[Help Thread] Blindsolving

[White KB]

Hey guys, WhiteKB here. In the past I've tried getting into BLD, but it hasn't worked out because it took too long and I almost never got a solved cube at the end. Any tips for getting into it? Currently, It takes 11-14 minutes for an attempt and I have a <10% success rate. Also currently I am using OP on edges and corners. I would preferably like something that will take me 1-2 weeks to do that will get me a 75% or greater success rate and will allow me to do it in 9-11 minutes (or less) by the end of the time frame. Again, if you have any tips, please let me know. Thanks!

 

[Skewber10]

i with not a lot of practice got around 75% accuracy averaging 7-9 min. Here is my method. Hope this helps.

I use a variation of the PAO system to memorize corners. Here is is.
The first letter you think of a person you know or who you have heard of example: a YouTuber.
The second letter you think of an action that the person can do
The third letter indicates an object that the person is doing the action to
Example of PAO system: GEW = Galen eats watermelon.

Next is the journey method. It is really easy but all the tutorials make it sound really complicated. Arrange a path that you often walk example: around your living area. Visualize your images in those places as you walk through your journey. It’s that easy. For me, if my first letter triplet was GEW I would imagine Galen eating w watermelon on my front porch where my journey starts.

I use these to systems to memorize corners. I memo and review a couple times by in my head visualizing the journey. After this I never review it during edge memo and. This is how well this memo system sticks.

Next , I memo edges. I first trace through my first two targets but don’t memo them. Then I use the audio system to memorize the rest of the edges. Once I have memoed the edges I go back and don’t memo but visualize my first to targets and then just execute quickly. This makes it so you don’t need to keep so much audio in your head.

Absolutely nothing is special about how I execute.

 

[schapel]

I wouldn't recommend this sort of PAO system, standard letter pairs will give you more variety and will help you in the long run
Journey also isn't too necessary, you can experiment with it but it's unlikely to be helpful
Best thing you can do is just practice as much as you can, and see DNFs as learning moments rather than failures

Also, any system where you have to retrace edges is complicated and unnecessary, just do pure audio the whole way through if you can

 

[AbsoRuud]

I am learning right now as well. I'm focusing on learning corners only right but and slowly working on edges. I find it it fairly easy to remember 8-10 letters in pairs. It took me two days to get my first couple of success doing only corners.I use the blind method on speedcubereview.com, it works well for me. Especially since he uses yellow on top and red in front, which is how I solve my cube with Roux. So it's a natural orientation for me. Edges are a few more letters but that isn't a big challenge either.

 

[tx789]

World Class 3BLD people don't have a 75% success rate, above 33.3% is an aim in the long term. Just do a lot of solves at this stage you can get down to 2 min with OP/OP.

Make sure you have a word for all letter pairs. Also switch to using audio for edges or corners if you don't already. So make sounds out of the letter pairs instead of making a image and say it in your head at first you'll need to repeat it a lot but with pratise this will get easier. Most people do audio edges. Memo the piece type you use with audio last and execute it first. So if you choose audio edges memo corners visually then memo edges with audio, solve the edges, solve the corners. Then just do a lot of solves. Do 100 over a few weeks and see where you're at.

The journey method is unessacry for 3BLD that is used for MBLD and bigBLD.

 

[White KB]

ok those sound like some good tips! thanks guys!

 

[PetrusQuber]

Hey guys, WhiteKB here. In the past I've tried getting into BLD, but it hasn't worked out because it took too long and I almost never got a solved cube at the end. Any tips for getting into it? Currently, It takes 11-14 minutes for an attempt and I have a <10% success rate. Also currently I am using OP on edges and corners. I would preferably like something that will take me 1-2 weeks to do that will get me a 75% or greater success rate and will allow me to do it in 9-11 minutes (or less) by the end of the time frame. Again, if you have any tips, please let me know. Thanks!

