Owen's Progress Thread | Finally some updates |

[AlphaCuber is awesome]

I have a really hard time believing that Kian has an 8.72 ao1000 OH. If you go to his WCA profile almost all his averages were 11's and then he got one 9.5 average. How can you say cfop was not built for oh when the TOP 2 OFFICIAL AVERAGES IN THE WORLD were with CFOP?!?!?! Maybe Kian is legit and does have an 8.72 ao1000 blah blah blah, but you cannot say cfop is not built for OH when is has the world records. Maybe Roux is better for OH than CFOP, but CFOP is certainly a great OH method as well, considering CFOP users have the top 2 averages in the world.

Corners first had the best average in the world in the 1980s but that doesn’t mean it was good for speedsolving

 

[ProStar]

CFOP average movecount:60-50
Roux average movecount: 40-55

CFOP TPS ceiling is much higher

CFOP: rotations
roux: no rotations

Rotations are around 2 a solve at max, which doesn't take that much time at all. Rotations are more important in TH than in OH. I will give you that Roux is better at that though

Max park best ao1000: 9.8 (the cfop UWR for OH)
kian mansour best ao1000: 8.72 (despite max park being faster 2h)

Did I not already tell you this isn't proof?

roux: mainly 2gen (RrU (this is kind of psuedo 2gen), MU)
cfop: mainly 3gen (for oh) (RUL, RUF, RUD)

Roux is not mostly 2gen, the solves are mostly RrUM. While this isn't bad, it is definitely not 2-gen
CFOP is RUF with maybe a couple D moves in PLL

roux first step: not good for OH
cfop first step: not good for OH

Not sure what this is for?

cfop OH pll: bad for OH ( i guess you could do stuff like CP control though)
roux cmll: good for OH

CFOP OH LL isn't bad
Roux CMLL is good for OH

Roux fb: average of 7 moves rotationless, can have 0 D or L moves
Cfop X-cross/cross, average of 8 moves, potential rotations ( L moves)

Roux FB can't be done RrU unless you rotate afterwards (this could be untrue, I'm not super familiar with this exact concept)
Cross doesn't have rotations and is mostly RUFL, which isn't bad for how short it is. And 8 moves is the limit, not the average. Even for me the average is around 6 moves, and pros do it in less on average

Roux SB average of 15 moves, rotationless
Cfop F2L average of 21 moves, definite rotations

F2L commonly doesn't have rotations, and at max it'll have around 2 (at pro levels)

Roux CMLL average of 11 moves, rotationless
cfop OLL average of 11 moves, rotationless usually

Name an OLL that has rotations?

roux LSE average of 13 moves, rotationless
cfop PLL average of 15 moves, rotationless usually

Again, name an OH PLL that has rotatations

honestly cfop was really not built for OH

Honestly you weren't built for debating



I can't speak to the average movecounts for each step, I don't remember each one, so I can't say if all of those you posted were true. Also, like I said, I'm not super familiar with Roux blockbuilding, so I could be wrong in saying FB can't be RrU without doing a y2 for SB

 

[AlphaCuber is awesome]

None of that evidence proves Roux is better. In fact, some of your "evidence" is completely untruthful.

It does and the only bit of his evidence that is slightly off is the cross move count

 

[Sub1Hour]

I would also like to mention that if roux automatically makes you great at OH then why does sean only have a 14 official avg? We all know that he is world-class for 2H but he is objectively worse at OH then Keaton who uses CFOP. Just because you know roux does not mean it makes you better than CFOP solvers at OH. For example, Keaton has a 7.28 official avg for 2H while Sean has a 5.98. By your logic, sean should crush Keaton in oh right? I mean, it's not like CFOP solvers can be faster than roux solvers at OH if they are worse then said roux solver at 2H. Once sean can get an 11 official avg and a 7 single, then we can talk about CFOP being an awful OH method.

 

[ProStar]

I would also like to mention that if roux automatically makes you great at OH then why does sean only have a 14 official avg? We all know that he is world-class for 2H but he is objectively worse at OH then Keaton who uses CFOP. Just because you know roux does not mean it makes you better than CFOP solvers at OH. For example, Keaton has a 7.28 official avg for 2H while Sean has a 5.98. By your logic, sean should crush Keaton in oh right? I mean, it's not like CFOP solvers can be faster than roux solvers at OH if they are worse then said roux solver at 2H. Once sean can get an 11 official avg and a 7 single, then we can talk about CFOP being an awful OH method.

