[Help Thread] Roux Discussion and Help

[shadowslice e]

i would think the algs would be pretty bad, it's already pretty hard to solve the bottom corner, and now you have to also solve cmll at the same time. plus it doesn't save many moves in second block.

well, a misoriented corner would save 2 moves. I generated a couple of the algs and they are 12 moves 3-gen which are nice and fingerteickable I found

 

[dominugget]

On a sort of off topic, wouldn't it be possible to solve the UL and UR edges during CMLL with different Algs (I know memorizations would be insane)


Sent from my iPhone using Tapatalk

 

[Berd]

On a sort of off topic, wouldn't it be possible to solve the UL and UR edges during CMLL with different Algs (I know memorizations would be insane)


Sent from my iPhone using Tapatalk

What about EO?

 

[GuRoux]

well, a misoriented corner would save 2 moves. I generated a couple of the algs and they are 12 moves 3-gen which are nice and fingerteickable I found

2 moves isn't a lot and my average cmll movecount is 9.5. and what is the shortest alg possible? i can't come up with anything less than 10.

 

[shadowslice e]

On a sort of off topic, wouldn't it be possible to solve the UL and UR edges during CMLL with different Algs (I know memorizations would be insane)


Sent from my iPhone using Tapatalk

Well, it's possible to solve the whole cube in one alg if you learned all of them

2 moves isn't a lot and my average cmll movecount is 9.5. and what is the shortest alg possible? i can't come up with anything less than 10.

I don't know. It was just an idea. it'd definately not have a higher (at least not much) movecount in total and the algs would also have applications in other methods like ZZ or CFCE or FMC

 

[TDM]

Zz, roux and petrus all have lower move counts than cfop

It's not quite as simple as that.

You're correct in STM, but in SQTM (which is what each move really is in a speedsolve), CFOP doesn't do too badly. Much of the F2L, and OLL, is predominantly quarter turns. This makes speedsolving much easier. In ZZ, you have to do L2/R2 moves in F2L, and if you don't (e.g. R U R to insert a pair and the DR edge), you often have to regrip. In Roux, during the SB, there can be a lot of R2/r2/M2 moves which also take time and require a regrip sometimes. These can (and do) slow both methods down and isn't something I really see discussed.

(I've ignored U2s since not only do most methods have a lot of them anyway, but they're usually fast enough to be insignificant in comparison with R2 etc.)

 

[dominugget]

It's not quite as simple as that.

You're correct in STM, but in SQTM (which is what each move really is in a speedsolve), CFOP doesn't do too badly. Much of the F2L, and OLL, is predominantly quarter turns. This makes speedsolving much easier. In ZZ, you have to do L2/R2 moves in F2L, and if you don't (e.g. R U R to insert a pair and the DR edge), you often have to regrip. In Roux, during the SB, there can be a lot of R2/r2/M2 moves which also take time and require a regrip sometimes. These can (and do) slow both methods down and isn't something I really see discussed.

(I've ignored U2s since not only do most methods have a lot of them anyway, but they're usually fast enough to be insignificant in comparison with R2 etc.)

I agree completely.


Sent from my iPhone using Tapatalk

 

[shadowslice e]

In Roux, during the SB, there can be a lot of R2/r2/M2 moves which also take time and require a regrip sometimes. These can (and do) slow both methods down and isn't something I really see discussed.

That's interesting, I don't really do that many double moves in my SB because for some reason I instinctively think in QTM (although I do a fair amount of M2s but they have the same reasoning as the U/U2 in that they are more or less the same in terms of speed). In a similar way, OBTM doesn't come to naturally to me except when I'm orienting centres while inserting the last pair.

 

[TDM]

That's interesting, I don't really do that many double moves in my SB because for some reason I instinctively think in QTM (although I do a fair amount of M2s but they have the same reasoning as the U/U2 in that they are more or less the same in terms of speed). In a similar way, OBTM doesn't come to naturally to me except when I'm orienting centres while inserting the last pair.

Hmm, it depends on how you build blocks. I think doing fewer double moves could make your SB a bit less efficient in STM, but I'm not sure about this - my blocks aren't really very efficient.

However there are pretty much no R2s with CFOP, unless you use specific algs for a few cases (e.g. R2 U2 R' U' R U' R2).

 

[shadowslice e]

Hmm, it depends on how you build blocks. I think doing fewer double moves could make your SB a bit less efficient in STM, but I'm not sure about this - my blocks aren't really very efficient.

However there are pretty much no R2s with CFOP, unless you use specific algs for a few cases (e.g. R2 U2 R' U' R U' R2).

True, but lower in QTM might be better as it can be executed more quickly

 

[TDM]

True, but lower in QTM might be better as it can be executed more quickly

Would it be though if you had more moves in STM? I think it could still be about the same.

 

[shadowslice e]

Would it be though if you had more moves in STM? I think it could still be about the same.

I think an interesting way to measure it would be R, Rw and L moves counting as 1 but U2 and M2 also count as 1.

For future reference, I usually do SB in a similar way to CFOP with "cross" piece (DR), pair, insert, pair, insert CMLL etc. I find it nice because you can of FB essentially BLD, track DR, then lookahead is only 4 pieces

 

[GuRoux]

Would it be though if you had more moves in STM? I think it could still be about the same.

i think so, U2, M2, R2 are much more like 2 moves than one move no matter how you are fingertricking it.

 

[TDM]

I think an interesting way to measure it would be R, Rw and L moves counting as 1 but U2 and M2 also count as 1.

i think so, U2, M2, R2 are much more like 2 moves than one move no matter how you are fingertricking it.

Hmm, for me I'd say M2 is more like 2 moves, since I don't doubleflick with the same hand in Roux (until 4c), but U2 is one. I guess that depends on which fingertricks you use.

 

[GuRoux]

Hmm, for me I'd say M2 is more like 2 moves, since I don't doubleflick with the same hand in Roux (until 4c), but U2 is one. I guess that depends on which fingertricks you use.

double flick is as slow as doing two moves, even R2 is.

 

[shadowslice e]

double flick is as slow as doing two moves, even R2 is.

Not for me; my U2 and M2 are much faster than if I did (U U) or (M M) although my R2 is definately more like 2 moves once you take into account regrips etc.

But overall, I would say that minimising regrips and rotations should be the biggest priority in Roux F2B rather than sheer tps.

 

[GuRoux]

Not for me; my U2 and M2 are much faster than if I did (U U) or (M M) although my R2 is definately more like 2 moves once you take into account regrips etc.

But overall, I would say that minimising regrips and rotations should be the biggest priority in Roux F2B rather than sheer tps.

that's not what i mean, U2 and U R are about the same speed. like do (R2 U2)x6 that is 12 moves stm, and 24 moves qstm. it's probably around you (R U R' U')x6 speed.

 

[shadowslice e]

that's not what i mean, U2 and U R are about the same speed. like do (R2 U2)x6 that is 12 moves stm, and 24 moves qstm. it's probably around you (R U R' U')x6 speed.

Yes, I said R2 is loads slower due to regrips but for me R U2 is about the same speed as R U. In the same way M U is about the same speed as M2 U2 as these are regripless

 

[dominugget]

Any tips for look ahead.


Sent from my iPhone using Tapatalk

 

[NeilH]

Any tips for look ahead.


Sent from my iPhone using Tapatalk

plan out your entire first block and and look ahead for your 5 sb pieces while doing fb