• 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 35,000+ people from around the world today!

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

My quest for Sub 8 Petrus

Should I do some example solves every once in a while?


  • Total voters
    67

PetrusQuber

Member
Joined
Jun 27, 2019
Messages
839
Location
my house, cubing.
WCA
2020XIEC01
Ok so I’m going to move on to finishing F2L this week, and learn the last of the H COLLs. After that, if F2L turns out OK, I might review my blockbuilding in Steps 1&2, and also focus on inspection usage. I should probably learn how to do COLl recognition from at least two angles as well.

Quick question, is it worth having the ability to choose an easy 2x2x1 from 4 of the different options when finishing F2L? Or should I just use the 2 on the R layer?
 

PapaSmurf

Member
Joined
Jan 4, 2018
Messages
467
WCA
2016TUDO02
YouTube
Visit Channel
Definitely do block in back for EO and initially build block in back prefereably. So solve 2x2x2 in BDL then, unless the 2x2x1 on F is really good, solve the 2x2x1 on R. That sets you up really nicely for EO and will allow you to lookahead much more easily. I would also work on drilling your algs. If you can't sub 2 all of the COLLs on that sheet, you should work on that. Same with PLL. That will then benefit the rest of your solve as you will have better fingertricks and tps etc. And definitely learn COLL recog from 2 angles (U2 away from each other).
 

PetrusQuber

Member
Joined
Jun 27, 2019
Messages
839
Location
my house, cubing.
WCA
2020XIEC01
After I finish COLL and 2 sided recognition, I think I’m going to be learning ZBLL. Since a quick search of ZBLL recognition yielded not much, I was thinking, how should I take the recognition? I was thinking COLL recognition, then looking at the UFR corner and it’s surrounding edges - no idea what it’s called.
 

Aerma

Premium Member
Joined
Apr 1, 2017
Messages
1,312
Location
Galar Region
WCA
2015MANN01
YouTube
Visit Channel
@Aerma, how do you do ZBLL recognition?
I've only learned 5/6 of Pi by now so I can only really speak for that I hold the cube so the headlights are on the left. I recognize the COLL case, then look at the 3 R stickers and depending on the case, URF/URB, to figure out what subset of the COLL I have. For example, if I have Pi 2GLL and there's a 1x2 block on the back on R, that narrows it down to 3 ZBLL cases. Then, I look at the two stickers on F and, depending on the case, I might look at UFL or UFR. For Pi, that's always enough to narrow it down. I might make a video or something once I finish Pi to go through every case and explain my recognition process, might make more sense that way.
 

Tao Yu

Member
Joined
Jun 29, 2011
Messages
1,071
Location
Ireland
WCA
2012YUTA01
YouTube
Visit Channel
EO with Block in back vs block on left is an interesting question. Up until now, pretty much all fast Petrus solvers have done block in back. Block on right had been thought of as bad since we didn't know any good algs for it. Now though, as some people have posted, there have been some decent algs found for block on right.

I've not heard of anyone who has gotten good at block on right to the point where it's faster than block in back, but I think it's something worth looking into and something you should experiment with yourself. Sub 8 is after all unexplored territory, and you should be prepared to try things people have not tried before, and do your own analyses.

I don't think there has been a unanimous agreement on how best to recognize ZBLL. My way is to is use COLL recognition and then pay attention to easy to recognize patterns such as the ones here. Then, once you have learned a basic kind of recognition, I think 2-sided recognition could be learned by doing a lot of training, perhaps by grouping similar cases together. This is much like how many people learn 2 look PLL without really trying, just by doing a lot of solves, except I feel that more targeted practise might be needed for a set as big as ZBLL.

I'm sure that not everyone will agree with this method, but I'm just posting it so that you can have a think about it and decide if you like this method or other methods. As well, ZBLL is a big set: you can experiment with several different recognition methods before you are even finished and see which works best for you. In fact I think it is very important to constantly adjust your strategy while learning ZBLL to account for discoveries you've made regarding how best you learn - it is quite difficult to get it right from the very start.
 

