To switch or not to switch...

[Swordsman Kirby]

Petrus takes much longer to get used to

I don't think you can really compare them, unless you've practised both equally much since you started cubing. How did the quote go... "Learning Lisp takes 1 day, except for those who already know Fortran. They'll need 2 days."

I average around 60 or so moves with Fridrich, and I average around 55 with Petrus. I have room to improve, but I don't think it's really THAT much more efficient...

You sure do. About 45 is possible doing just one step at a time. I'd like to know how many moves the best Fridrich cubers average for F2L.

Probably low 30's. I think Petrus + optimal ZBLL can do sub-40 total. Imagine, way into the future, when people can do optimal 2x2x3 starts... And when people can do optimal 2x2x3+fix bad edges...

And when people can do optimal solves period!

 

[Mike Hughey]

We already have a pretty good answer to the question of how many moves the best Fridrich cubers average for F2L - in this thread:
http://www.speedsolving.com/showthread.php?t=2504

It was just 3 solves, but a pretty large sample of top cubers. Average was 33.83, according to Kenneth.

 

[Kenneth]

"Learning Lisp takes 1 day, except for those who already know Fortran. They'll need 2 days."

But seriusly, the more cube methods you know the easier it is to learn more of them. You can reuse algs or even steps, I know this for sure because I know how to solve CF (at least four methods), Petrus (OLL + PLL or COLL if so) exept bad edges that I have not practise, (intuition works fine but maybe not optimal). Full Roux, Any LBL using OLL, PLL, CLL, COLL, ELL as steps (add to that a great number of CLL + EOLL and ZB's). For edges first, my record is sub 25

Keyhole F2L, Fridrich F2L, VHF2L (some ZB), Heise F2L + around 5 methods to solve F2L that I created myself.

Show me a method that I don't know that uses around 20-40 algs and I can promise a sub 60 solve in 3-5 tries.

And I'm not bragging, it is ment to be an inspiration to others, it is not that hard to learn more methods once you started.

 

[mizzle]

Kenneth's right. The more you know, the faster you learn new things, and the easier it is to recycle the old.

It's really hard to pin down my method these days, because it switches multiple times a solve. I'll start out with some variation of orient edges first, Petrus block/Xcross, Roux blocks, or standard cross, put in pairs, screw with the centers, orient edges, orient corners early, and swap between CLL+ELL, OLL+PLL, COLL+EPLL, and various chunks of each of those.

I don't think I've had an average with a single method solid through in over a year. And to be honest, I think that's a lot more fun than getting to be sub15 with straight Fridrich.

 

[Johannes91]

We already have a pretty good answer to the question of how many moves the best Fridrich cubers average for F2L - in this thread:
http://www.speedsolving.com/showthread.php?t=2504

It was just 3 solves, but a pretty large sample of top cubers. Average was 33.83, according to Kenneth.

But if you take only Harris' solves, for example, the average is 29.

"Learning Lisp takes 1 day, except for those who already know Fortran. They'll need 2 days."

But seriusly, the more cube methods you know the easier it is to learn more of them.

I agree in general, but if you know one method and want to learn a very different one, you might have to unlearn many things and get rid of some habits. It might've been easier without knowing the first method at all. For example, when giving tips to a few Fridrich cubers who are learning Petrus, I've noticed that they keep talking about "cross edges" and F2L ce-pairs. Not to mention color neutrality...

 

[Speedy McFastfast]

I double Johannes on that (too lazy to quote today!) because when I switched over to Fridrich, I realized that I could only see ways to block together things in "Fridrich style" C/E pairs. I am just now at sub 20 starting to see how to put together corners and edges that would belong to the U or D layer in Fridrich.

Johannes, you say 45 is possible? But is that during a full speed solve? Slow Petrus solves are insanely good in few moves, but I don't think that it's possible to be nearly that optimal at full speed. How many moves do you average on speedsolves on Gabbasoft? That's where I'm getting that 55 from. That's also using OLL/PLL, no nifty tricks

And while we're on the topic of few moves, how many moves does your F2L take, Johannes? I only worry about getting the F2L in few moves (I am for about 30) because the LL is all speed for me.

(Time for on topic!) For those who want to switch to Petrus, this is the kind of technical stuff you get to play with! Isn't it fun!

 

[Johannes91]

Johannes, you say 45 is possible? But is that during a full speed solve?

My average is around there, I think. Or at least I can average sub-20 and ~45 moves, in competitions where only speed matters I probably use a bit more.

How many moves do you average on speedsolves on Gabbasoft? That's where I'm getting that 55 from.

I haven't used Gabbasoft in a long time, so don't know.

And while we're on the topic of few moves, how many moves does your F2L take, Johannes?

The F2L/LL distinction isn't that clear to me, because I try to get a 1-look LL whenever I can. I did 5 speedsolves and got, in FTM: 32 25 22 32 25. So about 25-30 on average.

 

[Speedy McFastfast]

Ha, I guess it is a bit harder for you to get just the F2L because you know all of those LL tricks. I only use OLL and PLL, so I focus mainly on getting my F2L in the fewest moves I possibly can. Either way, I've got a lot of room to improve.