Older cubers discussions

Shaky Hands said:
I have a general question which I guess applies to all puzzles but particularly to big cubes. When you can't find a piece you're looking for (whatever the reason) do you tend to persist in finding it as you know it's there somewhere, or do you search for alternatives?
Click to expand...
Depends on the situation, but I have deliberately chosen my methods to make it easier, either to find pieces by having a smaller set of pieces to search and/or the search space more visible, or to let me solve the first thing I see rather than specific pieces.

Take 5x5. I use Hoya and during cross I solve one piece at a timee, solving the first cross piece that I see and keeping going that way, in any order. Only on the last piece am I looking for that one specific piece.

On pairing the other edges, I am usually looking for a specific piece, but I have the bottom and back slots filled so my search space is all easily visible without rotation and I use a pairing method that requires me to search for midges not wings, so there are only seven spots I have to scan.
 
mafergut said:
If somebody was trying the 11.50 scramble and thinking how on earth I got a PLL skip I tried to reconstruct and it took me a while until I realized I had made a stupid mistake during cross. Without that mistake and others it could have been a sub-10 but, again, without that mistake it would have ended on a U-perm so probably the final time would have been similar or worse.

Now that I have the reconstruction I'm ashamed that this execution disaster is actually my PB single:
Click to expand...

To me this sounds like an amazing solve :) I just tried the same scramble in VERY SLOW motion and still used 72 moves instead of 40... oh and at least 3 more cube rotations than you did :)

Technical question : how do you figure out reconstructions ? Is it based on a video or your memory of your solve ?

Second rookie question : for your fourth pair, I would have used y'F'LFL'F'LFL'. Since it's a trigger reapeated twice I can perform it fast and tend to use that strategy (this alg or its mirror) all the time when I have edge on U + corner on D with white sticker and edge on the same face. Can anyone discuss if this is a good strategy or bound to be slow in the long run ?

P.S. Talking about luck : my new 24.95 s PB happened with an OLL/PLL skip. What are the chances of that ?!?
 
Last edited:
teacher77 said:
To me this sounds like an amazing solve :-) I just tried the same scramble in VERY SLOW motion and still used 72 moves and at least 3 more cube rotations than you did :-)

Technical question : how do you figure out reconstructions ? Is it based on a video or your memory of your solve ?
Click to expand...
Typically memory. I have only reconstructed a couple including my amazingly lucky pb of 12.83. :) Normally it's easy to find the cross you started with, and from there not too hard to remember what came next. It's kind of fun you should try it.

Btw, I tried my pb a few times and I think the best I could manage was like 11 seconds, and that's basically knowing the whole solve. Of course I remember a time when I practiced someone's easy scramble and couldn't break 16. This was a while back, maybe I should look it up and see if I'm faster. :)
 
Hey everyone, I'm currently switching from 2look PLL to 1look PLL and I'm having a hard time with recognition more than alg memorization. Is there any way (apps ?) to practice recognition other than just solving full cubes ? I find solving inefficient as some cases only show up once in a day so it hardly does anything to get things in my brain... Can some of you please tell me how you first did it ?
 
teacher77 said:
Hey everyone, I'm currently switching from 2look PLL to 1look PLL and I'm having a hard time with recognition more than alg memorization. Is there any way (apps ?) to practice recognition other than just solving full cubes ? I find solving inefficient as some cases only show up once in a day so it hardly does anything to get things in my brain... Can some of you please tell me how you first did it ?
Click to expand...
I think there are apps, some written by guys here. I would actually just drill the alg on a solved cube. If my f2l was not messed up then I basically knew I did it right and I'd keep going. If I messed up I would solve the cube and start over. Some of them if you do twice it solves the cube again, others you can do multiple times and it will re-solve the cube (all I think?).
 
teacher77 said:
Hey everyone, I'm currently switching from 2look PLL to 1look PLL and I'm having a hard time with recognition more than alg memorization. Is there any way (apps ?) to practice recognition other than just solving full cubes ? I find solving inefficient as some cases only show up once in a day so it hardly does anything to get things in my brain... Can some of you please tell me how you first did it ?
Click to expand...
I learned quite early to recognise from two sides only, because it's faster, and in the long run that's what you'll want. I put together the guide linked below. In there is also a link to Rishi's tool I used for practising recognition.

https://www.speedsolving.com/forum/threads/two-sided-pll-recognition-guide.41108/

Having said that, you do need a basic knowledge of the overall patterns on all four sides first, and I can't think of a good resource for that, beyond things like the pictures on the wiki. For example, the two sided guide doesn't distinguish between mirrored cases like Ja and Jb so you need to learn which is which. And you need to learn how to orient the case for execution, which will be helped by knowing what's on the back sides even though you only recognise from the front.

