• 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!

Advanced Petrus / block building

Joined
Jun 20, 2010
Messages
8
Likes
0
Thread starter #1
Hi!

First some background: I've been cubing for a while now and even though I've learnt some other methods, I still use Petrus for F2L + EO, which gives me just a few OLL cases and then full PLL, which I know. With this, my best average is roughly 27 seconds.

I feel that what I need to improve on LL is mostly recognition and execution speed, both which will come with more practise. For F2L, however, I would be interested in tips, ideas or hints on how to get faster. I have understood and implemented most of the basic block-building stuff on Petrus' own website, but I feel that that is not enough and I know there have to be more than that.

So, over two my question:

Are there any other websites, threads or anything else that might help me improve my block building? Does anyone have any other suggestions on how to improve from here?
 
Joined
Nov 24, 2009
Messages
1,256
Likes
2
Location
Florida, in that one place...
#2
Hi!

First some background: I've been cubing for a while now and even though I've learnt some other methods, I still use Petrus for F2L + EO, which gives me just a few OLL cases and then full PLL, which I know. With this, my best average is roughly 27 seconds.

I feel that what I need to improve on LL is mostly recognition and execution speed, both which will come with more practise. For F2L, however, I would be interested in tips, ideas or hints on how to get faster. I have understood and implemented most of the basic block-building stuff on Petrus' own website, but I feel that that is not enough and I know there have to be more than that.

So, over two my question:

Are there any other websites, threads or anything else that might help me improve my block building? Does anyone have any other suggestions on how to improve from here?
For blockbuilding, I find that doing slow untimed solves to find the most efficient block is the best way to get better at it.

Reading won't get you better at something, practice will (in my opinion)
 
Joined
May 31, 2010
Messages
303
Likes
0
#3
(http://www.speedsolving.com/forum/showthread.php?27293-How-to-make-your-thread-look-useful)

Jk, but what I should say is that you really need to post your breakdowns if you want advice, because one person might have a decent last layer, but have a 20 second F2L, whereas another might have a really crapp LL but have a thirteen second F2L.
Unfortunately, I can't recommend any good websites for block building, although I suspect that even the more experienced Petrus users here will tell you that for Petrus, most of it is about being figured out by yourself. I could be wrong though, because I really don't know much about Petrus.

Good luck anyway- block building is cooler than Fridrich, IMO. ;)
 
Joined
Jun 20, 2010
Messages
8
Likes
0
Thread starter #5
Some more information, then!

I timed ten solves to get some idea of F2L/LL speed (of course not too accurate, but it should say something). I average 22.9 on F2L, which is probably really slow. The average total time for these solves was 30.6, which gives an LL average of 7.7 seconds (it should be a bit lower, because my recognition is a bit disturbed by glancing at the clock). I realise that F2L is my problem and timing it just confirms that.

I'm colour neutral, yes, and I don't think it's the 2x2x2 that's the problem. It's hard to count average for 2x2x2 when going for speed, but after trying some sample openings, it's definitely below 10 and I can usually visualise the whole sequence leading up to 2x2x2 before starting. I think my problem is extending the 2x2x2 along with fixing the last two pairs (edge orientation is quite fast, most of the time).
 
Joined
May 31, 2010
Messages
303
Likes
0
#6
Unfortunately, as I do not use Petrus, I can't give you advice on the individual steps of F2L, but I would agree with you that F2L is where the work should get done. For some immediate goals, I would say that you should aim for a fifteen second F2L. Once you get to that point, the last couple seconds before sub-20 should probably focus on both F2L and LL, because a seven second LL is a little slow, tbh. If you're aiming for sub-20, I would say get it down to five.
What LL system are you planning on using in the future, btw? I'm assuming that you're planning on sticking to OCLL / PLL, but I'm just making sure.
 
Joined
Jun 20, 2010
Messages
8
Likes
0
Thread starter #7
I think I should just stop timing for a while and focus on examining the various F2L steps and see where I can improve. I think I can gain quite a few seconds by just trying to look ahead more and find pieces before I actually need them. Regarding LL times, I think I'm a lot faster if we ignore the G-perms. I sometimes need a couple of seconds to recognise these cases, so some practise there would definitely be of help. I plan to continue using OCLL / PLL, yes, and sub-20 seems to be a reasonable goal.
 
Joined
May 31, 2010
Messages
303
Likes
0
#8
OCLL is good- just don't feel like it's the only thing that can be fast. Whenever you feel like it, look into other LL systems- I would suggest looking at COLL, WV, and CLS (particularly SS) first.
Good luck, and hopefully someone who knows more about Petrus can give you more specific advice.
 
Joined
Nov 24, 2009
Messages
1,256
Likes
2
Location
Florida, in that one place...
#9
Try to get to the point where you can execute 2x2x2 phase without looking, then try to lookahead to a premade pair that would fit, or an easy case for one of the 3 block spots you can place it.

I seem to be finding this very hard to describe...
 
Joined
Mar 23, 2008
Messages
1,871
Likes
26
Location
KC
WCA
2010POTT01
YouTube
rpotts1
#10
L' U' F2 B2 L D2 B' F2 R2 L' U' B F' U' B2 D U B F2 L' B2 L' B F D'

How would you solve this? Specifically the blockbuilding/EO steps, we can maybe show you some better examples.
 