That's another thing; mukerflap is saying that if you use Roux for OH you will automatically be good, which is even more ridiculous than claiming it's undoubtably better than CFOP.

 

[Nmile7300]

It does and the only bit of his evidence that is slightly off is the cross move count

All right. I'm sorry I got so heated. I have edited the post. I still stand by the fact that there is no way to prove a "best method" for either OH or 3x3.

 

[Sub1Hour]

That's another thing; mukerflap is saying that if you use Roux for OH you will automatically be good, which is even more ridiculous than claiming it's undoubtably better than CFOP.

Agreed. It's hard to make that case when the Roux poster child doesn't even break top wr400 OH with a wr3 3x3 avg.

 

[fun at the joy]

cfop OLL average of 11 moves, rotationless usually


cfop PLL average of 15 moves, rotationless usually


Beacuse of 0 rotations and better ergonomics you can reach a higher TPS with a lesser movecount

(according to the wiki)

CFOP OLL Average: 9.7 HTM
CFOP PLL Average: 11.8 HTM (Idk about that but definitely <13

CFOP OH TPS is higher. This is a fact, believe or not.

 

[ProStar]

Agreed. It's hard to make that case when the Roux poster child doesn't even break top wr400 OH with a wr3 3x3 avg.

I don't think times prove anything; Roux has a smaller sample size. I just think his claims are stupid, illogical, and ridiculous.

 

[Nmile7300]

That's another thing; mukerflap is saying that if you use Roux for OH you will automatically be good, which is even more ridiculous than claiming it's undoubtably better than CFOP.

Yeah. Even when someone switches from beginners method to something advanced, it still takes a while to get used to it.

 

[fun at the joy]

Did I not already tell you this isn't proof?

There is no proof for anyone's PBs, no proof for Max Parks Ao1000 PB.
You just have to believe them and I really don't understand why people don't believe Kian.
Remember he has competed in 21 rounds of OH since 2018, someone like Max Park has 71 Averages.
The argument "Why doesn't he have at least a sub-9.5 Ao5 or WR or whatever in comp then?" is just stupid.

 

[ProStar]

There is no proof for anyone's PBs, no proof for Max Parks Ao1000 PB.
You just have to believe them and I really don't understand why people don't believe Kian.
Remember he has competed in 21 rounds of OH since 2018, someone like Max Park has 71 Averages.
The argument "Why doesn't he have at least a sub-9.5 Ao5 or WR or whatever in comp then?" is just stupid.

I never said that it wasn't proof because it wasn't in comp or anything, I said it wasn't proof because one person's times doesn't show the potential of a method

 

[ProStar]

honestly cfop was really not built for OH, roux outclasses it for OH in ergonimics, movecounts, and rotations. Beacuse of 0 rotations and better ergonomics you can reach a higher TPS with a lesser movecount, lookahead doesnt really matter and is easy anyway, lookahead is solver specifics, feliks could easily lookahead with roux

Oh look you edited to your post.

Roux movecount is obviously better, but TPS ceiling is lower, making the speed not definitely better. You have to do rotations for FB unless you do LlU. You can't reach a higher TPS because lookahead is harder in Roux than in CFOP, and lookahead does matter for OH when you're going that fast. Feliks actually would have a lot of trouble with Roux since he doesn't use it, but anyway lookahead is(again) much harder with Roux than CFOP

 

[fun at the joy]

I never said that it wasn't proof because it wasn't in comp or anything, I said it wasn't proof because one person's times doesn't show the potential of a method

Oh okay
You meant that showing the best method's solvers PBs doesn't prove if one is better than the other method.
It's still worth it to show that the best CFOP OH solver's OH Ao1000 is 1s worse than the WB which was set with Roux though.

 

[Owen Morrison]

There is no proof for anyone's PBs, no proof for Max Parks Ao1000 PB.
You just have to believe them and I really don't understand why people don't believe Kian.
Remember he has competed in 21 rounds of OH since 2018, someone like Max Park has 71 Averages.
The argument "Why doesn't he have at least a sub-9.5 Ao5 or WR or whatever in comp then?" is just stupid.

