Blindsolving Accomplishment Thread

[cmhardw]

I use x' U' R' U R' F R F' r2 F R' F' R U' R U, it's a bit longer, but I think it's a lot faster.

It's nice, Maarten. Thanks
You should post it on Official BLD Algorithms thread.

If I understand correctly that is the A part of a commutator for the Per Special case? Have you guys tried Daniel Beyer's alg for this case? It's amazing!
L2 U' r2 U r2 U L2 U' r2 U' r2 U

It's a commutator and breaks down as [L2, U' r2 U r2 U]

That's the fastest angle for Daniel's way of doing it. The same angle your alg does would be:
U2 R U2 L' U2 R' U2 R U2 L U2 R'

or in commutator form: [U2, R U2 L' U2 R']

Chris

 

[trying-to-speedcube...]

Doesn't work, Chris. The A-part of that commutator changes 4 pieces of the L-layer, there's nothing special about it.

Anyway, new PB on 4x4BLD: 13:10.55.

I'm coming, Chuck

 

[Hyprul 9-ty2]

I just completed my first ever successful BLD attempt!
It was my first full BLD attempt, but I had been practicing "Edge only" BLD for a bit before this.
14 minutes 54.07 seconds.

 

[Mike Hughey]

I look forward to seeing what your breakdowns look like, Chuck.

Oh my... Now that I think about it, maybe I can try to give PAPA to my centers with the same PAPA I gave for the adjacent corners. For example, I can name Ubr center with the same name for UBR corner. I've never thought about it, because I learned 4x4x4 BLD from your How-To thread.

Yeah, I use the letters to identify images. And I use the same images for centers as for corners or edges. No reason why you can't do the same.

Yes, I think it's faster and stronger on a long term memorization (like multi and big cubes BLD). Only with a lot of practice we can use it fast for single BLD. You should try it sometime, it's only 86.

I really think I might try it, but the thousand images version. It's easy enough - many of my images can already easily translate to an action, and sometimes I use actions instead of objects. I do things a little funny, though - I always have a person acting on objects in my locations, so I'll probably go more like person-action-object-action-object. But I think this could really make the memory faster and stronger, like you say. And it should cut down a bit on the problem of doing images in the wrong order, which is probably responsible for 15% of my DNFs or so.

Your version is similar in approach to Tim Habermaas's in that you go with a single piece per image. And considering his (and your) success, it's obviously pretty powerful for multiBLD.

 

[Chuck]

Have you guys tried Daniel Beyer's alg for this case? It's amazing!
L2 U' r2 U r2 U L2 U' r2 U' r2 U
The same angle your alg does would be:
U2 R U2 L' U2 R' U2 R U2 L U2 R'

I don't understand. It's not working, Chris.

Anyway, new PB on 4x4BLD: 13:10.55.I'm coming, Chuck

Nice! I'm waiting

I really think I might try it, but the thousand images version.

How is the thousand images version? More than one image per piece?

I look forward to seeing what your breakdowns look like, Chuck.

I've got it!
Owwww I can't wait to get your response.

4x4x4 BLD Breakdown
1. Deciding orientation & counting solved centers - 00:14.50
2. Memorizing centers - 01:55.60
3. Memorizing edges - 01:50.70
4. Memorizing corners - 00:31.30
5. Solving centers - 03:04.70
6. Solving edges - 02:35.20
7. Solving corners - 00:54.60

TOTAL TIME - 11:06.50 (solved)
Memorization - 04:32.10
Execution - 06:34.50

I really appreciate if you, Maarten, and Chris can be bother to tell me your breakdowns too. Wow I'm so falling in love with 4x4x4 BLD.

Thank you.

 

[trying-to-speedcube...]

I haven't really timed yet. Estimations:

- Pick orientation 0:15
- Memo centers 2:30
- Memo edges 5:30
- Review centers and edges 7:00
- Memo corners 7:30
- Put on blindfold 7:31
- Solve corners 8:00
- Solve centers 12:00
- Solve edges 15:00

For a normal solve.

 

[Chuck]

- Review centers and edges 7:00

I've always memorize 4x4x4 BLD using the multi BLD way, not the single BLD way. It's hard to explain what's the difference, but I'm sure you understand. This way I can minimize review at the end. Sometimes I didn't review at all because the items already stored deep enough.

Maybe you can try it too, so you can cut much of your time.

- Put on blindfold 7:31

Oh my! How can I forget about it!

 

[trying-to-speedcube...]

- Review centers and edges 7:00

I've always memorize 4x4x4 BLD using the multi BLD way, not the single BLD way. It's hard to explain what's the difference, but I'm sure you understand. This way I can minimize review at the end. Sometimes I didn't review at all because the items already stored deep enough.

Maybe you can try it too, so you can cut much of your time.

So you mean taking more time to memorize the pieces so they are stored in your head directly, which reduces the time you need to review your memo? Sounds like a good idea, I'll try. Thanks.

Oh, and you cheat, you don't put on your blindfold.

