and read that its the same as a 5x5. I use the BigCubes.com Pdf as a resource for the 5x5 http://www.kungfoomanchu.com/guides/andy-klise-5x5x5-guide-v3.pdf. My question is:

1) Do all the algorithms carry over to the 7x7??

2) instead of turning 2 layers, do i need to turn 3??

3) Do the last 2 centers present in the same way (corner to corner, or keyhole to keyhole)?


Any help would be appreciated as all the videos don't really list any instructions and haven't found a pdf like andy klises for the 7x7..

Can't wait to scramble and solve

 

[Ollie]

Yeah, all 5x5x5 algs carry over. You need to add a bit of intuition when it comes to the Niklas comms for centers. But scramble and try it out yourself = discovery learning is the best