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.

What do you mean by "actual method"? This question has multiple answers depending on that.

1. A method that exists

Some years ago, Tomas Rokicki coded a 4×4×4 solver that did reduction in ~1 step (it's a bit more complicated than that) and then solved it like a 3×3×3 optimally. As far as we know, this is the closest to optimal we've ever been (I think it averaged under 40 moves), but the solver is very resource-intensive.

2. A method that exists and you can use

csTimer etc. includes a 4×4×4 solver, solving reduction in three phases before moving on to a 3×3×3 solve. This is used to generate random-state scrambles. The solutions are typically around 45 moves long.

3. A method that exists and you can use without a computer

The current 4×4×4 FMC records are done, once again, with reduction. This reduction method is similar to what's used in the aforementioned three-phase reduction method:
1. Sort the centres into opposite colours. (As in, top and bottom faces have white and yellow, possibly mixed; left and right faces have orange and red, possibly mixed.) Also ensure that edge parity is solved at this point to avoid needing an OLL parity alg later.
2. Solve the centres using Uw2, Rw2, Fw2 and outer-layer turns, while pairing as many edges as possible.
3. If it's not already finished, finish edge pairing in a way that avoids PLL parity.
4. Do a 3×3×3 FMC solve.

4. A method that exists and you can use without a computer in real time

Guess what, it's reduction again. But this time you just do it the way you normally do: solve two opposite centres, solve the rest, do edge pairing, etc. Use Petrus or Heise or another low move count method for the final 3×3×3 step and you're set.

---

("lowest move count" is also not exactly determined but I'm assuming you mean an average over the different scrambles. This is the only interpretation that makes sense; if you compare two methods with similar average move counts, about half of the scrambles will favour method A and about half will favour method B, so it doesn't make sense to compare the move counts on individual scrambles.)

What do you mean by "actual method"? This question has multiple answers depending on that.

1. A method that exists

Some years ago, Tomas Rokicki coded a 4×4×4 solver that did reduction in ~1 step (it's a bit more complicated than that) and then solved it like a 3×3×3 optimally. As far as we know, this is the closest to optimal we've ever been (I think it averaged under 40 moves), but the solver is very resource-intensive.

2. A method that exists and you can use

csTimer etc. includes a 4×4×4 solver, solving reduction in three phases before moving on to a 3×3×3 solve. This is used to generate random-state scrambles. The solutions are typically around 45 moves long.

3. A method that exists and you can use without a computer

The current 4×4×4 FMC records are done, once again, with reduction. This reduction method is similar to what's used in the aforementioned three-phase reduction method:
1. Sort the centres into opposite colours. (As in, top and bottom faces have white and yellow, possibly mixed; left and right faces have orange and red, possibly mixed.) Also ensure that edge parity is solved at this point to avoid needing an OLL parity alg later.
2. Solve the centres using Uw2, Rw2, Fw2 and outer-layer turns, while pairing as many edges as possible.
3. If it's not already finished, finish edge pairing in a way that avoids PLL parity.
4. Do a 3×3×3 FMC solve.

4. A method that exists and you can use without a computer in real time

Guess what, it's reduction again. But this time you just do it the way you normally do: solve two opposite centres, solve the rest, do edge pairing, etc. Use Petrus or Heise or another low move count method for the final 3×3×3 step and you're set.

---

("lowest move count" is also not exactly determined but I'm assuming you mean an average over the different scrambles. This is the only interpretation that makes sense; if you compare two methods with similar average move counts, about half of the scrambles will favour method A and about half will favour method B, so it doesn't make sense to compare the move counts on individual scrambles.)

Ok so here's the big deal.I bought a MoYu MeiLong 4x4(Non-Magnetic) recently,and decided to learn 4x4 since I never have.Managed to learn it in one evening,and solved it fully before going to bed.Next day I wake up and decide I want to practice 4x4.Everything goes well at first.I solve all the centers,and do all edge pairing.But then,as i learned,I start solving it like a 3x3(I use beginners method btw).But when I insert every corner,they are flipped.I mean,the white parts or all on the same side,therefore I have a full white side.But the the colours(eg. Red and Blue) are "flipped".I tried doing the OLL Parity Algorithm for corners but it didn't work.I tried re-solving the whole cube but corners,when inserted,would become flipped again.I tried corner twisting them and insert the corner again but it still didn't work.I got really p*ssed so I took the cube apart fully.Spent a couple of hard-working days to put it back together and when I finally did put it back together,I did a re-solve.And literally again it happened.My first full solve goes well and I solve the whole cube without any flipped corners or problems.But on my second solve the corners become flipped again.My tutorial pamphlet does not include a single mention for this case.I did every single thing to try and fix it like last time but again,nothing worked.I am actually losing my mind as I was really enjoying learning 4x4 and I like to solve it...WHEN THE CORNERS ARENT FLIPPED.It is also worth mentioning that if I try to do OLL,it also happens so I have flipped corners on the top and the bottom.