Just keep on practising (even though I'm a noob). That is all you need. Don't worry about timing for now, just take your time to think about setup moves and etc, or you'll probably make a mistake. Tapping all the pieces during inspection helps me think about which colour will be on top, you could try that. Also, what memorisation technique so you use? Letter pairs is easy enough. If you do do that, make sure you know your cube - SPEFFZ

 

[z3non]

Hey... I am brand new and atm I am learning 3BLD (with TuRbo for edges, buffer at C/I)(front red,right green) . My scramble is L' U' L2 R U' B2 U2 F D' B' R' F2 U2 B2 U B' F' .
After (UM) (XO) (WA) (PE) I am at a state ...
left with (LF) (flipped, right position) and (T R N ). [scramble to produce this state: L' B2 U B2 U' B' L B2 R2 U' L' U B' L B R2 (16f*)]
How do I proceed/ handle this flipped stuff ?? I won't be able to produce a 3-cycle for those, right?? TIA guys

 

[ThatGuy]

Hey... I am brand new and atm I am learning 3BLD (with TuRbo for edges, buffer at C/I)(front red,right green) . My scramble is L' U' L2 R U' B2 U2 F D' B' R' F2 U2 B2 U B' F' .
After (UM) (XO) (WA) (PE) I am at a state ...
left with (LF) (flipped, right position) and (T R N ). [scramble to produce this state: L' B2 U B2 U' B' L B2 R2 U' L' U B' L B R2 (16f*)]
How do I proceed/ handle this flipped stuff ?? I won't be able to produce a 3-cycle for those, right?? TIA guys

This video goes over parity, though fairly quickly. The overall method is different, but since turbo is a 3-cycle method like 3-style, you handle parity in conceptually the same way. (Note that if you were using Old Poch for both edges are corners, then parity would always be an Rperm)

Basically, in the most convenient scenario, you would have an even number of corner letter pairs and an even number of edge letter pairs, which is solvable by using 3-cycles. However, it is possible to get an odd number of corners and an odd number of edges, so you'll be able to solve everything with 3-cycles except the last corner and last edge. Then, you would perform a 'parity algorithm' to fix this. However, doing it this way can lead to some complicated situations. Instead, an easier way to deal with parity is:
1) Memorize corners. If you have an odd number of letters, you have parity.
2) Memorize edges. If you have parity, then you memorize A and D as swapped (I use Old Poch for corners still so A and D is most convenient.). So, in your case, if you saw the yellow orange sticker (normally A for you), you would pretend that it needed to go in the yellow blue slot (normally D for you).
3) Execute edges as you memorize them.
4) Execute corners. Now, in the case of parity, the last corner will not be a part of a letter pair--you will need to swap it with your buffer, which normally isn't possible since you cannot swap exactly two pieces. But since you memorized the A and D edges as swapped, you'll actually be swapping two corners and two edges, which is basically just PLL with setup moves.

It's a bit more confusing in text so here's an example of your scramble.
Corners:
IK (new cycle) BY CB UT U => you have parity, so memorize edges with A and D swapped.
Edges:
UM XO WD (new cycle) AP QR NH, then flip the FL piece with your buffer, for example doing U' L' U, then the algorithm M' U M' U M' U2 M U M U M U2, then reversing the setup moves. I've bolded the pieces you should memorized as swapped.
Execute edges, and execute corners until you get to U. Now, since A and D were swapped from how you memorized edges, you can just do F2 + Yperm to finish the solve.

Parity was a bit hard for me to wrap my head around at first so let me know if you have any questions.

