Petrus is like, so dead now.

[Petrus_EW]

​

I've always thought this way about blockbuilding, its just hard to explain to anyone that doesn't cube or even that does but uses a non-blockbuilding method.​

you learn to understand the cube beyond most people do, atleast in this aspect and it really feels like you have gained a skill through pure intuition and effort.​

understanding roux and petrus blockbuilding and being effective at both imo means you can basically be good with any type of blockbuilding, this makes using any method alot easier to get used to as you can take block concepts and apply> it to this other method.​

there is an (p)infinite amount of blockbuilding cases possible but they can all be simplified into another for finding an efficient solution.​

I guess "spatial awareness" of the cube is always how i've put it​

​

​

I totally agree with you, only people who get used to or prefer to use block building methods are the ones who can really understand the knowledge that is achieved in the Rubik's cube with these methods. This is why I really appreciate block building methods like Petrus, Roux, or Heise. You can really learn a lot about how the cube works, you can achieve a broader understanding and knowledge of the Rubik's cube. And all this happens because you are doing it intuitively, you are looking for the most efficient way to put it together, and that requires ingenuity and creativity.​

It is for the same reason that these methods are very important in FMC, where a high knowledge of the rubik's cube is required to carry out your solutions in the least number of movements.​

If any of the members of this community ask me if Petrus is dead, my answer is a resounding no.​

It is one of the best methods that could have been created and is also part of the history of Speedolving. Petrus deserves as much recognition as most methods.​

 

[minxer293]

Petrus is a cool method, but it had a lot of problems.

The entire point of speedcubing is to solve the cube fast. Due to how advanced and ergonomically friendly algorithms have gotten, it is ultimately faster to use algorithms as they require less time to think.

Petrus, Roux, etc, are methods that use fewer algorithms and therefore use more thinking, which isn't optimal in the grand scheme of things.

Also, Petrus has an extremely low move count compared to other methods, and during the early days of cubing where hardware was garbage, using fewer moves was better.

Basically, Petrus was phased out due to better hardware and a greater need for speed.

 

[PetrusQuber]

Petrus is a cool method, but it had a lot of problems.

The entire point of speedcubing is to solve the cube fast. Due to how advanced and ergonomically friendly algorithms have gotten, it is ultimately faster to use algorithms as they require less time to think.

Petrus, Roux, etc, are methods that use fewer algorithms and therefore use more thinking, which isn't optimal in the grand scheme of things.

Also, Petrus has an extremely low move count compared to other methods, and during the early days of cubing where hardware was garbage, using fewer moves was better.

Basically, Petrus was phased out due to better hardware and a greater need for speed.

I would argue that the definition of an algorithm is that of a sequence of memorised moves you can use to solve an area of the cube. And it’s generally the case that the more algorithms you learn, the faster you can proceed (such as full OLL and PLL). Even during F2L you’re not ‘learning’ algorithms, but you are still practising common moves that come up during F2L. Ideally as you go into F2L, you’d have a perfect idea of how to solve all 4 pairs and execute one long move sequence without pause to do so. But thats not something we learn as an algorithm, for the simple reason that there are too many of them and too many moves to reasonably memorise properly. Thats why things like 1LLL without EO are supposedly faster but doubted due to the lengthy identification process and execution (and learning 1.5k algs). The best way to solve the cube would be to memorise the countless numbers of solutions to every state in existence, but of course thats unfeasible. So we break down the cube into smaller sections, and use algorithms for even smaller sections within those.
These algorithms, while easy to learn and execute, restrict us, same with method steps. We don’t have total freedom over what we can do more efficiently, and actually, this is why X-Cross is becoming popular. Although it requires more thinking and planning to do, you’re more free to move and come up with better solutions instead of just constraining yourself to the same steps every solve.

Thats why ultimately, I believe at a high level, Petrus could become a feasible method for speedsolving. Blockbuilding has a lot of possibilities, but not nearly so many as we’d like to think, and certainly you can learn hundreds of triggers and subsolutions to steps, even doing so subconsciously. The main argument for Petrus is that while it’s more efficient, the thinking time to execute blockbuilding right now loses more time than the efficiency saves. Solution: we fit blockbuilding more into algs.