It is just a little weird that he is apparently globally 8.7, but his best official average is 9.5, and most of his averages are not even sub 11.

 

[ProStar]

Oh okay
You meant that showing the best method's solvers PBs doesn't prove if one is better than the other method.
It's still worth it to show that the best CFOP OH solver's OH Ao1000 is 1s worse than the WB which was set with Roux though.

Yeah, but I would like to go again to my reference about CF in the 1980s:

The only thing remotely close to evidence you have submitted is Kian's times, but this isn't proof. In 1982, CFOP existed, but the fastest solvers used corners first. Does that mean that CF is the best method? No. And we can say that because we have objective proof that CF isn't as good as CFOP, while no one has proof that Roux is better than CFOP or vice versa.

It is just a little weird that he is apparently globally 8.7, but his best official average is 9.5, and most of his averages are not even sub 11.

Nerves are a big deal for him apparently, although I admit it's weird how he is globally way faster than his comp averages

 

[fun at the joy]

It is just a little weird that he is apparently globally 8.7, but his best official average is 9.5, and most of his averages are not even sub 11.

Max Park has a 6.17 Comp Style Ao10000. Does he get sub-6.2 Averages in comp? Mostly not.

 

[Sub1Hour]

Max Park has a 6.17 Comp Style Ao10000. Does he get sub-6.2 Averages in comp? Mostly not.

Mostly is still better than not at all.

 

[mukerflap]

CFOP TPS ceiling is much higher



Rotations are around 2 a solve at max, which doesn't take that much time at all. Rotations are more important in TH than in OH. I will give you that Roux is better at that though



Did I not already tell you this isn't proof?



Roux is not mostly 2gen, the solves are mostly RrUM. While this isn't bad, it is definitely not 2-gen
CFOP is RUF with maybe a couple D moves in PLL



Not sure what this is for?



CFOP OH LL isn't bad
Roux CMLL is good for OH



Roux FB can't be done RrU unless you rotate afterwards (this could be untrue, I'm not super familiar with this exact concept)
Cross doesn't have rotations and is mostly RUFL, which isn't bad for how short it is. And 8 moves is the limit, not the average. Even for me the average is around 6 moves, and pros do it in less on average



F2L commonly doesn't have rotations, and at max it'll have around 2 (at pro levels)



Name an OLL that has rotations?



Again, name an OH PLL that has rotatations



Honestly you weren't built for debating



I can't speak to the average movecounts for each step, I don't remember each one, so I can't say if all of those you posted were true. Also, like I said, I'm not super familiar with Roux blockbuilding, so I could be wrong in saying FB can't be RrU without doing a y2 for SB

1 for 2H

2 absolutely not, feliks had an average of 3.8 rotations in 5 example solves in his OH advanced example solves

3 Yes this can be proof because its unofficial and we are close to the limits

4 CFOP is RUFL (source: feliks OH advanced exmaple solves from cubeskills

5 for in case you were to mention roux FB isnt good

6 Notice how i said X cross / cross, roux FB can be done RUFBur which is a good moveset

7 cfop oh LL, some algs require rotations (according to feliks oh OLL 7/57 oll algs have rotations, 5/21 pll have rotations (im not dumb i didnt count auf rotation as rotation)), also the algs could be RUD/RUL which isnt as good as RUF cmll (except in 2 rud cmlls)

8 refer to 2, and feliks is pro level

9 7/57 according to cubeskills

10 5/21 according to cubeskills

11 not an argument

12 i am referring to kian mansours seminars which showcased the possible movecount limits, i could be wrong, you never need to do y2 in FB

 

[mukerflap]

Oh look you edited to your post.

Roux movecount is obviously better, but TPS ceiling is lower, making the speed not definitely better. You have to do rotations for FB unless you do LlU. You can't reach a higher TPS because lookahead is harder in Roux than in CFOP, and lookahead does matter for OH when you're going that fast. Feliks actually would have a lot of trouble with Roux since he doesn't use it, but anyway lookahead is(again) much harder with Roux than CFOP

proof TPS ceiling is lower in OH? FB literally requires 0 rotations of L moves ever. Roux lookahead is only a problem when going from SB to cmll, as planning FB+SS every solve or at least FB+DR is something kian already does, and lse lookahead is extremely easy, cmll to lse is very easy too