 

[z3non]

This video goes over parity, though fairly quickly. The overall method is different, but since turbo is a 3-cycle method like 3-style, you handle parity in conceptually the same way. (Note that if you were using Old Poch for both edges are corners, then parity would always be an Rperm)

Basically, in the most convenient scenario, you would have an even number of corner letter pairs and an even number of edge letter pairs, which is solvable by using 3-cycles. However, it is possible to get an odd number of corners and an odd number of edges, so you'll be able to solve everything with 3-cycles except the last corner and last edge. Then, you would perform a 'parity algorithm' to fix this. However, doing it this way can lead to some complicated situations. Instead, an easier way to deal with parity is:
1) Memorize corners. If you have an odd number of letters, you have parity.
2) Memorize edges. If you have parity, then you memorize A and D as swapped (I use Old Poch for corners still so A and D is most convenient.). So, in your case, if you saw the yellow orange sticker (normally A for you), you would pretend that it needed to go in the yellow blue slot (normally D for you).
3) Execute edges as you memorize them.
4) Execute corners. Now, in the case of parity, the last corner will not be a part of a letter pair--you will need to swap it with your buffer, which normally isn't possible since you cannot swap exactly two pieces. But since you memorized the A and D edges as swapped, you'll actually be swapping two corners and two edges, which is basically just PLL with setup moves.

It's a bit more confusing in text so here's an example of your scramble.
Corners:
IK (new cycle) BY CB UT U => you have parity, so memorize edges with A and D swapped.
Edges:
UM XO WD (new cycle) AP QR NH, then flip the FL piece with your buffer, for example doing U' L' U, then the algorithm M' U M' U M' U2 M U M U M U2, then reversing the setup moves. I've bolded the pieces you should memorized as swapped.
Execute edges, and execute corners until you get to U. Now, since A and D were swapped from how you memorized edges, you can just do F2 + Yperm to finish the solve.

Parity was a bit hard for me to wrap my head around at first so let me know if you have any questions.

--edit--
I'll look into it, after some tests I may have some questions. Thanks a lot.

 

[z3non]

This video goes over parity, though fairly quickly. The overall method is different, but since turbo is a 3-cycle method like 3-style, you handle parity in conceptually the same way. (Note that if you were using Old Poch for both edges are corners, then parity would always be an Rperm)

Basically, in the most convenient scenario, you would have an even number of corner letter pairs and an even number of edge letter pairs, which is solvable by using 3-cycles. However, it is possible to get an odd number of corners and an odd number of edges, so you'll be able to solve everything with 3-cycles except the last corner and last edge. Then, you would perform a 'parity algorithm' to fix this. However, doing it this way can lead to some complicated situations. Instead, an easier way to deal with parity is:
1) Memorize corners. If you have an odd number of letters, you have parity.
2) Memorize edges. If you have parity, then you memorize A and D as swapped (I use Old Poch for corners still so A and D is most convenient.). So, in your case, if you saw the yellow orange sticker (normally A for you), you would pretend that it needed to go in the yellow blue slot (normally D for you).
3) Execute edges as you memorize them.
4) Execute corners. Now, in the case of parity, the last corner will not be a part of a letter pair--you will need to swap it with your buffer, which normally isn't possible since you cannot swap exactly two pieces. But since you memorized the A and D edges as swapped, you'll actually be swapping two corners and two edges, which is basically just PLL with setup moves.

It's a bit more confusing in text so here's an example of your scramble.
Corners:
IK (new cycle) BY CB UT U => you have parity, so memorize edges with A and D swapped.
Edges:
UM XO WD (new cycle) AP QR NH, then flip the FL piece with your buffer, for example doing U' L' U, then the algorithm M' U M' U M' U2 M U M U M U2, then reversing the setup moves. I've bolded the pieces you should memorized as swapped.
Execute edges, and execute corners until you get to U. Now, since A and D were swapped from how you memorized edges, you can just do F2 + Yperm to finish the solve.

Parity was a bit hard for me to wrap my head around at first so let me know if you have any questions.

