Older cubers discussions

[kbrune]

@mark49152 @h2f

I'm not even sure how to push my memo beyond where it is now. So far when I push memo faster. The info doesn't stay in my mind.

Is it possible that my brain just won't be able to retain the information?

A better way to explain this idea for me is this. So far. To be able to recall my targets. I have to repeat them a few times as I memo them. (I use images for corners and phonetic words for edges).
To speed up the memo I'd have to reduce the repetition meaning I'd have to settle for either one or 2 looks at most for each pair so to speak, rather then repeating them several times. I've been able to cut down repetition on the first few pairs because I end up repeating them anyway as I go. So usually the fist 2 or 3 pairs I don't repeat at all. Then as I add more I repeat the whole sequence in my mind before adding the next piece. Of course there's way more that goes on but this is a simplified account of what I do.

Is there a better way for me to look at this whole thing? Perhaps I need to look a better memo methods for single solves?

Apologies for being a broken record on this topic. If it really does come down to just pushing memo speeds and forgetting it all is part of the process of pushing your brain to be capable of one looking pairs. I will keep plugging away. Memo just seems to be so difficult to figure out how to improve for me.

@newtonbase
What's you bld times like? Are you focusing on the event?

 

[h2f]

Is it possible that my brain just won't be able to retain the information?

A better way to explain this idea for me is this. So far. To be able to recall my targets. I have to repeat them a few times as I memo them. (I use images for corners and phonetic words for edges).
To speed up the memo I'd have to reduce the repetition meaning I'd have to settle for either one or 2 looks at most for each pair so to speak, rather then repeating them several times. I've been able to cut down repetition on the first few pairs because I end up repeating them anyway as I go. So usually the fist 2 or 3 pairs I don't repeat at all. Then as I add more I repeat the whole sequence in my mind before adding the next piece. Of course there's way more that goes on but this is a simplified account of what I do.

In my case that was the other way. First - how long does it take for you to go through all targets? For example for me, when my memo was around 2 minutes it took me around 1:20. Next I was repeating it several times and went to corners which I did with audio pairs and I needed less time to remember. Now it takes me around 8-10 seconds at best moments and if I push and I repeat it (or imagine the image) once or two and go to corners. If I remember it well I did push sessions few times when my memo was around 1 minute but the main factor to speed up the memo proces was that I was comofortable with my letter scheme and my images/words plus few techniques like touching edges with left and right hand fingers. I think the biggest factor is just regular practice and pushing your memo when you feel comfortable. How many solves do you do per day or per week, if I may ask?

 

[mafergut]

In fact, here's my list, if anyone would care to comment on the order or the relevance, etc. :
1) master a satisfying alg for all 21 PLLs
2) master cross on bottom
3) improve PLL recognition (learn to recognize from 2 sides)
4) Other stuff, in whatever order I'll find relevant by then :
- learn some of the easy OLLs (I know 12 of them only)
- learn how to make an X-cross
- improve my lookahead (currently able only to find a corner piece ahead or a single pair at best)

I would re-order: cross on bottom, good 21 PLL algs, lookahead, learn more OLLs, x-cross, 2sided PLL recog,

 

[mark49152]

@kbrune - yes it takes hard work, just like improving at any event. To speed up memo you need to read the targets from the cube faster, map them to images quicker, form those into a memorable scene quicker, get it to stick and move on to edges quicker, trace the edges and form audio words quicker, and avoid reviews. Trying to go faster is really quite uncomfortable, but you do need to push beyond your comfort zone.

One technique I use is to set a countdown timer for say 30 seconds. Start timer and rush memo, and when the timer beeps, close eyes and execute. If you haven't finished memo, tough, it will be DNF, but execute as much as you can anyway. Having that time limit puts pressure on the memo and forces you to rush it.

 

[pglewis]

I've done about an hour to 90 mins a day of timed solves at the coffee shop this week for distraction practice. I think it was a good decision, I could feel a difference in my concentration the past two days vs. Tuesday. 3x3 times were about 10 seconds slower than normal today, though, just one of those days. Yesterday was fairly good including a couple more sub 30s.

The plan to start absorbing new OLLs next week may have hit a serious challenge: I just got a mirror blocks cube as my first shape-shifter. Totally fascinating.

 

[SenorJuan]

HeHe, fascinating indeed. Once you get the basics sorted, you have to try it unsighted, by touch alone. Rather satisfying when it's done for the first time.

 

[newtonbase]

@newtonbase
What's you bld times like? Are you focusing on the event?

I have a 1:32 PB but average around 2:20. Any sub 2 mins makes me happy at the moment. It's definitely my favourite event but I don't get to practice anywhere near enough.
@pglewis The mirror blocks is my favourite shape shifter. I have the solid pink Yuxin (amongst others) which is a pleasure to solve. I agree with @SenorJuan about no inspection blind solving. You must try it.

