• 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 35,000+ people from around the world today!

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

The Layer by Layer Podcast

BenChristman1

Member
Joined
Oct 26, 2019
Messages
2,756
Location
The Land of 11,842 Lakes
WCA
2019CHRI11
YouTube
Visit Channel
I don't know much (actually anything) about ZZ, so I can't say much about that, but like Kit, I sometimes take a FreeFOP/Petrus approach to certain scrambles. For example, if i see a really easy 2x2x2 block in inspection, I will plan out that and then move on to finishing the cross, then doing the rest of the solve like normal. I will also occasionally insert cross-edges with MU or rU moves if there was an easy F2L pair that I had created and inserted earlier. In my opinion, I think this is still CFOP, just doing about it a different way. A lot of other people have stated it, but even the WR solve was technically done with Petrus. Yusheng Du made a 2x2x3, got an EO skip, solved F2L, and did a ZBLL. He obviously uses CFOP as his main method, but that solve just happened to be Petrus-like. He meant for it to be a CFOP solve, but it just ended up being Petrus. As Andrew was saying in the episode, even if you alter a method slightly in the middle of the solve, if you eventually change it back at the end, it really isn't changing it to a different method.

I think that a person can't really consider themself "method neutral" unless they: 1) get approximately the same averages with the 2+ methods they use, and 2) use the methods somewhat evenly. For example, I wouldn't consider you method neutral if you use CFOP 75% of your solves and Roux the other 25%. I think there are a couple people on the forums who are method neutral who may be able to give more insight on this, though. (@RadicalMacaroni?)
CStimer has 1x1x2 scrambles as well.

1. (0.911)
2. (1.275)
3. 0.980
4. 1.017
5. 1.011
Average: 1.003
I am playing on mobile with 2 impostors.
Role: Crewmate

I got finished all of my tasks, went to security and went on cams then started solving cubes. I got killed while on cams, then kept solving. My body got reported, and one of the crewmates got voted out (so both impostors were still in the game), and eventually the impostors killed enough people and won.
Role: Crewmate

I finished all of my tasks and went to security just like last time, except this time, the impostors won by sabatoging O2.
Role: Crewmate

I finished all of my tasks except trash in case anybody accused me, then I went on cams to solve cubes, then with 5 people left, we tasked out and won, and I solved 13 cubes.
Role: Crewmate

I did my first task (garbage in cafeteria) then got killed in weapons when I was heading down to navigation. I started doing my tasks, but then the imposters won the game in literally 2 minutes by sabatoging the reactor.
Role: Crewmate

3 people left right at the start (those lame little kids who leave if they're not impostor), then I got killed right away, tried to finish doing my tasks, but the impostors won right afterwards by getting kills.
 
Joined
Jun 12, 2019
Messages
53
I think there are a couple people on the forums who are method neutral who may be able to give more insight on this, though. (@RadicalMacaroni?)
I'm not method neutral, but I'm sub 8 with ZZ and sub 9 with Roux so I've tried it before. And I have to say that honestly, its probably not worth it except special solves where you see something super easy with a method you don't main (5 move eocross or FB + DR for example)
 

Mr. McCubing

Member
Joined
Nov 8, 2020
Messages
256
YouTube
Visit Channel

Andrew Nathenson and I have been doing a cubing podcast for a couple years now, so I figured I'd start a thread here too for when we post new episodes of the podcast! You're welcome to discuss episodes here, but we also have an official subreddit for discussing each episode linked below. You can click the logo above to see our Anchor page, which shows all the podcasting apps that we're listed on.

Latest: Episode 39 - Oops, All U Moves
Subreddit: https://www.reddit.com/r/layerbylayer/
Yay
 
Joined
Feb 23, 2019
Messages
639
Location
The FitnessGram Pacer Test is a multi stage...
Just now listening to the most recent episode of the podcast. I thought your discussion about ZZ method “purity” was really interesting, and I noticed that you also mentioned Roux in that discussion.

From my personal experience as a Roux solver, there is pretty good consensus that you should do whatever is the most efficient. I think this may have something to do with Kian Mansour being a pretty clear leader in the Roux field for a while and actively encouraging people to play around with blocks and EO and LSE at the end of the solve. To my knowledge, there was never a cuber using ZZ of the same central prominence as Kian was to Roux, and as such the decentralization may have caused an air of rigidity among the solving styles with no one to push the method along.

