• 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 40,000+ people from around the world today!

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

Older cubers discussions

Tom Joad

Member
Joined
Jun 29, 2016
Messages
137
3x3: My potential seems to have improved about a second. Where I was getting 2 or 3 full step 17s a day on my best solves, now it's full step 16s. I also landed a full step 14.9 a couple days ago, 6-move T-OLL and U-Perm which is probably the best possible two look LL for me. I'm starting to shrug at anything over 18 when I'm in a groove even though I'm still good for following up my sub 20s with a couple mid 20s. I think most of the improvement is lookahead along with a little bit of my ongoing F2L refinement coming together. I still haven't returned to unlimited inspection and first pair prediction but I have made one minor change at the start of my solves the past few weeks. Watching Feliks's recent WR average he consistently sits on the armed timer for a few seconds, visualizing the physical starting moves. Such a seemingly tiny little thing.

3bld: Thankfully I didn't lose much on execution. My first couple sighted attempts were rather hilarious but after 3 or 4 of those all my setups were back like riding a bike. I haven't re-incorporated any of the advanced M slice tricks yet but I have all my notes. First day back I got to better than 50% on slow, careful, untimed solves that were probably in the 8-10 min range. Second day I went for a timed solve to see where I'm sitting and got a 5:09 success. Execution split was close to where I was before the hiatus and I made a tracing mistake I had to track down in memo, so I'm not terribly far off where I was. Pushing tracing and memo really hard is the prescription for the next week.

Good point about Feliks, I’ll try focusing on that myself.

Thanks for the Teng Yun recommendation. I really like it.

I know what you mean about the just shrugging at average solves. I think i’m Just two seconds ahead of you there. Anything over 16 and I’m unimpressed but any “15er” or less makes me feel contented.

Planning the first pair in inspection: I think I’ve stumbled upon a compromise. I generally don’t do this but I do when it “looks easy”. However, every now and then I do a session of about twenty where I always plan first pair (unlimited inspection) I think this is slowly increasing the percentage of solves where I see a “looks easy” first pair. Maybe I’ll reach a point where I see an easy first pair 30% or 40% of the time...

In particular I keep looking at the rU’r’ insert for a bad cross piece and looking at how it affects every cube piece (from all four angles). Still slow from some angles but this seems crucial to planning first pair because so many cross solutions include a bad cross piece like this. Will keep working on this.

Carry on
 

pglewis

Member
Joined
Sep 23, 2016
Messages
1,268
Location
Cincinnati
WCA
2016LEWI07
Good point about Feliks, I’ll try focusing on that myself.

Of course I was already sort of concentrating on the first few moves but watching his Ao5 made me realize I don't hold for as long nor do I have that laser focus I see in him. It stems from the same laziness as good planning in inspection for me: I'm often doing the bulk of my practice in one big batch and start to skimp on individual solves with a "get through 'em" mindset. One of many bad habits to ditch now that fractions of a second really matter.

Keep being my inspiration on deeper inspection :). Even though I haven't done it much or in ages I still occasionally spot a corner I realize isn't going to move in the top layer, one small vestige from the time I did practice it.
 

Tom Joad

Member
Joined
Jun 29, 2016
Messages
137
New PB single 11.27

After my 12.3/12.4 lasted for over a year, it was certainly due and switching to the Teng Yun seems to have given me a new motivation.

However, taking a full second off like that makes me think it will be at least another year before I ever solve a cube that fast again! Indeed, it may never happen. I wish I had it on video! Alas

It was a full solve, the f2l was easy, from what I remember there was at least one three move case and maybe two. I’ve finished f2l in about 8 seconds a few times before but always been stung by a bad oll pll combo at the end but this time I got sune then u perm with just an easy u’ auf which flowed really nicely.

I will now focus on beating my Ao5 PB which stands at 15.56. I guess the long term aim is to Ao100 sub-15 which now seems like a feasible two-year project.

Onwards and upwards
 

DLeiber

Member
Joined
May 16, 2019
Messages
16
Location
Southern California, USA
YouTube
Visit Channel
... and I made a tracing mistake I had to track down in memo, so I'm not terribly far off where I was. Pushing tracing and memo really hard is the prescription for the next week.

I'm brand new at 3BLD, so please forgive me if this is a stupid question, should be asked elsewhere, or has been asked a million times, but you said something that intrigues me. As I trace, I memorize, but it sounds maybe like you do them in two different steps. Do you trace all the pieces somehow just to get an idea of the progression, but then go back and come up with actual letter pair sounds/images and then memorize? I'm not sure I understand how you knew you made a tracing mistake and then tracked it down in memo.
 

mark49152

Premium Member
Joined
Oct 29, 2012
Messages
4,719
Location
UK
WCA
2015RIVE05
YouTube
Visit Channel
I'm brand new at 3BLD, so please forgive me if this is a stupid question, should be asked elsewhere, or has been asked a million times, but you said something that intrigues me. As I trace, I memorize, but it sounds maybe like you do them in two different steps. Do you trace all the pieces somehow just to get an idea of the progression, but then go back and come up with actual letter pair sounds/images and then memorize? I'm not sure I understand how you knew you made a tracing mistake and then tracked it down in memo.
This is the right place :).