PetrusQuber

Member
Joined
Jun 27, 2019
Messages
839
Location
my house, cubing.
WCA
2020XIEC01
Thanks! So COLL, then just looking for patterns and blocks like in full PLL?
About the block in back, I feel R U/F L gen is much better than F U/R B gen, but there leaves much to be researched.
 

Tao Yu

Member
Joined
Jun 29, 2011
Messages
1,071
Location
Ireland
WCA
2012YUTA01
YouTube
Visit Channel
Thanks! So COLL, then just looking for patterns and blocks like in full PLL?
Yes, that is an accurate summary of what I said. As I said, it's just my opinion on the best way to do approach ZBLL recognition.

I think that using easy to recognize patterns like the ones in the document I linked should be relatively uncontroversial, I don't think it makes sense to ignore them even if you primarily use a different recognition method, and it's kinda hard to ignore them anyway once you've seen them. What people might disagree on is how to eventually learn 2-sided recognition, some people might argue that you should learn a system that allows for 2-sided recognition from the very start (using a system like the one on the wiki), which my proposal would not do. My counter to this is that doing it my way ensures that your brain picks up on the patterns that are naturally easy and fast to recognize.

About the block in back, I feel R U/F L gen is much better than F U/R B gen, but there leaves much to be researched.
While movegroup is a good informal way to decide which method is better, it's important to remember that at the end of the day, it's the time that matters. The deciding factor is whether block on right algs are on average faster than block in back algs (taking the rotation for block in back into account) - movegroup doesn't matter if you can simply do the algs for one faster. It's pretty likely that the fastest block on right alg are not mostly FURB, as you can see, a few people have posted algs that include r moves.
 

PapaSmurf

Member
Joined
Jan 4, 2018
Messages
467
WCA
2016TUDO02
YouTube
Visit Channel
I think he means block on left, not right, but it could be good, but it would require optimisation for it to be faster. It could be the thing where you need less work to be good at EO with block in back but overall it's slower but you need more work for block on L, but overall it's faster.
 

PetrusQuber

Member
Joined
Jun 27, 2019
Messages
839
Location
my house, cubing.
WCA
2020XIEC01
Hmmm, with all this discussion going on, I’m going to make a debate thing here, and also list pros and cons, e.g. no rotation needed, fingertrickier, etc.
———————————————————————————————————————
Block in Back
Pros: Fingertricks come easily
After alg, no regrip needed

Cons: Needs rotation after for RU F2L
Certain wide move algs require Fs
———————————————————————————————————————
Block on right
Pros: Doesn’t need rotation after for RU F2L
Certain wide move algs are better (r vs f)

Cons: Triggers like F R F’ leave you not in home grip
Arguable whether F is slower than L and B is slower


Please let me know about better arguments, why certain arguments are invalid, new arguments, etc.
 
Last edited:

PetrusQuber

Member
Joined
Jun 27, 2019
Messages
839
Location
my house, cubing.
WCA
2020XIEC01
Just discovered M‘ U M U2 M’ U M and all it’s mirrors, reverses, and translations :). For people who dislike using F and B. Even faster than H Perm (sub 1 alg)
(Block on right)
 

xcross

Member
Joined
Dec 24, 2019
Messages
6
Location
VC, Canada
Hey! Thanks for replying to my thread! I really think this is a great project to see if petrus can get a low time! I just watched a few videos on petrus and I think its a great method! Good luck!
 

PetrusQuber

Member
Joined
Jun 27, 2019
Messages
839
Location
my house, cubing.
WCA
2020XIEC01
I broke my average PB a couple days back, with a sub 20 times :). Nice lookahead and TPS combo, which lead to extremely good solves. I think there was a 16, counting 18, and the rest were 20s and 21s. Next two weeks will just be practising mindlessly because I’ve got a competition coming up and would like to be sub 20 solidly by then (which I’m sure is possible since I just got a sub 20 average). Plus, learning new things is not good for short term improvement.
Happy cubing! :)
 
Top