Secondly, move count in Roux is almost divine. Especially at a higher level, the amount of moves you spend to complete a task on the cube is almost always more important than the ergonomics of it (unless the optimal solution is very awkward to solve). Thus, people are very willing to play around with different block orders and alg sets to get a better movecount as long as they can still execute it relatively quickly, making experimentation part of the game.

Thirdly, many of the more fringe methods (ZZ being at the top of those, then Petrus, then LEOR and others) are similar to more common methods like Roux and CFOP, so when someone decides to use one of these methods, their “method identity,” as you mentioned, is often very important to them and they don’t want to stray from the orthodox version of the method that they learned. I tend to be a little more freeform with blockbuilding in my Roux solves (again, because of the things I previously outlined) but many people who use other methods may not be so inclined.

I wanted to mention this because I think the discussion about “method culture” is really interesting and deserves more time, and I’d like to hear y’all dive into this some more!
 

abunickabhi

Member
Joined
Jan 9, 2014
Messages
2,078
Location
Yo
WCA
2013GHOD01
YouTube
Visit Channel
I don't know much (actually anything) about ZZ, so I can't say much about that, but like Kit, I sometimes take a FreeFOP/Petrus approach to certain scrambles. For example, if i see a really easy 2x2x2 block in inspection, I will plan out that and then move on to finishing the cross, then doing the rest of the solve like normal. I will also occasionally insert cross-edges with MU or rU moves if there was an easy F2L pair that I had created and inserted earlier. In my opinion, I think this is still CFOP, just doing about it a different way. A lot of other people have stated it, but even the WR solve was technically done with Petrus. Yusheng Du made a 2x2x3, got an EO skip, solved F2L, and did a ZBLL. He obviously uses CFOP as his main method, but that solve just happened to be Petrus-like. He meant for it to be a CFOP solve, but it just ended up being Petrus. As Andrew was saying in the episode, even if you alter a method slightly in the middle of the solve, if you eventually change it back at the end, it really isn't changing it to a different method.

I think that a person can't really consider themself "method neutral" unless they: 1) get approximately the same averages with the 2+ methods they use, and 2) use the methods somewhat evenly. For example, I wouldn't consider you method neutral if you use CFOP 75% of your solves and Roux the other 25%. I think there are a couple people on the forums who are method neutral who may be able to give more insight on this, though. (@RadicalMacaroni?)
I agree with your assumptions about method neutrality. For the next decade atleast, I do not expect speedcubers to become method neutral. Maybe after the competition gets crazy tough, becoming method neutrality can provide an edge.

Currently, I do 65% CFOP on 2H and 30% Roux, and 5% 3-style or some hybrid (2H CFOP global-10, 2H Roux global-10.5). In OH, it is 80% Roux and 20% CFOP (OH Roux global 17, OH CFOP global-19).
So, achieving the 50-50 split with both method is not ideal.
 

PapaSmurf

Member
Joined
Jan 4, 2018
Messages
928
WCA
2016TUDO02
YouTube
Visit Channel
Seeing as it's kinda a big theme of the episode, ZZ.
The big thing (as identified) for ZZ is EO. That differentiates it most from CFOP and Roux, so doing EOCross is really just EOLine but blocks are made generally more ergonomic. Rotating for cases that are better with rotations (I don't think there are any for TH, but triple sexy from BR for OH is probably one) is still keeping EO and not 'breaking ZZ', it's just doing something in a more ergonomic way and if it's better, do it. Same for F moves. For example, F' R U R' U' R' F R (or inverse) is the best alg for the given case. There's no point delibrately restricting yourself to a subset of moves just for the sake of method purity, just as most CFOPers don't do OLL/PLL exclusively.
ZZ is faster in general with EOCross, but I agree that doing EOLine can sometimes be better, but there's a major difference between the DR of Roux and EOCross in ZZ. The fastest solvers (Simon K from Germany, RadMac from earlier on in this thread, me) all say that EOCross is better as the thing you do 95% of the time and that's because it is. On the other hand, Rouxers who are faster tend to say it could help, but don't do it exclusively. If we assume constant EOCross with ZBLL for most cases ie. not sune/anti (this ignores things like pseudoslotting, multislotting, OLS tricks and other things that can reduce movecount) you get a movecount of 50-55 which is lower than CFOP. You also have really good ergonomics. Switch EOCross out for EOLine and while you save maybe 5 moves, you probably add at least 2 regrips in plus more blindspots. So comparing ZZ with EOLine purely to CFOP will show that CFOP is better, but if you compare with EOCross you get 2 very similar and competitive methods. If you wanna see more, read this post to see more of my reasoning. Also ZZ has none of the bad F2L cases like edge flipped in slot, corner solved, so that's pretty good.
Andrew, don't look at my signature.
 

