• 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 40,000+ people from around the world today!

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

Is Petrus a dead method? Does the community still think it can be "fast"?

Is Petrus still a "fast" method?

  • No

    Votes: 41 30.8%
  • Yes

    Votes: 78 58.6%
  • Other

    Votes: 14 10.5%

  • Total voters
    133

JLarsen

Premium Member
Joined
Sep 30, 2008
Messages
1,880
Location
Dover, New Hampshire, USA
WCA
2009LARS03
YouTube
Visit Channel
The title is pretty descriptive. I've noticed since my long break from cubing that interest in the Petrus method has really dwindled. Back maybe a year or two when I was really trying to get fast, before Roux really caught on in popularity, there were lots of people interested in Petrus. I was constantly asked questions, and so was ErikJ. It was commonplace for newbies to say something like "Petrus is slow". And when anybody said that, they were just overwhelmingly smashed into the ground by people saying the opposite. I don't really see any of that today. I see a TON of Roux solvers nowadays, and a lot of them are realllly fast. But I'm not really seeing any new, fast Petrus solvers.

I think that Petrus just isn't as fast as Roux or Fridrich anymore. Let's try not to argue semantics or technicalities here either. We do that wayyyyy too much on the forum. I'm not attacking anything, or anyone. Now, when I say it's not as fast, I mean that if you were to put an equal time into practicing Petrus, Roux, and Fridrich, that you would have the highest times using Petrus. I'm really curious to see what the community thinks nowadays. Is Petrus still a viable method? As fast as Fridrich/Roux?

Edit: I know I left the poll answers pretty cut and dry with the yes/no, but I wasn't sure what to put otherwise. I just wanted a general idea.
 
Last edited:

emolover

Member
Joined
Aug 15, 2010
Messages
3,096
Location
Carmel, IN
WCA
2009MAHO01
YouTube
Visit Channel
I think it could be a fast method. But I think the orientation of the F2L edges for the last part is stupid. One cube rotation wont hurt.

Honestly though, I think Petrus and CFOP have converged because of X-cross.
 

JLarsen

Premium Member
Joined
Sep 30, 2008
Messages
1,880
Location
Dover, New Hampshire, USA
WCA
2009LARS03
YouTube
Visit Channel
I've tried X cross before it's really fun. The idea behind good edges is a good one though...at least in theory? Every step counts towards solving the cube. 2x2 reduces to 3 sides, 2x2x3 to 2. Step 4 is pew pew pew. Then only having to learn 7 OLLs is awesome possum. Could EO/ bad edges be over restrictive? F2L pairs are like the epitome of freedom....
 

Specs112

Member
Joined
Dec 19, 2010
Messages
322
Location
Ithaca, NY
WCA
2011ANDE03
Petrus is a perfectly serviceable method on the same tier as CFOP and Roux.

What does seem different to me is that unlike the deluge of "lol fried rice is best" people and the few vocal Roux elitists, there really aren't Petrus users who go out of their way to sell their method at all. Which in part explains the whole popularity thing.
 

Escher

Babby
Joined
Jul 23, 2008
Messages
3,374
WCA
2008KINN01
YouTube
Visit Channel
Petrus is just too 'big' for speed imo - the lack of restrictions (in terms of initial blockbuilding) make it efficient but also difficult to make a framework out of. There are just too many situations to take into account in the early stages/middle of a solve - when you want things to be simple but flexible.
Petrus/CFOP haven't converged imo - CFOP is just badly named these days ;)

I think another point is the restriction to 2-gen for blockbuilding while finishing f2l is bad - 2gen just isn't fashionable these days, as people have discovered why it isn't as OP as they thought it was :p

If the last year is anything to go by, hopefully we should be seeing a bunch more people using Roux. It needs a bunch of development, in bad and good directions (like CFOP has experienced), but I think one day it could feasibly be the most-used method.
 

qqwref

Member
Joined
Dec 18, 2007
Messages
7,834
Location
a <script> tag near you
WCA
2006GOTT01
YouTube
Visit Channel
I think Petrus definitely has some serious potential, but not quite as much as Roux and Fridrich do. That's not to say you can't get fast with Petrus, though; if someone with very good tps posted a sub-10 Petrus average I'd probably believe it, but I really doubt you could get near the best Fridrich times that have been achieved. As Escher says, 2gen isn't as overpowered as people thought - it's true that you can get very high turnspeed with it, but you can also get high turnspeed with other substeps, and 2gen F2L slots often use significantly more moves. And, if you ask me, having oriented edges for the LL doesn't save time if you're still doing it in two steps - especially considering the extra moves/time you need to set the oriented edges up.
 

