How fast can Petrus method go?

[jackolanternsoup]

Ok then: if 99% of people used that method... You're fighting a losing battle, I would stop if I were you.

I can't lose to that argument until I see a 30 move T-perm. In a few minutes I will memorize it then I will give you an A5 for your uber inneficient method.

Be sure to take a video of your A5 and post it.

 

[Dene]

I bet that if my physical speed was better for step 4 and the last layer then Petrus method could be just as fast as Fridrich.

If 99% of cubers (thought experiment) used Method A,
Then Method A would be the fastest method.

Ok, then consider this method:

Step 1: Before looking at the cube, you do an A perm. There is no reason for this, you just do it.

Step 2: You then randomly do another 10 moves, why not?

Step 3: Proceed to solve the cube 1 piece at a time using a 30 move T perm.

Step 4: With 4 pieces to go, randomly do another 10 move scramble.

Step 5: Finally, finish solving the cube.

If everyone did this method would it be faster than Fridrich?

This analogy is not apt. If everyone did this method, then people who do Fridrich in this imaginary world would be the fastest in the world.

It's not that 99% of people do Fridrich. It's that EVERYONE who is top in the world uses a variation of it. Granted, this doesn't prove beyond any doubt that Fridrich is fastest...but it certainly implies it.

What do you mean? He explicitly said: If 99% of people use a method, it will be the fastest one. I gave a method. Imagine if 99% of people used it? They would certainly not be faster than the 1% who use Fridrich (or some other more efficient method). Of course by 99%, I mean 99% of speedcubers, not of some other morons. You haven't changed anything.

 

[Speedy McFastfast]

I chose the sub 10 button. Does everyone know why Petrus isn't as fast as Fridrich yet? Because nobody has used it that would be good with it.

Take it this way: Anyone could learn to speak German, but for some people, it sortof just "clicks." It's the same way with cube methods.

Anyone can learn Petrus, but there are only a few people where it will really click in their heads. We haven't found those people yet. Fridrich method has found people like Chan, however, so for now it's faster.

 

[AvGalen]

I bet that if my physical speed was better for step 4 and the last layer then Petrus method could be just as fast as Fridrich.

If 99% of cubers (thought experiment) used Method A,
Then Method A would be the fastest method.

Ok, then consider this method:

Step 1: Before looking at the cube, you do an A perm. There is no reason for this, you just do it.

Step 2: You then randomly do another 10 moves, why not?

Step 3: Proceed to solve the cube 1 piece at a time using a 30 move T perm.

Step 4: With 4 pieces to go, randomly do another 10 move scramble.

Step 5: Finally, finish solving the cube.

If everyone did this method would it be faster than Fridrich?

This analogy is not apt. If everyone did this method, then people who do Fridrich in this imaginary world would be the fastest in the world.

It's not that 99% of people do Fridrich. It's that EVERYONE who is top in the world uses a variation of it. Granted, this doesn't prove beyond any doubt that Fridrich is fastest...but it certainly implies it.

What do you mean? He explicitly said: If 99% of people use a method, it will be the fastest one. I gave a method. Imagine if 99% of people used it? They would certainly not be faster than the 1% who use Fridrich (or some other more efficient method). Of course by 99%, I mean 99% of speedcubers, not of some other morons. You haven't changed anything.

I am pretty sure that 99% or more of the cubers (not speedsolvers) that can solve a cube use a beginners method like layer-by-layer. No way to prove that though

 

[fanwuq]

I think Petrus is for smart people and Fridrich is something that us regular humans can use to beat the smart people. I'm just too stupid to see good blocks in 15 seconds and recognize EO quickly. If some genius with perfect lookahead comes out, he/she would probably use something crazy like Petrus+ZBLL without any recognition delay and easily sub-8. If only I can use Petrus, then I'd have no problem at all with learning all of MGLS...

 

[ErikJ]

I'm just too stupid to see good blocks in 15 seconds

that's something I have to work on.