It might be more advanced than you were looking for, but anyway, Rishi's tool is what I used for practice.
 
teacher77 said:
Hey everyone, I'm currently switching from 2look PLL to 1look PLL and I'm having a hard time with recognition more than alg memorization. Is there any way (apps ?) to practice recognition other than just solving full cubes ? I find solving inefficient as some cases only show up once in a day so it hardly does anything to get things in my brain... Can some of you please tell me how you first did it ?
Click to expand...

I learned a lot from CrazyBadCuber's tutorial on PLL recognition:

 
@teacher77 Thanks but it is a bit sad that I started the solve with a cross mistake and the only reason this is a PB is because that mistake ended up in a PLL skip :) but I'm not going to complain.

Regarding the 4th pair, I tend to avoid F2L algs based on repeated triggers (e.g. the triple sexy one for edge in place with corner on top with cross color up) because if I execute them fast I will easily lose my lookahead and if I execute them slow they are usually not better in move count than other more intuitive solutions. I also try to avoid F moves as much as I can, because they are awkward to execute for me. In this case I prefer what I did, which is 2-gen and comfortable for me but your solution is nice so keep doing that if it works for you.

Regarding PLL recognition I didn't do anything specific. Just spam the alg over and over. Most of them have a 2 cycle, others like A&U-perms a 3 cycle. G-perms are a bit more confusing but can be practiced in pairs. If what you want is getting a random PLL every time and see how well you recognize it you can use a PLL scramble like those on csTimer. I also recommend Mark's 2-sided PLL recognition guide. It's the best you'll find. I tried to learn it last Christmas and, even though I didn't fully integrate it in my solves, it has improved a lot my recognition from 2 sides (I still need a 3rd one now and then).
 
Last time I looked at Mark's PLL recognition tutorial was before I learned full PLL and it seemed quite complicated then. Had another look now and it looks a lot more manageable. :) I shall try to learn it all now.

Yesterday I got a bundle of Traxxas 10k, 30k and 50k and did some experiments with those. First victim was my Thunderclap, which felt odd and had pretty bad spring noises out of the box. I lubed the core with 50k and put some 10k on the pieces. The spring noises went away and it became a very nice cube. Reminds me a lot of the Aolong v2. I need to do some more solves with it to find out if I can end up really liking it.

Next up was a Yuxin I use for blind. Again I used 50k for the core, but since I didn't want it quite as fast I tried using 30k on the pieces. My initial reaction was that I had put way too much lube and ruined it. It was really slow and sticky immediately after lubing. I didn't give up, kept working in the lube by playing around with it for about an hour while watching TV and eventually it got better. I can still feel some of the stickiness and also hear a sticky sound while turning. Not sure if I like it or not. If I end up not liking it and want it less sticky, do I need to clean off all the old lube from the pieces before relubing? Or could I try by adding some 10k to see if it becomes faster?

Anyone else here use 30k for lubing? If so, how do you use it? How much to apply and where?
 
@JanW if you have gummed it up, don't put more in, not even lighter lube. Just take off the top layer and wipe the pieces off, then reassemble and work the remaining lube around, and repeat until you like it.
 
Well, I have improved but not as much as I would have wanted. Anyway, some nice inspiration in this thread. Not sure you have had time to watch Chris' sub-1min Ao5 video from yesterday. That's what I really aim to but I still feel so far from it. Regarding 5x5 I don't seem to improve since 1 month ago but now I have some weeks of vacation where, on top of swimming pool and relaxing, I plan on putting some big cube practice time.
 
mark49152 said:
I learned quite early to recognise from two sides only, because it's faster, and in the long run that's what you'll want. I put together the guide linked below. In there is also a link to Rishi's tool I used for practising recognition.

https://www.speedsolving.com/forum/threads/two-sided-pll-recognition-guide.41108/

Having said that, you do need a basic knowledge of the overall patterns on all four sides first, and I can't think of a good resource for that, beyond things like the pictures on the wiki. For example, the two sided guide doesn't distinguish between mirrored cases like Ja and Jb so you need to learn which is which. And you need to learn how to orient the case for execution, which will be helped by knowing what's on the back sides even though you only recognise from the front.

It might be more advanced than you were looking for, but anyway, Rishi's tool is what I used for practice.
Click to expand...

RiDo's tool seems nice but I can't seem to get it working. I tried both in Excel 2013 and in Excel 2003 and either will make it crash. How did you run it ?

And the guide is great too but definitly advanced :-)
 
You must log in or register to reply here.

Similar threads

Burrito
Any older kid/teen cubers in Durham Region, Ontario?
Replies
0
Views
796
Burrito
Burrito
Jilli0
  • Locked
Any Teen/Older Kid Cubers in Wayne County, Michigan?
Replies
11
Views
2,508
Deleted member 74176
D
Mr Cubism
How many older members (registered before 2010) are still active here?
Replies
1
Views
102
abunickabhi
abunickabhi
O
[Member Intro] Older beginner in scotland
Replies
17
Views
3,175
Cuberto333
C
Humble Cuber
Selling Older Puzzles | 2-5's & Megaminx's
Replies
6
Views
2,268
Humble Cuber
Humble Cuber
Back
Top