Noahaha

blindmod
Joined
Dec 19, 2011
Messages
3,015
Location
CT
WCA
2012ARTH01
YouTube
Visit Channel
I stumbled across Petrus right after I learned Ortega as my first speedsolving method. I loved learning from Lars's site. I got down to about 40 seconds and then stopped for a few months. I picked the cube back up in December, and since then I have moved from 40 seconds to 25. I am still steadily improving with Petrus, and I don't plan on switching any time soon. I feel like for me when I drop time, it's because I have gotten better at combing two steps into one. For example, my times dropped considerably when I started finding and fixing bad edges while making the 2x2x3, and again when I started planning step 4 during EO. I think that if someone were able to blend together the steps extremely well, sub 10 averages would be possible. Perhaps I'm being a little too optimistic since I've been improving so fast recently, but I don't think the Petrus method will move anywhere without optimism. I think that the method could seriously benefit if more people started using it. I'm positive that there are many improvements to be made, such as forcing EO skips every time and possibly a missing link solution during EO. It just frustrates me that Petrus discussion and ideas end up getting so stifled in general, since I see a world of possibilities for the method.

Anyway, Josh, I think that you should not lose faith in the method at all, but challenge yourself to break this horrible plateau you've hit. I'm rooting for you.
 

JLarsen

Premium Member
Joined
Sep 30, 2008
Messages
1,880
Location
Dover, New Hampshire, USA
WCA
2009LARS03
YouTube
Visit Channel
I stumbled across Petrus right after I learned Ortega as my first speedsolving method. I loved learning from Lars's site. I got down to about 40 seconds and then stopped for a few months. I picked the cube back up in December, and since then I have moved from 40 seconds to 25. I am still steadily improving with Petrus, and I don't plan on switching any time soon. I feel like for me when I drop time, it's because I have gotten better at combing two steps into one. For example, my times dropped considerably when I started finding and fixing bad edges while making the 2x2x3, and again when I started planning step 4 during EO. I think that if someone were able to blend together the steps extremely well, sub 10 averages would be possible. Perhaps I'm being a little too optimistic since I've been improving so fast recently, but I don't think the Petrus method will move anywhere without optimism. I think that the method could seriously benefit if more people started using it. I'm positive that there are many improvements to be made, such as forcing EO skips every time and possibly a missing link solution during EO. It just frustrates me that Petrus discussion and ideas end up getting so stifled in general, since I see a world of possibilities for the method.

Anyway, Josh, I think that you should not lose faith in the method at all, but challenge yourself to break this horrible plateau you've hit. I'm rooting for you.

You give me hope! Maybe there is something that could be improved in the method...but I've never been much of a pioneer myself. When people really get into the cube theory/algorithm generation jargon I just get completely lost. I can hardly grasp FMC and commutators, so I really don't think I would be able to do much as far as finding this "missing link". I'm not even really sure what it would entail. I know I've heard the term in reference to ZZ before, but that's about it.

Would someone be willing to try and come up with some ideas? I'm really not a major intellect when it comes to stuff like this. I've contributed hardly at all to the method. I've just put in a lot of time and effort into turning blockbuilding into a series of "cases" in my mind. Like Escher was saying there's just so many possibilities. I'm somewhat fast because I just don't use them.
 

Noahaha

blindmod
Joined
Dec 19, 2011
Messages
3,015
Location
CT
WCA
2012ARTH01
YouTube
Visit Channel
You give me hope! Maybe there is something that could be improved in the method...but I've never been much of a pioneer myself. When people really get into the cube theory/algorithm generation jargon I just get completely lost. I can hardly grasp FMC and commutators, so I really don't think I would be able to do much as far as finding this "missing link". I'm not even really sure what it would entail. I know I've heard the term in reference to ZZ before, but that's about it.

Would someone be willing to try and come up with some ideas? I'm really not a major intellect when it comes to stuff like this. I've contributed hardly at all to the method. I've just put in a lot of time and effort into turning blockbuilding into a series of "cases" in my mind. Like Escher was saying there's just so many possibilities. I'm somewhat fast because I just don't use them.

I think we are opposites, because i spend a lot of time thinking about the method and how it works, and not enough time practicing my cases and such. I came up with a missing link solution in January where corners are permuted during EO. It was very inefficient, and a better solution than mine that accomplished the same thing was discussed in an old thread. I don't think either of these could beat CPLS, but if someone found a good way to permute the corners during EO, the LL would be down to 2GLL, and this would make the EO step useful, which was one of qqwref's problems. This is why I answered yes to your poll, because I feel that petrus is on the verge of becoming a sub 10 method if only enough thought was put into it.

pseudo blocks.
 

