One-Answer 3x3x3 Question Thread

[Duncan Bannon]

Oh, so it has nothing to do with ao1000 or comp average? Thanks

 

[Hazel]

Oh, so it has nothing to do with ao1000 or comp average? Thanks

an Ao1000 can be a pretty good estimate of your global average if you feel it accurately represents your usual solve times

 

[CloudNight]

I am looking for something that is like COLL but can be used on all OLL cases, is there one?

 

[DGraciaRubik]

OLLCP

 

[CloudNight]

Thank you

 

[REDICUBER10]

I have learned
roux
cfop
begginers
waterman
hiese {commulatars}
petrus

and understanded ZB , ZZ , VLS ,etc

I CAN SOLVE IN 80 MOVES
BUT STILL I M SLOE JUST SUB 50 USING CFOP.

 

[greentgoatgal]

First off, pick one method and focus on it. If you want to do CFOP, some questions for ya

How much of OLL and PLL do you know?
What are your splits?
Do you know intuitive or algorithmic F2L?

Also, what kind of cube do you have?

 

[AlphaSheep]

How many solves have you done with each of those methods? Pick one and do 1000 solves using that method over the next month. It sounds like a lot, but it's really not one you get going. You should then see plenty of room for improvement.

 

[Hazel]

I have a program on my calculator that generates a random-move scramble for 3x3 of my desired length, since it's random-move as apposed to random-state (and can have things like B2 B'), what length should I have the scrambles be to make them good scrambles?

 

[xyzzy]

I have a program on my calculator that generates a random-move scramble for 3x3 of my desired length, since it's random-move as apposed to random-state (and can have things like B2 B'), what length should I have the scrambles be to make them good scrambles?

40 or so should be fine.

(It should be an easy exercise to modify it so that you don't get immediate move cancellations; say you have a list of all six faces, and you keep track of which was the last face used. Add a random number between 1 and 5, then mod 6, then look up which face that is in the list; repeat until you have the desired number of moves.)

 

[Hazel]

40 or so should be fine.

(It should be an easy exercise to modify it so that you don't get immediate move cancellations; say you have a list of all six faces, and you keep track of which was the last face used. Add a random number between 1 and 5, then mod 6, then look up which face that is in the list; repeat until you have the desired number of moves.)

Well I didn't write the program, just copied it down... I don't understand much programmy things

 

[Lapis]

Is there such a thing as zbll for olls? If there is, where can find good algs (or possibly generate them myself)? Particularly oll 47.

 

[shadowslice e]

Is there such a thing as zbll for olls? If there is, where can find good algs (or possibly generate them myself)? Particularly oll 47.

Do you mean OLLCP or 1LLL?

For the former there's quite a few sets (including one on algdb iirc). For the latter I don't believe there is a full set which is good for speedsolving though there are a few partial sets such as @Bindedsa 's

 

[Lapis]

Do you mean OLLCP or 1LLL?

For the former there's quite a few sets (including one on algdb iirc). For the latter I don't believe there is a full set which is good for speedsolving though there are a few partial sets such as @Bindedsa 's

I meant 1LL. Where exactly could I find Bindedsa's algs?

 

[TDM]

I meant 1LL. Where exactly could I find Bindedsa's algs?

He has a collection of his algorithm lists/tables here.

 

[Ulico]

VHLS + COLL + EPLL vs OLL + PLL
Which would be more efficient? Sorry if I messed something up, this is my first post. Thanks!

 

[Underwatercuber]

The question is in the title. Which would be more efficient? Sorry if I messed something up, this is my first post. Thanks!

I used to use zbls, coll and epll but I found it slower than just doing oll and pll and doing coll cases when I encountered them.

 

[Ulico]

Really! I had this idea because I thought it would be perfect for moving on to ZBLLs. Do you think ZBLLs aren't worth learning?

 

[Sue Doenim]

Welcome!
Yeah, it's best to use OLL/PLL, then learn COLL and use it when applicable. If you really like COLL, I'd look into ZZ. I think that it is probably how you'd transition to ZBLL, though. I'd still look into ZZ, I think it works better than ZBLS.

 

[Oatch]

Personally, I use COLL + EPLL for last layer every time. This comes down to a few reasons for me:
1. Since I use ZZ, I can use COLL every time, and it turns out that COLL + EPLL is more efficient than O(C)LL + PLL on average. Also, 1/12 chance of EPLL skip is always nice.
2. COLL recognition isn't that much slower imo, and also as you mentioned learning how to recognize corners is crucial for transitioning to learning ZBLL.
3. I don't like PLL. Recognising cases is harder for me, my algs are kinda lackluster for some cases and I can't be bothered relearning nicer ones at this stage, and I don't even know the full alg set anymore.

Again, the most significant factor of why COLL is most useful for me is because I use ZZ. COLL and EPLL is less moves than OLL + PLL, but once you add in VHLS, I imagine OLL + PLL will win out (you can check average movecount stats for LL sets on the wiki - I just can't recall the numbers off the top of my head). I think the best way to go about it is to, as previously mentioned, just execute COLL in cases where all of the edges are oriented, but don't put too much effort in otherwise if they aren't. Additionally, to increase the chance of this occurring, there's always edge control tricks that you can put in during your F2L to orient edges that will probably be less moves than using VHLS at the end, but harder to recognise. (Or you could just learn ZZ - there's some excellent resources online on it! And even if you don't use the method, the EO principles will still be highly useful for CFOP).