It worked well. THANKS!! Although " IK (new cycle) BY CB UT U " wasn't my solution for corners.... I have no Y.. just uvwx... my corners: (IK)(CBXC)(LOL) and setting up last L to U face with D F'

 

[OreKehStrah]

Okay guys I have a bit of an odd question because of some odd circumstances. Basically my vision is so poor that without glasses my vision is worse than what’s considered legally blind. As a result, when I try to do blind solves without my glasses I can’t even tell separation of pieces. To do that I have to have the cube 2-5 inches from my face, which is not convenient. So I have a question.
Instead of taking my glasses off and using a blindfold, could I use no blindfold and just look straight up? I know it’s a really weird question but I’m curious since they usually use a piece of paper or something to cover the cube anyway, and dealing with this issue is part of why I don’t do blind solves.

 

[PetrusQuber]

Can’t you just inspect with glasses, then take them off and don the blindfold? Or even just find a blindfold which is comfortable with glasses on? It’s not a big issue... I guess that’s sacrificing speed though.

 

[OreKehStrah]

Can’t you just inspect with glasses, then take them off and don the blindfold? Or even just find a blindfold which is comfortable with glasses on? It’s not a big issue... I guess that’s sacrificing speed though.

No. If I take the glasses off that’s technically some additional time added to the solve since I’m not going to just toss my glasses lol. And then if you have a blindfold on over glasses there’s an area in between your face, the glasses, and blindfold from which you can see so I don’t think that it’s allowed.

 

[ThatGuy]

It worked well. THANKS!! Although " IK (new cycle) BY CB UT U " wasn't my solution for corners.... I have no Y.. just uvwx... my corners: (IK)(CBXC)(LOL) and setting up last L to U face with D F'

Nice! Also, I forgot I use Y instead of X, whatever works though.

No. If I take the glasses off that’s technically some additional time added to the solve since I’m not going to just toss my glasses lol. And then if you have a blindfold on over glasses there’s an area in between your face, the glasses, and blindfold from which you can see so I don’t think that it’s allowed.

You may look into blindfold goggles that can fit over glasses and are designed to block out the gaps as well. Not sure if they are WCA legal but I feel like they should be.

 

[ctan]

Hello guys

just a beginner in BLD, using OP
I usually have issues knowing if I finished my memorization, mostly on edges
Is there a way to check if I didn't miss anything ?

Thanks !

 

[PugCuber]

Hello guys

just a beginner in BLD, using OP
I usually have issues knowing if I finished my memorization, mostly on edges
Is there a way to check if I didn't miss anything ?

Thanks !

The way almost everyone does it is by simply reviewing the memo. If you memo Corners then Edges, it’s best to execute Edges then Corners, which means you can put edges in short term memory by using a memory method like Audio which reduces the chance of you forgetting edges.

 

[mark49152]

Hello guys

just a beginner in BLD, using OP
I usually have issues knowing if I finished my memorization, mostly on edges
Is there a way to check if I didn't miss anything ?

Thanks !

Yes, you can add up the targets. For edges:
T + P - B = 11
Where T is the number of letter targets, P is the number of edges already solved or flipped in the correct position, and B is the number of cycle breaks. Count each as you go along, and if the above expression comes to less than 11, there's something you haven't found yet.

Same applies to corners but using 7 not 11.

 

[z3non]

Yep. You can check if you covered all edges. We can look at my example above with the following memorization for edges:
(UM XO WD)(AP Q)(R NH)(FL)
The question: Did I miss any edges here?
You ALWAYS cover ([1st cycle positions(letters)]+1) + ([2nd cycle positions(letters)]-1) + (...) + ([Nth cycle positions(letters)]-1) NUMBER OF EDGES/CORNERS!!!

In my example I have (UMXOWD) for my first cycle (covering 6+1 edges). For my 2nd cycle (APQ) I cover 3-1 edges. For my 3rd cycle (RNH) I cover 3-1 edges. For my 4th cycle (FL) I cover 2-1 edges. (6+1)+(3-1)+(3-1)+(2-1)=12
So I covered all 12 edges and didn't miss anything.

 

[ctan]

Wow thank you guys I will for sure try both your methods, which at the end are both logical

Thanks you so much !