I haven’t written a long post in a while, probably tripped up several times there

 

[Athefre]

Thats why ultimately, I believe at a high level, Petrus could become a feasible method for speedsolving. Blockbuilding has a lot of possibilities, but not nearly so many as we’d like to think, and certainly you can learn hundreds of triggers and subsolutions to steps, even doing so subconsciously. The main argument for Petrus is that while it’s more efficient, the thinking time to execute blockbuilding right now loses more time than the efficiency saves. Solution: we fit blockbuilding more into algs.

I've been developing this solution for around a week. It takes care of one of the big weaknesses of Petrus. I also plan to accompany it with an additional improvement plan. The document should be ready to show soon.

 

[Deleted member 51471]

I've been developing this solution for around a week. It takes care of one of the big weaknesses of Petrus. I also plan to accompany it with an additional improvement plan. The document should be ready to show soon.

Can't wait!

 

[PetraPine]

Petrus is a cool method, but it had a lot of problems.

The entire point of speedcubing is to solve the cube fast. Due to how advanced and ergonomically friendly algorithms have gotten, it is ultimately faster to use algorithms as they require less time to think.

Petrus, Roux, etc, are methods that use fewer algorithms and therefore use more thinking, which isn't optimal in the grand scheme of things.

Also, Petrus has an extremely low move count compared to other methods, and during the early days of cubing where hardware was garbage, using fewer moves was better.

Basically, Petrus was phased out due to better hardware and a greater need for speed.

petrus doesn't have "extremely" low movecount its only like around 44~
the hardware argument isn't even an argument against the method lol, just saying it used to be more popular (which is a straight lie because it was most popular with like ~500 users in the mid 2000s long past the hardware would give any advantage to petrus)
the first thing you said is true, but like wrong? everything past inspection (223) in petrus can be done "algorithmically" like f2l, so there is not more thinking required

 

[xyzzy]

Also, Petrus has an extremely low move count compared to other methods, and during the early days of cubing where hardware was garbage, using fewer moves was better.

Using fewer moves is still better now. It didn't stop being better just because you could turn the cubes faster.

 

[minxer293]

Using fewer moves is still better now. It didn't stop being better just because you could turn the cubes faster.

But as hardware became less garbage, ergonomically friendly movesets needed to be implemented more.

 

[xyzzy]

But as hardware became less garbage, ergonomically friendly movesets needed to be implemented more.

Your wording frames "ergonomically friendly" as some kind of objective or universal ideal, when it seems that it's more of something you have to adapt for the specific cube and cuber combination.

Regrips and rotations (mostly) don't rely on any properties of the hardware (size is the main factor, but aside from novelty items every cube has about the same size), so as turning got better, the relative cost of regrips and rotations went up. IOW, trading off regrips for more moves started to make sense… right? But there are two things you're missing from this equation.

Firstly, Petrus never had that many rotations or regrips in the first place. With good blockbuilding, the 222 is roughly comparable to cross, extending to 223 takes maybe one regrip, doing EO might take one rotation, and finishing F2L might take one regrip. This is all comparable to CFOP's 0.5 rotations per pair, and CFOP even has LU-RU grip shifts that Petrus doesn't.

Secondly, as cubes got even better, it turns out that move sets that were originally considered "bad" were in fact… not all that bad actually! See for example the recent resurgence of S-move algs.

My understanding is that people didn't want to use Petrus for the same reason they didn't want to use Roux: learning good blockbuilding is difficult and essentially a never-ending journey. Roux eventually won out as the blockbuilding method to use, especially since it has competitive performance without needing to learn hundreds of long algs, but in a parallel universe it wouldn't've been strange to see Petrus win out instead.

 

[PetraPine]

Your wording frames "ergonomically friendly" as some kind of objective or universal ideal, when it seems that it's more of something you have to adapt for the specific cube and cuber combination.

Regrips and rotations (mostly) don't rely on any properties of the hardware (size is the main factor, but aside from novelty items every cube has about the same size), so as turning got better, the relative cost of regrips and rotations went up. IOW, trading off regrips for more moves started to make sense… right? But there are two things you're missing from this equation.