xyzzy

Member
Joined
Dec 24, 2015
Messages
2,228
For some bizarre reason, in my younger years I did a lot of number crunching on paper, and that gave me a decent amount of "number sense" to work with. I made light work of most of the high school maths curriculum just cruising on my intuition; the exceptions were geometry (draw triangles and circles, chase angles, etc.) and integration because I didn't find those fun and so I hadn't practised much of those outside of school. I still mostly cruised through an undergraduate curriculum with this strat, but things got pretty dicey near the end. Turns out you need to actually do things to get good at them!

I also used to dislike trigonometry (since it's geometry-adjacent), and I'm still not exactly hot on it, but I pretty much had to force myself through it to get to the actually fun stuff like Fourier transforms. (Protip for anyone struggling with trig: learn complex numbers and use Euler's formula e^ix = cos x + i sin x as much as you can. Stuff like the double-angle/half-angle formulae just drop out of that with very little work.)

Integration by parts is basically just the product rule for derivatives: write (fg)' = f' g + f g', then integrate both sides and transfer one term to the other side. Well, I know the intuition for it, but I was doing some integration by parts earlier (after I knew there was a new podcast ep, before I listened to it) and I still had to keep checking to not mess up the signs. True calculus struggles.
I've been following this very-long-running blog that mostly touches graph theory, algorithms, and related topics (including some of computational geometry), 11011110. Might be of interest to Andrew!

A Cayley graph is the graph you get by taking the group's elements as the vertex set, and drawing edges whenever there is a single "move" to go between two vertices. (You have to decide what count as legal moves ahead of time, of course.) Cayley graphs are relevant to cubing in that they describe how you move from one cube state to another: an "optimal solution" is just a shortest path on the graph; "god's number" is just the graph's diameter.

Naturally, if you draw more edges, shortest paths might get shorter but never longer, and that's why optimal solutions in STM are usually shorter than those in FTM, and god's number in STM is (believed to be) smaller than god's number in FTM.
 
Joined
Feb 23, 2019
Messages
639
Location
The FitnessGram Pacer Test is a multi stage...
Just want to let you folks know (I am aware this was from the previous episode) that ice cream cake is usually mostly ice cream. There may be a layer or two of cake-style cake, but for the most part, the whole point of ice cream cake is the ice cream! I strongly encourage you to try ice cream cake from a local ice cream shop (not from a big chain).

I am also aware that I live on the east coast, and that may have something to do with it.
 

xyzzy

Member
Joined
Dec 24, 2015
Messages
2,228
Regarding the G5 timer and 4-pad mode:

The regulations currently have laxer requirements for a legal timer stop compared to timer starts (e.g. can't start with your palms, but can stop with your palms), and switching to 4-pad would basically make starts and stops symmetrical. Which is aesthetically pleasing, but maybe not an entirely good idea.

In comps I've been to, there was maybe one or two instances where the timer wasn't reset before I started inspecting; I don't practise with a timer at home, and I go to non-FMC-only comps only once a year, so this was a bit disorienting. (I didn't ask for extras; I think I noticed this either before inspecting or just before starting the solve (with plenty of inspection time to spare), so I just hit the reset button myself.) With the G5, having the timer accidentally be in 4-pad mode is likely to be even more disorienting. Competitors unfamiliar with the G5 might not know why they can't start the timer, and even competitors who are familiar would have to spend quite a few seconds pressing-and-holding the reset button to switch to 2-pad mode or to just use the timer in 4-pad mode and hope to remember to stop it with their thumbs too.

Looking at the WCA forums, there's been quite a few angry posts by Michał Halczuk in a thread about the G5. On one hand, I am extremely sympathetic to Michał's point of view; on the other hand, there are indeed legal risks if the WCA starts allowing non-(Speed Stacks) timers, and nobody really wants to take on that risk. Tear down IP law! (it's obvious that I'm a bit of an extremist on that front, isn't it)

---

Regarding WCA forums: I still don't understand the need for a real name policy there, but then again maybe I'm the weird one for wanting to be pseudonymous/anonymous. I guess people are less inclined to shitpost if it's irrevocably linked to their name, or something.
 
Top