PandaCuber

Member
Joined
Jun 24, 2011
Messages
1,689
Location
Uruguay
YouTube
Visit Channel
I plan on making Petrus my Main method for Big Cubes. I dont like CFOP and im slow at it and Roux is just a hassle(For BC) , so Petrus is the best.

I really wish you had a website where you explained the method(I didnt understand Lars site) and make it fun.

Waffle made a fun site to learn Roux and people loved it.
Think about it?
 

Noahaha

blindmod
Joined
Dec 19, 2011
Messages
3,015
Location
CT
WCA
2012ARTH01
YouTube
Visit Channel
I plan on making Petrus my Main method for Big Cubes. I dont like CFOP and im slow at it and Roux is just a hassle(For BC) , so Petrus is the best.

I really wish you had a website where you explained the method(I didnt understand Lars site) and make it fun.

Waffle made a fun site to learn Roux and people loved it.
Think about it?

Yeah, that's the big problem with petrus. Unfortunately, there's a ton you have to figure out for yourself. I think that even if you did learn from a new site, you should still try to understand Lars's site because there are a few gems in there. Petrus is pretty simple actually though. An experienced cuber should be able to learn it quickly. Just build a 2x2x3 as fast as you can, then watch Josh's bad edge tutorial and then fill in two pairs.
 

JLarsen

Premium Member
Joined
Sep 30, 2008
Messages
1,880
Location
Dover, New Hampshire, USA
WCA
2009LARS03
YouTube
Visit Channel
I did make a Petrus website tutorial thing before but I really tried to focus on making Petrus faster so most of my tutorials were "advanced". If someone with some website skills wanted to make a website I wouldn't mind supplying videos/written content to supplement it but only if there was actually a demand for it. There isn't really anybody who would go there except you. (ironic that your name is panda cuber =P)

I like Petrus because it is a more intuitive approach than Fridrich and thusly a lottt more fun. And I like it over Roux because M slices on big cubes are horrible. I just did some playing around with weird free form variations of petrus solves that didn't include EO and I didn't really find anything that I thought was too useful. One idea I had was make a 2x2 block and then solving F2L with FRU, which was doable, but all I really wound up doing was some strange freefop/blockbuilding stuff with some edge control. It was cool, but definitely not great for speedsolving. It has wayyy to much freedom. Freeform methods are not optimal for speedsolving.

Now Noah; you were talking about putting something like psuedo blocks into a speedsolve. Really all I got out of the past half hour messing around was this; I need to turn faster. My problem isn't in my efficiency (at least I think). Back in the day when I was trying to improve I assumed that my problem was lookahead and efficiency. This was largely fueled by ErikJs famous "advanced petrus tutorial" pt.1 and 2, which if you ask me, are definitely solutions that are not geared towards a speedsolve. They're probably more a reflection of his blockbuilding skills in FMC more than speedsolving. So here's what I'm gonna do. I'm gonna turn fast, and see what I can do. Maybe I'll actually learn a couple new PLLs since some of mine could be a lot better. I'll just throw everything I have into TPS. Assuming my move counts are fine. Is 60 alright for a speedsolve? If you check the solution to the slow motion vid I put up recently Brest did a reconstruction and it was about on par with most of my solutions. Pretty typical.

TLDR; gotta go fast. I'm just going to try to improve TPS and see where it gets me.
 

PandaCuber

Member
Joined
Jun 24, 2011
Messages
1,689
Location
Uruguay
YouTube
Visit Channel
Yeah, that's the big problem with petrus. Unfortunately, there's a ton you have to figure out for yourself. I think that even if you did learn from a new site, you should still try to understand Lars's site because there are a few gems in there. Petrus is pretty simple actually though. An experienced cuber should be able to learn it quickly. Just build a 2x2x3 as fast as you can, then watch Josh's bad edge tutorial and then fill in two pairs.

My problem isnt doing the method. My problem is making it efficient.
As a Roux user, I look for the most efficient solution possible. (That I can think of).
When doing a Petrus solve, My EO takes forever and so does my 2 gen pairs cause im not used to it. And I dont know the shortcuts to make it efficient.
 

qqwref

Member
Joined
Dec 18, 2007
Messages
7,834
Location
a <script> tag near you
WCA
2006GOTT01
YouTube
Visit Channel
So here's what I'm gonna do. I'm gonna turn fast, and see what I can do. Maybe I'll actually learn a couple new PLLs since some of mine could be a lot better. I'll just throw everything I have into TPS. Assuming my move counts are fine. Is 60 alright for a speedsolve? If you check the solution to the slow motion vid I put up recently Brest did a reconstruction and it was about on par with most of my solutions. Pretty typical.
Sounds good. And 60 is fine - normal, even. Gogogo 6tps :]
 
Top