• 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.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

What approach is best for CFOP - High TPS or Efficiency

Efficiency Or Lookahed and TPS


  • Total voters
    61

lawofthecube

Member
Joined
Aug 14, 2020
Messages
141
I like Max's solves and think he is great solver and exciting and fun to watch but my heart is with Tymon with his "mad scientist" approach/solves and given I might be ok with LL TPS someday I feel like for me to eventually solve very efficiant and smoothly I'll have to know the more complicated shorter path to getting through first two layers...

I have finally admitted, which is hard for me to hold onto thoughts in a certain order, that I am not naturally gifted in cross so I might post a help with cross post as it's clearly what is holding me up and my doing xcross and seeing into F2L etc...

I might be a Tymon fan...lol

:cool:
 

Swagrid

Member
Joined
Jun 5, 2018
Messages
716
Location
Downtown Swagistan
YouTube
Visit Channel
Gotta agree with the masses here. Tymon style CFOP has a much higher skill ceiling. Using simple, standard solutions, and doing them very quickly is all good until you can't do them any quicker. Using tricks, influencing, and pseudo, well the sky is the limit for that as far as cfop is concerned. It can bring cfop down to be luckier and more efficient than it has ever been before.

As for the future of cubing, efficiency is absolutely the way. Imo it'll be a lot harder to keep turning faster and faster, so solvers like Ruihang that use incredible TPS to get their results will only be successful for so long. Eventually, it's not gonna be about how fast you do the moves, but how many moves you do. Which is why Roux is the best method.
 

Dan the Beginner

Premium Member
Joined
Jun 4, 2021
Messages
648
Location
Australia
Gotta agree with the masses here. Tymon style CFOP has a much higher skill ceiling. Using simple, standard solutions, and doing them very quickly is all good until you can't do them any quicker. Using tricks, influencing, and pseudo, well the sky is the limit for that as far as cfop is concerned. It can bring cfop down to be luckier and more efficient than it has ever been before.

As for the future of cubing, efficiency is absolutely the way. Imo it'll be a lot harder to keep turning faster and faster, so solvers like Ruihang that use incredible TPS to get their results will only be successful for so long. Eventually, it's not gonna be about how fast you do the moves, but how many moves you do. Which is why Roux is the best method.

Me too! To this beginner, the difference between averages of 5.2 sec, 5.4 sec or even 4.9 is not much at all. They are equally very, very impressive to me. I would be even more impressed and interested to watch them if it's done with innovative new tricks and fresh ideas.
 

Swagrid

Member
Joined
Jun 5, 2018
Messages
716
Location
Downtown Swagistan
YouTube
Visit Channel
Unpopular opinion: This might not be as one-sided as people think, because high-efficiency solutions are more difficult to inspect. Brute-force spamming may in the end be more consistent.
Counter arguments:

1) with how unreasonably optimized 3x3 is, I think taking the more reliable approach with the lower ceiling is probably something we should be leaving in the past. We're simply getting too good to be taking the easy route.

2) Just because something is difficult doesn't mean we should leave it. Tymon level inspection is hard, sure, but so were simple xcrosses when they were new, so is cross+1 to most people, ZBLL used to be considered super hard and now a load of people know it in full, even more without sunes, someone has learnt it oh, and tonnes of people use TUL.
 

OtterCuber

Member
Joined
Jul 26, 2021
Messages
638
Location
USA
YouTube
Visit Channel
Counter arguments:

1) with how unreasonably optimized 3x3 is, I think taking the more reliable approach with the lower ceiling is probably something we should be leaving in the past. We're simply getting too good to be taking the easy route.

2) Just because something is difficult doesn't mean we should leave it. Tymon level inspection is hard, sure, but so were simple xcrosses when they were new, so is cross+1 to most people, ZBLL used to be considered super hard and now a load of people know it in full, even more without sunes, someone has learnt it oh, and tonnes of people use TUL.
This sounds logical, but in practice the extra cognitive muscles needed to recall and execute large alg sets like ZBLL may incur a penalty that "easier" methods are immune to. It's for this reason that no one uses FMC methods to speedsolve. That's an extreme example, but I don't think we could push for xxcross or xxxcross for every inspection either. You can't optimize endlessly without sacrificing speed. There has to be an optimal point somewhere in between.
 
Joined
Dec 12, 2018
Messages
73
Gotta agree with the masses here. Tymon style CFOP has a much higher skill ceiling. Using simple, standard solutions, and doing them very quickly is all good until you can't do them any quicker. Using tricks, influencing, and pseudo, well the sky is the limit for that as far as cfop is concerned. It can bring cfop down to be luckier and more efficient than it has ever been before.

As for the future of cubing, efficiency is absolutely the way. Imo it'll be a lot harder to keep turning faster and faster, so solvers like Ruihang that use incredible TPS to get their results will only be successful for so long. Eventually, it's not gonna be about how fast you do the moves, but how many moves you do. Which is why Roux is the best method.

According to large-scale analysis of world-class solves reconstructions they can do 41% more outer layer moves than m-axis moves in the same amount of time. You just cannot reach that level of TPS with roux.
 

Swagrid

Member
Joined
Jun 5, 2018
Messages
716
Location
Downtown Swagistan
YouTube
Visit Channel
According to large-scale analysis of world-class solves reconstructions they can do 41% more outer layer moves than m-axis moves in the same amount of time. You just cannot reach that level of TPS with roux.
While the slightly lower TPS during LSE is a small drawback, we also have to acknowledge that Roux uses less moves overall, as well as M moves being a small part of that.
If we look at a fairly well sized cubeast session from a top Roux solver, LSE takes just 13 moves on average. Half of those being slices. So, 7.5 slice moves on average. With Roux being ~45 moves and top cfop being ~57-58, we can see roux being about 22.58% more efficient, although if we count slice moves as 1.5 of a move, it's still 20% more efficient.


Consider also that top CFOP solvers, are not top roux solvers, and as such, are far far less practiced in slice moves. I think a top rouxer wouldn't see such a vast difference between outer block and slice turns.
 

Blau

Member
Joined
Jun 3, 2022
Messages
15
Location
Earth
Max almost never uses ZBLL, while Tymon knows at least 60% of ZBLLs.
@Klaudiusz Szyprocinski that document was a really interesting read.
View attachment 19629
tymon actually knows around 69% of zblls (TULH pi)

A good comparison of max park style solving vs tymon style solving will be directly comparing their solutions and times for the same scrambles.
They competed head to head in red bull finals 2021 and max won 3-2.

reconstructions for your reference:
solve 1: (tymon 6.19)
tymon
max

solve 2: (tymon 5.66)
tymon
max

solve 3: (max 4.94)
max
tymon

solve 4: (max 4.83)
max
tymon

solve 5: (max 6.10)
max
tymon
 
Top