Swagrid
Veteran
me when I spread misinformation on the internet
What's the best 3x3 method?
- Thread starter Xatu
- Start date
Xatu
Regular
You cannot, and i mean cannot, and I mean physically impossible to memorize 42 quintillion algs.
D
Deleted member 36445
Guest
you just said there is less thinking and more algs is better lol
he actually learned all the cases, don't you know?
ruffleduck
Veteran
I consider ZZ and CFOP to be equal. Read my paper in my signature for a detailed analysis.
ProStar
Legend
This has been discussed at length countless times, but since you clearly can't be bothered to research and/or read, here's it in very simplified form:
ZZ is a full pair behind CFOP and the only benefit is EO on the LL (no, the RULD F2L is not actually much of a benefit over top solvers' F2Ls). The only benefit of this is to open up feasible 1LLL in the form of ZBLL. The issue with that is that 1. it is so many algs, and that 2. top solvers already use a large chunk of ZBLL in CFOP, and most cases they don't learn is because they aren't that good. Couple that with the fact that 1LLL is now proven to be possible to learn and thus will start to be optimized and sets of it will be learned, and the time lost in EO is never saved later. You may save a fraction of a second in F2L and another fraction of a second in LL, but that is still less than you lose to EO.
And before you say "thats too many algs", I said *some sets* of 1LLL. Plus, you're advocating for ZZ which means you've thrown out alg feasibility entirely, since ZZ is only decent with nearly 500 ZBLL algs down perfectly.
Roux has potential to top CFOP, but ZZ is simply worse.
ProStar
Legend
Human augmentation is the future
Swagrid
Veteran
Do you not see how this is contradictory?
_The issues with zbll are that it is too large to use and that top people are already using it_
If top cfoppers can handle good ZBLL, as well as usually some OLLCP and EO control, I'm sure ZZers can handle some more ZBLL
ruffleduck
Veteran
Please read my paper in my sig. There are so many problems with your analysis, I am not even willing to list them out.
ProStar
Legend
That tends to happen when it gets condensed so much. Feel free to read the dozens of other posts I and many others have made on this
My mistake. I meant it was too many algs to be feasible to the “typical cuber”, but you are correct that the main point of my post was at the world-class level, not at a “lower” (not world class) level. I shouldn’t have included that
All of my other points still stand though. You’re a pair behind and get no real benefit from it.
Swagrid
Veteran
Do these small optimisations matter too much to the typical cuber? Is a pair a deal-breaker to an average solver? ZZ has been sub 7 without ZBLL so I'd argue that any average solver won't have their abilities inhibited by the method at all. And once you get slower than a certain point, everything is so loose, weird weak points and strength, stylistic differences. I average mid 9 and consider my solving to be HEAVILY unoptimised. At a 'typical' speed, I don't think objective optimizations as small as ZZ vs CFOP matter. When every step is being done sub optimally in multiple ways, I'd say let people choose their favourite so that their passion can stay and keep pushing them further.
ProStar
Legend
I have no issue with ZZ. I have said countless times that it is a perfectly viable method that can easily be sub-10 and far further (as has been proven). I also used ZZ for a time, although I was never very good with it. If someone enjoys ZZ and isn’t as interested by other methods I’d tell them to go for it. But if we’re talking at the world class level (at this point, sub-6 and rapidly approaching low 5), I don’t think ZZ can compete with CFOP or Roux. Is ZZ impossible to do well with? Absolutely not. There are many people who are far faster than I’ll ever be who use ZZ. But if we’re talking about the best method at the absolute highest level, ZZ isn’t that.
Again, I do apologize for previously bringing into the conversation a point about “average cubers”, as it was not the intended topic
ruffleduck
Veteran
Feel free to read my paper. I outline the benefits of solving EO early with as much objectivity as I can, and conclude that it effectively compensations for the 1 pair lag.
Perhaps this may seem rather egotistical, but I consider my analysis on ZZ and CFOP to be more valid and objective than the "dozens of other posts" I imagine you are referencing. I have achieved global sub 9 with both methods and have thought through and discussed comparison of these two methods extensively, as well as having broad experience with methods and method development.
Edit: I suspect you may not have interest in reading my paper (I probably should have concluded this earlier; this is an error on my part). I will briefly summarize:
ZZ is best thought of as EOcross, 4 ZZ F2L pairs, ZBLL
CFOP is best thought of as xcross, 2 CFOP F2L pairs, ZBLS, ZBLL (LS/1LLL is not nearly as good as ZBLS/ZBLL)
EOcross is equal to xcross.
First 2 ZZ pairs is equal to 2 CFOP pairs.
Last 2 ZZ pairs is equal to ZBLS.
ZBLL is obviously the same in both.
Last edited:
ProStar
Legend
This is where I disagree. Even with some EO/OLL control (not full ZBLS; top solvers don’t use 100% of ZBLL because not all cases are worth it. Sometimes OLL/PLL is faster, plus there are *some* 1LLL sets that are worth it), LS is faster than the last two pairs in ZZ.
ruffleduck
Veteran
Top solvers don't use 100% of ZBLL because they haven't learned it yet. They're actively working on it. It's 100% the future of all F2L-based methods. It's OP.
Absolutely not. ZBLS/ZBLL is miles ahead of OLL/PLL, and 1LLL is not worth it. ZBLS/ZBLL is as efficient as LS/1LLL while being a much smaller and thus more reliable alg set.
Please quote my paper instead of my posts if you wish to counter any points, for as you pointed out earlier, things "tend to happen when [ideas] get condensed so much." I am but summarizing my points here. My paper addresses pretty much everything I've seen you've ever said about ZZ vs CFOP.
Last edited:
