Zbroux application

[u Cube]

Is it worth it to learn Zbroux? I'm not talking about using it all the time as I know that it's definately not optimal in most cases and applying zb recognition into the equation makes things even worse. But for especially bad CMLL+EO cases where the alternate alg to for instance skip six flip just isn't worth it, would it be good to learn some zbroux for those very specific applications? Please let me know your input and why or why not it would be good, because I think for certain cmlleo cases it could be decent.

 

[mukerflap]

it sucks and i dont use alt CMLL algs

 

[Tao Yu]

EODFDB also has six flip cases, so I don't think you really avoid any bad cases using this. I think that learning 6-flip EOLR is probably the best option.

It's also not really possible to just learn some ZBroux. ZBroux needs full ZBLL to average around the same movecount as CMLL + LSE. If you don't use full ZBLL, ZBroux is definitely slower than getting all the worst cases for CMLL and LSE.

 

[Nir1213]

EODFDB also has six flip cases, so I don't think you really avoid any bad cases using this. I think that learning 6-flip EOLR is probably the best option.

It's also not really possible to just learn some ZBroux. ZBroux needs full ZBLL to average around the same movecount as CMLL + LSE. If you don't use full ZBLL, ZBroux is definitely slower than getting all the worst cases for CMLL and LSE.

so its only possible for a god like you oof

 

[Kaneki Uchiha]

even then it isnt better than l10p

 

[abunickabhi]

I do not think it is worth it. It is better to be method neutral and continue with freeFOP or CFOP, if the L10P looks like a bad case, if you started your solve with Roux.

 

[Tao Yu]

so its only possible for a god like you oof

You'd need to be way better than me in fact. My Roux L10P is miles faster than doing EODFDB+ZBLL

 

[Nir1213]

You'd need to be way better than me in fact. My Roux L10P is miles faster than doing EODFDB+ZBLL

oh wow

 

[Kaneki Uchiha]

l10p can easily be done sub 4 with practice. zbrouux just isnt good enough to be viable

 

[Nir1213]

l10p can easily be done sub 4 with practice. zbrouux just isnt good enough to be viable

yep when you mash two methods together it doesnt always mix well.

 

[mukerflap]

l10p can easily be done sub 4 with practice. zbrouux just isnt good enough to be viable

sub 3

 

[Kaneki Uchiha]

sub 3

Easily

 

[mukerflap]

Easily

yes l10p is easy to improve

 

[efattah]

I find it hilarious that people can make blanket statements like 'zbroux isn't good enough to be viable.' People said the same thing about Roux in the early days. Until someone actually masters ZBroux (which no one has), there is only speculation, which history has proven is usually wrong, when it comes to cubing.

 

[Kaneki Uchiha]

people can sub 2.5 l10p 2h you're telling me someone can solve dfdb and then recog the zbll and exec that fast?

 

[efattah]

Of course, and even more is possible. If you solve DFDB in 0.8, ZBLL has zero recognition time because you can see it during DFDB, then you execute ZBLL in 1.7, and there is your 2.5 seconds.
If you are talking about doing L10P in classic Roux style in 2.5 seconds, how much time are you allowing, to recognize the corners case and recognize which edges are disoriented, to select the CMLL that avoids a 6-flip? I'm just curious how much time you are allowing for recognition there, since CMLL+EO recognition time is considered one of the worst aspects of Roux.

 

[Athefre]

I've also always found it interesting that the community largely dismisses this method combination. The same goes for the very common F2B -> DF+DB -> OLL+PLL idea. Each time these methods are mentioned, someone says "terrible method". It's absolutely possible that these method combinations can't be as fast as the individual methods of which they are made. But to call them terrible is going a bit far. It's possible they they can be just as fast. And, if not, the time difference will be minimal. Not the 4-5 second difference that is implied by the word "terrible".

No one really knows how much of a time difference there is. There are already several popular, and highly regarded, ways of completing F2L. Is F2B -> DF+DB another good option? No one yet really knows for sure. F2B is a proven set of steps. There are also many good techniques for solving DF+DB. Maybe they actually combine well or maybe they don't. It would just take a few dedicated solvers to show either way.