Learning the Petrus System

[dChan]

I've been looking for some ways to get quicker times for special cases so I am [continuing] learning of the Petrus systeM. Pretty much everyone knows how to do the 2x2x3 block and I figured outhow to orient the edges in step 3 by myself. But I don't know exactly what a Niklas, Eve, etc. is? Is there a list of algs written down already somewhere that give me the notation for these algs? Right now I just permute the corners with the usual PLL algs then I orient them with a Sune. So I assume it would be more efficient to use a Niklas but I don't know it. Could somebody help? I just need like a list of the algs for the last layer of Petrus.

Thanks,
dChan

 

[choipster]

i'm having the same problem as you are. i've been looking at the step 5 + 6 index to cut some time, but when i write out the algs in notation and perform them when i see one of the cases on my cube, it never works out. ive been using this page:

http://lar5.com/cube/yMain.html

 

[dChan]

Yeah, I wish he just wrote down the algs in notation instead of using only applets as if people don't know what notation is.

 

[joey]

The booklet that comes with the cube actually shows the nikklas.

Its
R U' L' U R' U' L U

 

[dChan]

Yeah, I figured that out after remembering what I learned. Thanks man. Now I can solve full Petrus in 40 seconds. I need to improve though. Step 3 sometimes takes more moves than is really neded for me.

 

[Fred]

Some of the really fast Petrus users actually do OLL/PLL for the last layer. If you already know the corner orientation algs and PLL, you should just use them and focus on learning Petrus f2l.

 

[dChan]

Ah, I see. Great idea. Because it cancels out repeating algorithms. So I guess I can just do OLL and PLL instead of Niklas and Sune + Edge PLL. It will probably increase my times with Petrus. Thanks Fred!

 

[Me]

as far as i can say the Petrus method is just something fun to play around with, as for speed the method barely ever breaks the 25 second barrier. Just see Lars Pertus' profile on the official records database to see what i mean.

otherwise you have to give the guy credit though for coming up with his own method rather then becoming a Fridrich method clone like most of us.

 

[Fred]

Actually, Anthony Hsu uses Petrus and recently won the Cornell Spring competition with an average of 14.62. So the Petrus method can be blazingly fast.

 

[Arakron]

Yes, Petrus can be as fast as Fridrich, although Petrus himself is not competitive anymore. The LL on the Petrus site is pretty basic. I don't know exactly which LL algs you know right now, but 2-look for Petrus is pretty easy - 1 of 7 to orient corners, then PLL. Either that or COLL and an edge perm.

 

[Cubinator]

How feasable is 1-look LL for Petrus? (7 OLL x 21 PLL = 147 algorithms). It's probably doable, although recognition might not be great.

My times with Petrus stink: around 1:30 or so for me. My times are much better with the method I made up, and of course better again for my normal speedsolve method (Fridrich).

 

[Harris Chan]

Actually, there's more than 147....you need to consider that there's 4 PLLs per orientation. So it's 4x7x21=588.

Ok so may be I'm wrong with the exact number, but it's around 500s algos, AKA ZBLL. That's what ZB does, doing LL in one look, by orientating the edges first.
It's just the F2L part that differs.

 

[tenderchkn]

It's at least 500 when you include mirrors and reflections. But you can just learn COLL to do Petrus - some of them are Niklas/Sune combos - the recognition is so much faster than if you're using pure Petrus.

 

[joey]

Originally posted by Arakron@May 2 2007, 02:00 AM
Yes, Petrus can be as fast as Fridrich, although Petrus himself is not competitive anymore.

He does still compete.

 

[Johannes91]

Originally posted by tenderchkn@May 2 2007, 04:44 AM
It's at least 500 when you include mirrors and reflections.

I'm pretty sure it's 493. I know a little over 40%.

 

[Arakron]

Originally posted by joey+May 2 2007, 06:39 AM--></div><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td>QUOTE (joey @ May 2 2007, 06:39 AM)</td></tr><tr><td id='QUOTE'> <!--QuoteBegin-Arakron@May 2 2007, 02:00 AM
Yes, Petrus can be as fast as Fridrich, although Petrus himself is not competitive anymore.

He does still compete. [/b][/quote]
Right, but I meant that he probably won't be winning competitions any time soon, which is not the case with Anthony Hsu and Johannes.

 

[dChan]

I started learning Petrus about Monday and now I average 35 seconds. So I believe Petrus can be pretty fast. Of course I've had two months of cubing practice so that could account for the fast acceleration. But it is a fast method I believe.

Right now I do Petrus + OLL/PLL. Kind of a bad time to be switching my cubing method right before the day of the competion(I'll probably be doing my primary method during the comp because right now 35 seconds isn't good for me to be doing in a competition).

Anyway, I was wondering if it was good for my step 3 to be correcting two bad edges at a time? There can be at most 6 bad edges and it takes me 3 moves to correct each two sets of edges which is 9 moves . When I get to step 3 the clor facing me is usually yellow so I permute the yellow/green edge(green is my bottom) first then I correct the edges in the FR and UB position and the FL and UB position using the three move correction. This allows me to go straight into inserting the FR and FL edges and using one of three basic ZBF2L algs that preserve LL edge orientation to permut the corners.

Is this a good way to take care of step 3? My fastest time with this method is 30 seconds which is pretty close to the 20s. With a bit more practice it could possibly alllow me to get my 26 sec average that I have using my other method.

 

[joey]

Originally posted by dChan@May 4 2007, 08:43 PM
I started learning Petrus about Monday and now I average 35 seconds. So I believe Petrus can be pretty fast. Of course I've had two months of cubing practice so that could account for the fast acceleration. But it is a fast method I believe.

Right now I do Petrus + OLL/PLL. Kind of a bad time to be switching my cubing method right before the day of the competion(I'll probably be doing my primary method during the comp because right now 35 seconds isn't good for me to be doing in a competition).

Anyway, I was wondering if it was good for my step 3 to be correcting two bad edges at a time? There can be at most 6 bad edges and it takes me 3 moves to correct each two sets of edges which is 9 moves . When I get to step 3 the clor facing me is usually yellow so I permute the yellow/green edge(green is my bottom) first then I correct the edges in the FR and UB position and the FL and UB position using the three move correction. This allows me to go straight into inserting the FR and FL edges and using one of three basic ZBF2L algs that preserve LL edge orientation to permut the corners.

Is this a good way to take care of step 3? My fastest time with this method is 30 seconds which is pretty close to the 20s. With a bit more practice it could possibly alllow me to get my 26 sec average that I have using my other method.

You should try learn the other ways of orienting 2,4, or 6, with one set of moves. You can learn all the positions, 48 I think, but if you look at a few of the examples on his site, you can make them up.