 

[muchacho]

I don't know why but today my OH times (now averaging 33-34, last days was 36 or more) are too good, wondering if maybe sub-25 by year's end was not so far-fetched after all #keepdreaming

New PBs
Single: 21.543
Ao5: 26.883
Ao12: 29.546

 

[Lid]

First PB of the year. SQ-1 of cause
(12.225), 12.674, 14.781, 16.040, (18.315[p]) = 14.498
which almost also became a PB for a12; (12.225), 12.674, 14.781, 16.040, 18.315[p], 18.861[p], 17.460, (20.114[p]), 15.350, 16.913[p], 16.125, 16.838 = 16.336

 

[pipkiksass]

Got my old Valk 3 (white) out because I was hiding one of my daughter's toys behind a mountain of cubes (fun!), and decided to do a comparison of my times on the white and stickerless (main). Second solve on the stickerless was a NL PB!

12.01 - F2 D F2 U' L2 U2 R2 D F2 U' F2 B U' L2 R D R B D B2 - nice scramble - 3 move cross. I think this shows how much difference a good cross makes to my solves.

Back to learning full OLL now. For any who are interested, I've decided on a change of method. Rather than learning 1/2 algs at a time, I've gone for what I did when learning full PLL many years ago - I've used @Logiqx DIY sheets, deleted the cases I know. I've replaced a couple of algs, leaving all of my favourites from 4 years of scribbles, and a very neat and very select group of OLLs. I'm annotating with directions (such as 'bar on left', etc.) for recognition and execution.

The way I figure is that when I learned full PLL I didn't care so much about times - 45 seconds was a good solve, so I didn't mind stopping and looking up an alg mid-solve. I'd like to get back to that, so when it comes to learning them, I'm just slow solving (other than this cube vs. cube comparison) with my alg sheets in front of me, and using the sheets when I hit an unknown case. Over time (as with PLL many moons ago), I'll use the sheets less and less, but the absence of time/timer pressure means that I can just solve and refer to the sheets without worrying about messing up a good solve. I'm taking the opportunity also to practice cross + 1 and general lookahead, intending to do very few timed solves until I can recall the vast majority of the remaining OLLs.

Anyway, the outcome of my experiment is that I'm over a second faster with the stickerless - 18.32 vs. 19.35.

Totally scientific. Ao12 FTW!!!

 

[JanW]

Tried some PLL attacks again. First 5 between 1:00.006-1:08.7, then finally a sub-1 with 59.92 on the 6th attempt. That would be 6th completed attempt, there was lots of DNFs in between as well that I didn't complete if I messed up badly.

Next I wanted to see how well I can do it blind. First one ended up with a complete mess, 2nd attempt was successful at 1:12.76. The blind PLL attack is interesting. It reveals quite well which PLL algs I'm still not 100% sure about.

 

[pglewis]

Second comp in the books for me, much better than the first go-round.

3x3: 44.90/48.36. Still about 10 seconds off my at-home times but my concentration was obviously much better. Table practice was actually running fairly close to my home times. Botched the very first solve with a 1:03 but didn't let it faze me too much. Some little F2L mistakes on all the others, coulda-shoulda been sub 40 on at least 2 but, meh... I'll be quite a bit faster in 4-6 more months anyway.

2x2: 12.94/18.70. DNF'ed the first solve. Misread PBL (my biggest weakness), did the wrong alg, and stopped the timer before I saw it. Had to be extra careful on the rest because I was determined to get an average. Ended with a counting 25 when I blacked out on the PBL alg and had to fix things up. Still, that's PBs across the board and leaves me tons of room to beat it next time out.

 

[newtonbase]

Second comp in the books for me, much better than the first go-round.

3x3: 44.90/48.36. Still about 10 seconds off my at-home times but my concentration was obviously much better. Table practice was actually running fairly close to my home times. Botched the very first solve with a 1:03 but didn't let it faze me too much. Some little F2L mistakes on all the others, coulda-shoulda been sub 40 on at least 2 but, meh... I'll be quite a bit faster in 4-6 more months anyway.

2x2: 12.94/18.70. DNF'ed the first solve. Misread PBL (my biggest weakness), did the wrong alg, and stopped the timer before I saw it. Had to be extra careful on the rest because I was determined to get an average. Ended with a counting 25 when I blacked out on the PBL alg and had to fix things up. Still, that's PBs across the board and leaves me tons of room to beat it next time out.

Congratulations. PBs across the board is ideal. You'll feel much more at home at competitions from now on.