Please help me guys,I am really annoyed.
See you all soon.

Ok so here's the big deal.I bought a MoYu MeiLong 4x4(Non-Magnetic) recently,and decided to learn 4x4 since I never have.Managed to learn it in one evening,and solved it fully before going to bed.Next day I wake up and decide I want to practice 4x4.Everything goes well at first.I solve all the centers,and do all edge pairing.But then,as i learned,I start solving it like a 3x3(I use beginners method btw).But when I insert every corner,they are flipped.I mean,the white parts or all on the same side,therefore I have a full white side.But the the colours(eg. Red and Blue) are "flipped".I tried doing the OLL Parity Algorithm for corners but it didn't work.I tried re-solving the whole cube but corners,when inserted,would become flipped again.I tried corner twisting them and insert the corner again but it still didn't work.I got really p*ssed so I took the cube apart fully.Spent a couple of hard-working days to put it back together and when I finally did put it back together,I did a re-solve.And literally again it happened.My first full solve goes well and I solve the whole cube without any flipped corners or problems.But on my second solve the corners become flipped again.My tutorial pamphlet does not include a single mention for this case.I did every single thing to try and fix it like last time but again,nothing worked.I am actually losing my mind as I was really enjoying learning 4x4 and I like to solve it...WHEN THE CORNERS ARENT FLIPPED.It is also worth mentioning that if I try to do OLL,it also happens so I have flipped corners on the top and the bottom.

Please help me guys,I am really annoyed.
See you all soon.

If you use Yau or Hoya, where starting centers dictate your cross colour, you certainly can(Feliks and Sebastian) but it is harder to learn than CN on 3x3, since that is required.

If you use Redux, go for it. Learn the colour scheme and do the fastest center first.

Yes, but your cube and your practice amount play a role too. I suggest doing slow solves but whenever you do a set of turns, do them fast. This helps a lot.

If you use Yau or Hoya, where starting centers dictate your cross colour, you certainly can(Feliks and Sebastian) but it is harder to learn than CN on 3x3, since that is required.

If you use Redux, go for it. Learn the colour scheme and do the fastest center first.

Sebastian Weyer does either white only or white/yellow on 3x3, so I don't think he's CN on 4x4. Looking at some reconstructions on speedcubedb, he always starts with white and yellow centers.

If you're not sub-45, I expect there are some tricks here and there that you don't know yet. There are some good walkthrough solves by Sebastian and Feliks where you could probably learn some.

Rw U2 x Rw U2 Rw U2 Rw' U2 Lw U2 Rw' U2 Rw U2 Rw' U2 Rw' and Rw' U2 Rw' U2 Lw U2 Rw' U2 Rw U2 x U2 Rw2 U2 Rw' U2 Rw U2

The first one is the default parity alg that everybody uses, which (if you didn't know) affects PLL parity. The second one, which I found from Hashtag Cuber's video, doesn't affect PLL parity in anyway. If you did the second one and you had PLL parity during OLL, it would still give you PLL parity, but if you did the first one, you would avoid PLL parity after doing OLL.

So now my question is, is there a way to know if you have PLL parity during OLL Parity so that you can do the "right" OLL parity alg to be able to always avoid PLL parity?

So now my question is, is there a way to know if you have PLL parity during OLL Parity so that you can do the "right" OLL parity alg to be able to always avoid PLL parity?

Not sure, but I don't think so, unless it's a very easy case (CP solved, 2 edges solved, other two swapped, or something like that), or you memorized every 1LLL case's orientation.

Rw U2 x Rw U2 Rw U2 Rw' U2 Lw U2 Rw' U2 Rw U2 Rw' U2 Rw' and Rw' U2 Rw' U2 Lw U2 Rw' U2 Rw U2 x U2 Rw2 U2 Rw' U2 Rw U2

The first one is the default parity alg that everybody uses, which (if you didn't know) affects PLL parity. The second one, which I found from Hashtag Cuber's video, doesn't affect PLL parity in anyway. If you did the second one and you had PLL parity during OLL, it would still give you PLL parity, but if you did the first one, you would avoid PLL parity after doing OLL.

So now my question is, is there a way to know if you have PLL parity during OLL Parity so that you can do the "right" OLL parity alg to be able to always avoid PLL parity?

I'm pretty sure you have it the wrong way. The first alg does not affect PLL parity while the second one does. Feliks has a video about this. tldr, you should only do it if it's easy to recognize. In my opinion, given your speed at 3x3 and 4x4, I don't think it's a thing you have to worry about too much. The time it would take you to recognize that you have double parity is probably not worth the tiny bit of time you'd save in almost all cases.