Joined
Jul 8, 2010
Messages
298
Likes
0
#11
Most Roux guides suggest that you practice being efficient, and Petrus is fairly similar. Do a lot of untimed solves and aim for low move counts up to the EO step.
 
Joined
Jun 20, 2010
Messages
8
Likes
0
Thread starter #12
I actually think my original post was about misleading, mostly because I didn't really understand the problem. Reading your replies have set me on what I think is the right track and I now at least where the problem is and I have some ideas what to do about it. I'll describe my thoughts briefly.

First and foremost, the problem is not finding smart solutions. I have done FMC before and although I'm not good at at it, I can usually find ~35 solutions and I don't think blockbuilding is my problem there. Rather, the problem is recognising these cases while going for speed. I think I've been suffering from the classic look -> burst of speed -> look burst of speed -> look problem. I really need to learn to look a head, especially when expanding the 2x2x2 and the step right after fixing EO, ie.e. extending the 3x2x2 block.

I think there are three ways to do this. First, I can try metronome solving, which I really suck at doing. I just tried doing 50/minute and I can't managed F2L without having to skip 2-5 beats when I don't find the pieces I want. That's probably really bad, right? Second, I'll try solving the 2x2x2 blind more so I can execute that with my hands while looking for the next pair to extend, which should make the transition smoother. Third, I can just play around with F2L without thinking about speed. I just think I need to not start my cube timing program for a month and see what happens after that.

Does anyone have any other suggestions? If not, I'll get back to you in a month or so and tell you how it has turned out. Thanks for all the helpful input!
 
Joined
May 31, 2010
Messages
303
Likes
0
#13
Well, ZZ is a blockbuilding method, and even though there are a few key differences, maybe what I say can help you.

When I'm doing a ZZ solve, looking ahead for pieces gets easier as the solve progresses because as you complete more blocks, there are fewer places where the sneaky bastards can be hiding. I assume that it's similar for Petrus.
For the first pair, though, I often find that the position of pieces is much more important than how many moves it'll take them to solve. Take, for example, this F2L case; It's a misconnected case, which makes it a pain, move-count wise. But it's also highly visible- the two pieces that I'm pairing up are right there, in plain view. There might be a 1-move pair elsewhere on the cube, but oftentimes I find that it's simply not worth it looking for it if I can instantly recall the alg for this case, execute it quickly, and move on. There's a happy medium to be found between speed of recog and efficiency, IMO.

What I'm saying might help you. It also might also be the result of my own ignorance, though, or (more likely) it might be more applicable to ZZ than Petrus. God knows that I've seen a lot of really fast users of blockbuilding methods stress efficiency. I guess all I can say for sure is that efficiency is ideal, but with *a* blockbuilding method, valuing speed of recognition just a tiny bit more than pure efficiency *can* produce *at least* mediocre times (14 - 17 seconds).
 
Joined
Jun 24, 2010
Messages
3,921
Likes
6
Location
Tucson, Arizona, USA
WCA
2011THOM01
#14
because a seven second LL is a little slow, tbh. If you're aiming for sub-20, I would say get it down to five.
:( I guess I need to work on LL a lot, because I wasn't sub-7 LL when I got sub-20, and now, confidently sub-17.5, I definitely do not average 5 seconds on LL

Well, I guess using myself as an example can 100% back up everything people say about practicing F2L being the most important thing. With a method like Petrus, doing really slow practice solves to work at lower move count is the best thing you can do when getting used to blockbuilding.
 
Joined
Jun 20, 2010
Messages
8
Likes
0
Thread starter #15
First a small update: I've been doing around 25 solves with a metronome and it turns out I really, really suck. I need to skip on average 6 beats on F2L, but nothing for LL except sometimes a single beat to have time to identify a G-perm. And this is done at 50 beats/minute! I've looked around on the forum and I see people averaging 3x3x3 at around 20 seconds who say they can keep up with 100 beats/minute! I don't know if they're incredibly good or I'm incredibly bad, but the latter is probably more likely.

Anonymous said:
When I'm doing a ZZ solve, looking ahead for pieces gets easier as the solve progresses because as you complete more blocks, there are fewer places where the sneaky bastards can be hiding. I assume that it's similar for Petrus.
Yes, this is indeed the case. I have no or little trouble looking ahead once I have a 2x2x3 block, because, as you say, I can see all pieces quite easily. I also agree on what you say about visibility. My "fast" ordinary F2Ls usually come when I don't need to look for pieces during F2L because they are plainly in sight.
 
Joined
May 31, 2010
Messages
303
Likes
0
#16
:( I guess I need to work on LL a lot, because I wasn't sub-7 LL when I got sub-20, and now, confidently sub-17.5, I definitely do not average 5 seconds on LL

Well, I guess using myself as an example can 100% back up everything people say about practicing F2L being the most important thing. With a method like Petrus, doing really slow practice solves to work at lower move count is the best thing you can do when getting used to blockbuilding.

Wow, so you average, what, 10 second F2L?
 
Top