 

[One Wheel]

First PB of the year. SQ-1 of cause
(12.225), 12.674, 14.781, 16.040, (18.315[p]) = 14.498
which almost also became a PB for a12; (12.225), 12.674, 14.781, 16.040, 18.315[p], 18.861[p], 17.460, (20.114[p]), 15.350, 16.913[p], 16.125, 16.838 = 16.336

Nice. I can only wonder if I'll ever get close. I timed my parity alg again today: down to about 14 from the 15.5 last time I timed it. Getting more under 2:00 than over when I solve square-1 these days. It's a very satisfying puzzle.

 

[mark49152]

Congrats @pglewis!

 

[teacher77]

I would re-order: cross on bottom, good 21 PLL algs, lookahead, learn more OLLs, x-cross, 2sided PLL recog,

You and @JanW both recommended to drop whatever I was doing and start with the cross on bottom so I took your word for it. Today I timed 50 solves with cross on top (as a reference) AND 50 solves with cross on bottom.

This leads to a few questions... But first, the results :

For crosses on TOP, my average for 50 solves was 4.23 sec. I screwed up 2 solves (stopped the timer thinking I had a cross when I didn't), which corresponds to a 4% failure rate. The measured time for these two solves is included in the 4.23 sec average and I didn't add any "penalty" or whatever. Also, and perhaps more importantly, there were many solves (maybe 20-30%, I didn't count them) where I couldn't predict the entire sequence and actually needed to look to place the last piece or sometimes the two last pieces. I'm pretty sure I never can plan more than 4-5 moves in advance.

For crosses on BOTTOM, my average for 50 solves was 5.20 sec. But my screw up rate was 22% (I stopped the timer 9 times thinking I had a cross when I didn't and there were also 2 solves where I stopped the timer although I was pretty sure I had screwed up). Again, the measured times for those 11 screw ups is part of the 5.20 sec average.

Now, I'm thinking that "one more second" is a fairly good start. Especially since I save 0.3-0.4 sec for turning the cube over.

I'm much more concerned with two issues :
1) Screwing up 22% of the time seems to fit with the number of cross-on-top solves where I NEEDED to look. How can you plan 7-8 moves in advance and remember them ? Do you remember an actual letter sequence like U-F-D2-etc ? Or do you remember a visual image of those moves ? Any specific tricks or maybe ressources I could read ?
2) Concentrating on remembering those moves left me little place to actually look ahead. In 50 solves, there's only a single one, which happened to be a 3-move cross, where I actually had found a pair before the cross was finished. Did it start like that for you guys ? How much time did it take you to be able to solve a cross on the bottom while actually looking ahead ?

As always, thanks everyone for your great input

 

[h2f]

Anyway, the outcome of my experiment is that I'm over a second faster with the stickerless - 18.32 vs. 19.35.

I got my stickerless few days ago. I've started to break in the cube and my times are little down but I'm a huge fan of stickerelss version. I've noticed two things: colors and plastic is a little bit different (more gummy). The second - the inside pieces are yellow not white as in a previous one I got. My times are a little bit down now.

@teacher77, In a long run doing crosses on bottom pays off a lot. The most important issue you got doing it - you can track the first pair of f2l when doing cross. With cross on the left or top you cant do it. And one of the exercices you can make is doing crosses with closed eyes after you plan how to do it.

 

[teacher77]

And one of the exercices you can make is doing crosses with closed eyes after you plan how to do it.

That was my question though : how do you proceed to remember 6-7 moves ? Do you remember a sequence like U-F-L2-etc ? Or more an image of the moves ?

 

[pipkiksass]

That was my question though : how do you proceed to remember 6-7 moves ? Do you remember a sequence like U-F-L2-etc ? Or more an image of the moves ?

Personally, I remember visually, rather than in notation. It really does just get easier with practice. As with all things cubing, I'd highly recommend turning the timer off and just playing - scramble, inspect, close your eyes, solve.

It's about pushing yourself, it SHOULD be really hard at first, because your brain is learning new stuff. Once things become more comfortable, it'll be easier. That might mean, for now, predicting 4-5 moves is your limit, but push the limit and soon (with practice) 6, 7, or more will be possible.

That said, I still have cases where the cross is REALLY awkward and I don't plan the full solution, just through laziness. Probably costs me a second or more.

There are a few cross example videos on YouTube, but just watch the first few turns of a few example solve vids. Feliks shared a "white cross example solves" vid a couple of years ago.

Also, don't feel the need to time your inspection. I don't, ever. Take as long as you need. Again, this will come down with practice.


Sent from my iPhone using Tapatalk

 

[h2f]

I think I was just imaging the moves I do. At the moment I just do it. I must add that still the cross and the transition to first pair is the weakest part of my solves. If I do it properly my times are around 15-17 seconds. But with bad cross - 19-21 or worse.