Firstly, Petrus never had that many rotations or regrips in the first place. With good blockbuilding, the 222 is roughly comparable to cross, extending to 223 takes maybe one regrip, doing EO might take one rotation, and finishing F2L might take one regrip. This is all comparable to CFOP's 0.5 rotations per pair, and CFOP even has LU-RU grip shifts that Petrus doesn't.

Secondly, as cubes got even better, it turns out that move sets that were originally considered "bad" were in fact… not all that bad actually! See for example the recent resurgence of S-move algs.

My understanding is that people didn't want to use Petrus for the same reason they didn't want to use Roux: learning good blockbuilding is difficult and essentially a never-ending journey. Roux eventually won out as the blockbuilding method to use, especially since it has competitive performance without needing to learn hundreds of long algs, but in a parallel universe it wouldn't've been strange to see Petrus win out instead.

I think the main reason roux won is because
A.The blockbuilding is easier, mainly due to FB being shorted and relying on more basic 2d blocks
B.Roux is much more algorithmic, during sb+ which cfopers see and recognize as a good thing.
C.Roux was a method invented on forums much later than petrus, leading to it having a more consistent interest as it was new.

 

[Cubing Forever]

Tao Yu got a sub 9 ao12.

Read the description lol

 

[PetrusQuber]

Tao Yu got a sub 9 ao12.

Read the description lol

Interesting!

 

[6 by 6]

I use Petrus, and I have noticed nobody is really using it anymore. I, along with a couple of other people, am like the solitary Petrus user on this forum which is still active. The majority of users seem to be CFOP, Roux, or ZZ . So there isn't much discussion on how to improve on specific areas here for me. What are your thoughts on this? Why has usage of the Petrus method dwindles? Anymore people using Petrus?

yes, i use petrus method, I’ve recently posted a detailed guide on the Petrus Method for Big Cubes from 4 by 4 to 20 by 20 in my thread — feel free to check it out in general discussion! i have made a pdf and experimented this method in 9 by 9.

 

[FenTheGoat]

yes, i use petrus method, I’ve recently posted a detailed guide on the Petrus Method for Big Cubes from 4 by 4 to 20 by 20 in my thread — feel free to check it out in general discussion! i have made a pdf and experimented this method in 9 by 9.

I don’t want to be rude, but I feel like a lot of people would be ruder than me trying to tell you this. Please don’t comment on threads that haven’t been commented on in years. It’s against forum policy. Like this thread was last touched over 4 years ago. That’s all!

 

[efattah]

So what happens when other people want to chime in as well? Is the thread still dead then?
I think the definition of a thread is a bunch of people exchanging ideas.

 

[Deleted member 36445]

"Petrus is dead" is a fallacy

Any knowledge you have on methods developed before the era of modern speedcubing can add you great knowledge.

You can't get good on FMC or cube theory in general without knowing the basics of blockbuilding and edge control.

We won't be seeing a huge percentage of people choose Petrus as their main method but the method is still useful and we can learn a lot by studying it

 

[6 by 6]

Thanks everyone for the responses! I really appreciate the clarification -I didn’t realize the thread was that old, so that’s my mistake. I’ll keep that in mind for future posts.
Also, I completely agree that Petrus still has a lot to teach us about blockbuilding and efficiency. That’s exactly why I experimented with it on big cubes and shared my guide — there’s so much depth in its logic that can still inspire new methods today. My main goal is just to share the Petrus Method for Big Cubes guide so more people can try it and experiment with it. I think it’s fun exploring how Petrus-style blockbuilding works on larger cubes, and I’d love to hear what others discover too!

Thanks again for the insights and for keeping the discussion constructive!

 

[Michael Wimberly]

I use a variation of Petrus, just not as my main method (it's CFOP).

 

[6 by 6]

I use a variation of Petrus, just not as my main method (it's CFOP).

oh, i also use 2 look CFOP in big cubes, I practice Petrus method rarely, but its a great method.

 

[Caleb Valenzuela]

Is APB a variation of Petrus?