 

[Mike Hughey]

I really think I might try it, but the thousand images version.

How is the thousand images version? More than one image per piece?

Two pieces per image. Right now I use 552 images, one for each letter pair AB to XW (not bothering with double letters, although I actually have images for most of those too). The idea would be to have 2 sets of images - one an object and the other an action. Then alternate, like you do with PAPA.

I look forward to seeing what your breakdowns look like, Chuck.

I've got it!
Owwww I can't wait to get your response.

4x4x4 BLD Breakdown
1. Deciding orientation & counting solved centers - 00:14.50
2. Memorizing centers - 01:55.60
3. Memorizing edges - 01:50.70
4. Memorizing corners - 00:31.30
5. Solving centers - 03:04.70
6. Solving edges - 02:35.20
7. Solving corners - 00:54.60

TOTAL TIME - 11:06.50 (solved)
Memorization - 04:32.10
Execution - 06:34.50

Wow, your memorization is fast. You're just about as fast as me at memorization, so I can't help you too much there. However, I do notice that you took longer to memorize centers than edges, and I'm usually the other way around, since there are fewer centers than edges to memorize. How do you keep track of where you are on the centers? I place a finger on the last-used one on each face as I go, so I know what's next all the time. I can't do that on edges - that's usually what gets me in trouble memorizing edges, when I have to break a new cycle.

For solving, again your centers seem comparatively slow, but that will probably just improve with practice - they're the newest thing for you, so they probably need more practice. Also, the corners seem rather slow - you might consider working on them a bit. They'll also help you with 3x3x3 BLD.

But overall, you're doing quite well - it's amazing how fast you've improved.

I really appreciate if you, Maarten, and Chris can be bother to tell me your breakdowns too. Wow I'm so falling in love with 4x4x4 BLD.

I haven't taken any breakdowns lately, but I'll post a set here the next time I get a chance to do one. Hopefully soon.

 

[Chuck]

So you mean taking more time to memorize the pieces so they are stored in your head directly, which reduces the time you need to review your memo?

Yes, exactly. You can see that I memorized the 4x4x4 corners in 31 seconds, that's twice slower than I memorize corners in a single 3x3x3 BLD.

Oh, and you cheat, you don't put on your blindfold.

Oh nnnooo... you found out about it!! LOL

 

[Mike Hughey]

Yes, exactly. You can see that I memorized the 4x4x4 corners in 31 seconds, that's twice slower than I memorize corners in a single 3x3x3 BLD.

This is why I solve corners first. It cuts quite a bit of time off the corners - if there are no twisted-in-place corners, I usually memorize them in 10 seconds or so. That part is like doing a 2x2x2 BLD.

 

[Chuck]

This is why I solve corners first. It cuts quite a bit of time off the corners - if there are no twisted-in-place corners, I usually memorize them in 10 seconds or so. That part is like doing a 2x2x2 BLD.

You mean by direct visualization?
I can't, I'm very bad at it, and tapping the piece never helps too.

However, I do notice that you took longer to memorize centers than edges, and I'm usually the other way around, since there are fewer centers than edges to memorize. How do you keep track of where you are on the centers? I place a finger on the last-used one on each face as I go, so I know what's next all the time.

I place my fingers too, I'm doing exactly what you recommended on the How-To thread. It's slower because I've always have a hard time pairing random letters. Yes, other than 8 move commutator, letter pair is new thing for me too. Thank you for noticing it. I will experiment with PAPA for centers.

I can't do that on edges - that's usually what gets me in trouble memorizing edges, when I have to break a new cycle.

Yes, we will always running out of fingers With only 86 PA, usually I can easily notice what Person/Action that haven't joined in my story. And it usually ended with 6 rooms (1 room for 2 pairs of PA. That's why it's called PAPA). When I have cycle breaking(s), it will be more rooms. When I have already solved edges, it will be less rooms. This way I can know whether I've covered all the edges. Sorry if my explanation is confusing.

For solving, again your centers seem comparatively slow, but that will probably just improve with practice - they're the newest thing for you, so they probably need more practice. Also, the corners seem rather slow - you might consider working on them a bit. They'll also help you with 3x3x3 BLD.

I know it! It's time for me to say, "well, let's just practice more!"
Stackmat timer, just wait for me a little bit more...
Thank you so much for your help, Mike.

 

[Mike Hughey]

I decided to go ahead and do a quick one. Well, not so quick - it was a bit of a pain - twisted corners that confused me and both parities. So sorry if my corners were really slow here - they're not normally this bad. (This would probably have been a 1:00 2x2x2 BLD for me, which is really bad corners for me.)

Orient: 0:11
Memorize edges: 1:52
Memorize centers: 1:11
Review edges and centers: 0:41
Memorize corners: 0:35
Solve corners: 0:37
Solve centers: 1:21
Fix corner parity: 0:12
Solve edges: 2:00
Fix edge parity: 0:17

So just under 9 minutes total - solve time was 8:57.82; I think a little was lost in the above splits because I rounded them. You can see we do have a pretty big difference in centers. Looks like you just need more practice, which is not surprising considering how little time you've been doing this so far.