Generally, we memo during trace - or at least translate the letters to images and try to remember them as we go, with perhaps a pause afterwards to focus purely on strengthening the memorization.

However, you can think of trace and memo as two separate components. Firstly, because mistakes can be clearly attributed to one or the other: encoding the wrong letters, versus encoding correctly but recalling them wrongly because the memo was too weak. Omissions are another common tracing error. Often, tracing errors become apparent before you finish memo, because something just won't add up (I think this is what Phil was referring to). Recall errors of course usually happen during execution and are fatal. It's also possible to trace correctly but encode wrongly, but that becomes much less common with experience.

The second reason to view them as separate components is performance. Memorization does take time, to group images together and visualise. We can measure our performance on both trace and memo, and train them separately as needed (memo cost can be measured as total average 3BLD memo time minus average trace time). It's easy to practise tracing without memorizing, in order to become faster and more fluid at it, and it can be helpful to do that for long sessions without taxing the memory. It's possible to train memo independently by revising letter pair images or by training memory using other online memory sports tools, for example memorizing sequences of images against the clock.
 

pglewis

Member
Joined
Sep 23, 2016
Messages
1,268
Location
Cincinnati
WCA
2016LEWI07
Do you trace all the pieces somehow just to get an idea of the progression, but then go back and come up with actual letter pair sounds/images and then memorize? I'm not sure I understand how you knew you made a tracing mistake and then tracked it down in memo.

I realized there was an error because I came up with an odd number of corners but an even number of edges.

I do memorize as I trace but because my tracing is very slow in and of itself I plan to focus on just the mechanics of tracing for about a week.
 

DLeiber

Member
Joined
May 16, 2019
Messages
16
Location
Southern California, USA
YouTube
Visit Channel
I realized there was an error because I came up with an odd number of corners but an even number of edges.

I do memorize as I trace but because my tracing is very slow in and of itself I plan to focus on just the mechanics of tracing for about a week.
Thanks for the explanation. My tracing is horribly slow; I should probably concentrate on it, too.
 

DLeiber

Member
Joined
May 16, 2019
Messages
16
Location
Southern California, USA
YouTube
Visit Channel
Thanks, Mark. The tracing error of odd vs even corners/edges makes sense. I've encountered that, too. I think my biggest problem is tracing incorrectly. This will improve with experience, but some more focused practice on that wouldn't hurt.
 

pglewis

Member
Joined
Sep 23, 2016
Messages
1,268
Location
Cincinnati
WCA
2016LEWI07
Thanks, Mark. The tracing error of odd vs even corners/edges makes sense. I've encountered that, too. I think my biggest problem is tracing incorrectly. This will improve with experience, but some more focused practice on that wouldn't hurt.

In my case I just forgot to shoot the final target when I was doing the corner trace but I had to go back and carefully retrace corners to find it.

I did some timed tests to see how quickly I can trace-- just saying the Speffz letters out loud without memorizing-- and my initial benchmark is 50-60 seconds. I'd like to be memo'ing completely in 90 seconds max. I think I can cut my tracing time in half fairly quickly with dedicated practice. I think the top guys do full memo for 3bld in like 8-10 seconds, I figure I ought to be able to trace in 25 seconds lol.
 

pglewis

Member
Joined
Sep 23, 2016
Messages
1,268
Location
Cincinnati
WCA
2016LEWI07
However, taking a full second off like that makes me think it will be at least another year before I ever solve a cube that fast again! Indeed, it may never happen.

Outliers are weird and it's not a lot of data points, but your two fastest solves in the past week? I'm gonna at least call that "really promising".

I’ve finished f2l in about 8 seconds a few times before

Okay, I'm in no position to critique but I can't help myself here: stop peeking at the timer! ;) There'll be time enough for countin' when the dealin' is done.

Hats off though, you were long overdue, and good luck on the longer-term sub 15 goal.
 

mark49152

Premium Member
Joined
Oct 29, 2012
Messages
4,719
Location
UK
WCA
2015RIVE05
YouTube
Visit Channel
I did some timed tests to see how quickly I can trace-- just saying the Speffz letters out loud without memorizing-- and my initial benchmark is 50-60 seconds. I'd like to be memo'ing completely in 90 seconds max. I think I can cut my tracing time in half fairly quickly with dedicated practice. I think the top guys do full memo for 3bld in like 8-10 seconds, I figure I ought to be able to trace in 25 seconds lol.
Yeah that was what motivated me - when I realised that it took me several times as long to even trace the cube as others took to memorize it. I even tried tracing the same scramble over and over - so just physically looking at the stickers for a known trace. Even doing it over and over, I was slower than them. I can't even LOOK at the cube as fast as they can memo! That really made me feel old :).

One tip that helped me: try a metronome. Set it to say 1.5 seconds, and try to move to the next sticker on every beat. It helps establish a smooth rhythm and break the habit of pausing. Ratchet up the beat whenever it starts to feel easy.
 
Top