I spend too much time looking for a 2x2x2. I should be able to find one then look to see which 2x2x3 would be best and find the pieces for that. Direct 2x2x3 make Petrus super fast.

 

[mrbiggs]

What do you mean? He explicitly said: If 99% of people use a method, it will be the fastest one. I gave a method. Imagine if 99% of people used it? They would certainly not be faster than the 1% who use Fridrich (or some other more efficient method). Of course by 99%, I mean 99% of speedcubers, not of some other morons. You haven't changed anything.

The specifics of his sentence were wrong, but the overall point is correct. Fridrich should be currently considered best because most of the best use it. You're correct in that popularity as a whole is irrelevant, but I think the gist of his argument was along the right lines.

 

[enigmahack]

The only point I'd like to make, and I don't know if anyone brought it up:
Best method indicates that whatever is best for you will also be best for me.

Best for xxxx is very subjective for just that reason.

The real question would be: Which method is more likely to get faster averages by those who practice it.

From a scientific perspective, you would need to have a group of people all only learning Fredrich, another Petrus, another Roux, etc.
All groups are only allowed to study/practice these methods x hours per day for a period of y days. Even that is VERY biased because a 13 year old is going to learn WAY faster than a 29 year old. There are too many variables to be able to specify which is "better"

Roux seems to be best for fewest moves "generally", Petrus very shortly behind it. Fredrich is a very fast brute-force alg crunching method, but I think they ALL could be equally fast given the right circumstances.

 

[JLarsen]

I think Petrus is for smart people and Fridrich is something that us regular humans can use to beat the smart people. I'm just too stupid to see good blocks in 15 seconds and recognize EO quickly. If some genius with perfect lookahead comes out, he/she would probably use something crazy like Petrus+ZBLL without any recognition delay and easily sub-8. If only I can use Petrus, then I'd have no problem at all with learning all of MGLS...

Hahahaha, I could get along with you. Anyway MGLS isn't even complicated though. And one final, thing, seeing blocks, you just aquire a really different view. You have to do it a lot more than you would a cross to be good at it, but I think it's freakin cool. Crosses just make me want to eat grenades, so for me, Petrus is a more interesting approach, than ...well........SPAM SPAM SPAM SPAM SPAM. Ha, I have no clue what I just did, but it's done.

 

[qqwref]

Here's what I think. There are two options. Either Petrus is at least as good as Fridrich for speedsolving, or Fridrich is faster on an absolute scale. I'm talking about potential speed, which determines what is considered good.

- If Petrus is at least as good as Fridrich, then I suggest we judge Petrus users by Fridrich standards. Makes sense, right? But then all you Petrus users who are not sub-20, or barely? You wouldn't make it to a competition final round, go practice more. How about if you're one of the best Petrus users in the world? That's not that impressive, you're just starting to be decent. My point of view is that that the only reason ErikJ and Johannes are impressive 3x3 solvers is because they use Petrus. If you really think Petrus is just as good as Fridrich, you should find them no more impressive than all the Fridrich guys who can average in the 13s. Those guys aren't called gods of the method - they're not even considered to be world-class. I think 13.41 is a truly amazing Petrus average, but if you did a 13.41 average of 5 in competition you'd be ranked, like, 40th in the world. That's not godly.

- If Fridrich is better than Petrus... well, we don't even have to talk about that, because nobody who uses Petrus would agree with me. But I really do think this is a possibility. (As for Roux, I'm really not sure.) You can't trade tips on Petrus as easily as Fridrich, because it's so intuitive. Once you know how to do the edge flip and LL as efficiently as possible short of ZBLL, there's really nothing left for anyone else to teach you. So if you want to complain about not having enough resources for your method, stop - it's irrelevant. You have to learn intuition on your own. And if Johannes and Erik could spend years doing Petrus only, and still not be sub-12, I'm not sure anyone can. I simply don't think you can be as fast at Petrus as the current fastest cubers are at Fridrich.

 

[fanwuq]

I've done Petrus all day today until now with 90% of the time doing 1 or 2 look CLS and getting times ranging from 19 to 45. The 19 was easy blocks, and decent recognition of EO and good LL. After doing so much Petrus, Fridrich just feels so easy. It's feels like doing a 2x2 after doing a 4x4 cube or Beginner's LBL after Fridrich. There was great improvement in my lookahead because the method is just so brain dead. Can someone who use Petrus compare the times of their Petrus and Fridrich F2L only?

Fridrich:
5: 00:10.71 x This one was continued as a 1LLL with ELL I stopped the timer for the F2L, but the whole solve should have been around 13 seconds.
4: 00:15.02 x
3: 00:15.40 x
2: 00:15.34 x
1: 00:13.98 x
Avg. 5:
00:14.09
3 of 5:
00:14.78
Petrus:
5: 00:15.37 x
4: 00:31.04 x Last 2 slots mess up
3: 00:19.01 x
2: 00:22.67 x
1: 00:22.89 x EO mess up
Avg. 5:
00:22.19
3 of 5:
00:21.52
This is a bad average for me.
Fridrich trial2:
5: 00:13.47 x
4: 00:12.43 x
3: 00:12.76 x
2: 00:11.04 x
1: 00:14.30 x
Avg. 5:
00:12.80
3 of 5:
00:12.89
This is more typical when I warmed up.
Petrus trial2:
5: 00:13.73 x
4: 00:21.13 x
3: 00:19.43 x
2: 00:19.22 x
1: 00:25.40 x
Avg. 5:
00:19.78
3 of 5:
00:19.93
At least one good F2L! It was EO skip. Apparently I suck more at EO than I thought.

 

[krazedkat]

It can only go as fast as YOU can go ...

 

[JLarsen]

I've done Petrus all day today until now with 90% of the time doing 1 or 2 look CLS and getting times ranging from 19 to 45. The 19 was easy blocks, and decent recognition of EO and good LL. After doing so much Petrus, Fridrich just feels so easy. It's feels like doing a 2x2 after doing a 4x4 cube or Beginner's LBL after Fridrich. There was great improvement in my lookahead because the method is just so brain dead. Can someone who use Petrus compare the times of their Petrus and Fridrich F2L only?

Fridrich:
5: 00:10.71 x This one was continued as a 1LLL with ELL I stopped the timer for the F2L, but the whole solve should have been around 13 seconds.
4: 00:15.02 x
3: 00:15.40 x
2: 00:15.34 x
1: 00:13.98 x
Avg. 5:
00:14.09
3 of 5:
00:14.78
Petrus:
5: 00:15.37 x
4: 00:31.04 x Last 2 slots mess up
3: 00:19.01 x
2: 00:22.67 x
1: 00:22.89 x EO mess up
Avg. 5:
00:22.19
3 of 5:
00:21.52
This is a bad average for me.
Fridrich trial2:
5: 00:13.47 x
4: 00:12.43 x
3: 00:12.76 x
2: 00:11.04 x
1: 00:14.30 x
Avg. 5:
00:12.80
3 of 5:
00:12.89
This is more typical when I warmed up.
Petrus trial2:
5: 00:13.73 x
4: 00:21.13 x
3: 00:19.43 x
2: 00:19.22 x
1: 00:25.40 x
Avg. 5:
00:19.78
3 of 5:
00:19.93
At least one good F2L! It was EO skip. Apparently I suck more at EO than I thought.

Both me and Erik, and all the other Petrus users I know use OLL/PLL. Often our last layer is just a sune =].

 

[nitrocan]

Both me and Erik, and all the other Petrus users I know use OLL/PLL. Often our last layer is just a sune =].

Last layer, or OLL?

 

[JLarsen]

Both me and Erik, and all the other Petrus users I know use OLL/PLL. Often our last layer is just a sune =].

Last layer, or OLL?

Last layer, that is if you get a pll skip, which is what i meant.