And when I say memorizing the corners fast, I don't do direct visual; I memorize the images verbally, by simply saying them. Since I do pairs of pieces per image, that usually only means 3 or 4 words, which are easy to remember verbally. It might be harder with 6 or 7 words. And I do use visual for twisted-in-place corners, but sometimes that trips me up, like on the solve above.

Chuck, I suspect you'll get a stackmatted solve sometime in the next couple of weeks. (And please note that today counts as sometime in the next couple of weeks - I say that because I wouldn't be surprised if it happens today!)

 

[trying-to-speedcube...]

Cheater, you didn't put on the blindfold.

Anyway, do you have so few recall delays? Fixing an edge parity in 17 seconds isn't very fast, I can do it in 8. That makes me think you turn very slowly. But still your solving time is a lot faster than mine. Is that because you are so much more efficient or because you have fewer recall delays?

 

[Chuck]

Solve centers: 1:21
You can see we do have a pretty big difference in centers. Looks like you just need more practice, which is not surprising considering how little time you've been doing this so far.

Wow Just wow!
Yes... I should practice more.

Chuck, I suspect you'll get a stackmatted solve sometime in the next couple of weeks. (And please note that today counts as sometime in the next couple of weeks - I say that because I wouldn't be surprised if it happens today!)

Thank you for all the troubles, Mike. I kinda feel that I've hijacked this BLD Accomplishment Thread with all of my questions. Sorry, everyone...

Fixing an edge parity in 17 seconds isn't very fast, I can do it in 8. That makes me think you turn very slowly.

I do it in like, 25 seconds :fp

Are we using same cube for 4x4x4 BLD?
(and now I'm asking something again) :fp

I'm using Mefferts, which doesn't have a smooth slice turns.
I've tried using Eastsheen, but it locks like mad.
I've tried using Mini QJ too, the slice turns are very nice, but its outer layers are way too smooth so I ended up doing many overturns on the outer layers.

 

[Mike Hughey]

Cheater, you didn't put on the blindfold.

I mislabeled. "Solve corners" should have read "Put on blindfold and solve corners".

Anyway, do you have so few recall delays? Fixing an edge parity in 17 seconds isn't very fast, I can do it in 8. That makes me think you turn very slowly. But still your solving time is a lot faster than mine. Is that because you are so much more efficient or because you have fewer recall delays?

I'm bad at edge parity. I use a stupid algorithm, and then I use stupid methods to set up for it. But I've practiced it so much this way that I don't trust changing it (I almost never mess it up, which is why I trust it so much), and I haven't been so hungry for an extra few seconds on average that I went for a better way. Someday hopefully I'll fix it. This particular parity fix had a 5-move setup, then my 15-move parity fix algorithm, then 5 more moves to undo the setup. So 25 moves in 17 seconds, which is admittedly slow, but not perhaps quite as slow as you thought, since it's also inefficient.

I really strive for few recall delays. The goal is to know what my next pieces are while still doing the previous algorithm. That's easy to do on the (typically) 3 images I have at any given location. I probably only hit that half of the time between locations, but a good solve (sub-8) usually means I manage that between almost all the locations. That's why I spend the extra time reviewing edges and centers - without that review time I almost invariably have bad recall delays. And I do still have bad recall delays fairly often - that's where most of my solves over 10 minutes come from.

I do it in like, 25 seconds

Don't feel bad, I'm sure I was 25 seconds on it a year ago. I've gotten faster since then. (But I'm still slow enough to amaze people. )

I've tried using Eastsheen, but it locks like mad.
I've tried using Mini QJ too, the slice turns are very nice, but its outer layers are way too smooth so I ended up doing many overturns on the outer layers.

My Eastsheen locks like mad, but I've used it on all of my competition 4x4x4 BLD solves. I basically just got used to it. As for my QJ (not a mini - I think I would prefer the mini, but oh well), it does tend to turn way too smooth, but I'm just trying to get used to that too.

 

[martijn_cube]

new PB 3x3: (i'm getting crazy) 2:00.08
Next PB will sub2.

 

[PEZenfuego]

My first blind 3x3 today!

Wow, it wasn't hard at all!

I just took my time (yeah it took a total of 20 minutes to do). lol, I am going to continue to practice. I can't wait to get my time down to something impressive.

Thank you forum, I couldn't have done it without you guys (especially Stefan Pochmann...after all it was his method).

 

[PEZenfuego]

Congrats to you as well Sandbest!

 

[blah]

3 consecutive 4x4x4 BLD successes It's on cubemania.org

10:30.90, 9:09.87, 10:37.62

Should I go for a fourth attempt?

Yeah - go for an average 10/12!?!

Congratulations - nice results.

Progress so far: 9:11.46, 7:55.28 , 11:12.34

That's 6/6 With a new PB 7:55.28.

They're all on Cubemania, so if you want the scrambles... :rolleyes: