Blindsolving Discussion and Help

[RageCuber]

Over time you will be able to do more solves in a row before this interference will begin to affect you. You can switch to a memory method that uses loci (Roman Rooms, journeys) and this will substantially increase the number of solves you can do in a day before hitting your interference limit. Make sure you have a large number of memory locations available if you go this route. If you use sentences/visual then just practice often and fight through the interference and your interference limit will increase over time.

If I need to shoot to the target I know as G I just use a word like Goat, let's say the next target is C I would probably use the word Can, leaving me with "goats can..." Then I just continue with targets until I have working sentence(s), mostly I repeat them in my head or in a low whisper. Takes me about two mins. Is that how people who use sentences do it?

 

[obelisk477]

If I need to shoot to the target I know as G I just use a word like Goat, let's say the next target is C I would probably use the word Can, leaving me with "goats can..." Then I just continue with targets until I have working sentence(s), mostly I repeat them in my head or in a low whisper. Takes me about two mins. Is that how people who use sentences do it?

No, they combine the letters. You shoot to G then C, so you memorize letter pair GC as a word like 'Gucci'. And so on.

 

[Feryll]

I'm been interested in big-cube BLD recently, using U2 for all centers and R2/M2 for edges and corners. I've gotten used to making slice moves by holding adjacent slices fixed and only pushing the slice in question, instead of e.g. r2 R2 on a 4x4. I'm doing it this way because it feels safer than risking losing track of my grips and without practice this technique is quicker and easier, but I notice that nobody else seems to do it this way. If I want speed to be any kind of priority, should I switch and get used to it, or keep doing what's most natural?

 

[Mike Hughey]

I bought a Megaminx and want to learn to solve it blindfolded.

But just to come up with a memory system for that many pieces...
Ok, I think I have ideas as to that.

Then I am trying to figure out how to solve, commutators of course.
Is there a list of comms somewhere, I have not found it? Corner comms
I can figure out some, but edge comms? There are no slice moves!!!!

You who have actually done it (or at least tried), do you use comms all
over the Minx or do you use LL-comms with a lot of setup moves?
(Sort of like Turbo-bld for 3x3).

I use a single specific set of commutators with lots of setup moves. It's not that bad. You already should have your buffer piece on the top layer. Come up with a set of ways to get the second piece's proper sticker to the top layer first, and then get the third piece to wherever it has to go (All of my commutators have two pieces on the top layer, one just below somewhere. 8 movers for corners, 12 moves for edges. And the analogues for gigaminx centers.)

Note that the second piece on the top layer can go anywhere on that layer; the algs are the same. So that helps quite a bit in getting a piece up there quickly.

When you do setups this way, you can come up with ways to do the first one to the top layer that don't interfere with the second setup piece in a large percentage of cases. For the few where it's a problem, you just come up with a standard way of handling those too. It takes a number of setup moves for some cases, but it doesn't take too much work to get it fairly consistent. (Hint: I do a lot of "D" turns for setup of the second piece, as in the D++ or D-- "most of the puzzle" turns used in scrambling. You get the first piece to the top, then you turn the "D" ultralayer so the piece is on the right face, then you move it into place with a consistent set of turns.)

For a while I experimented with solving a piece at a time, but I found that two pieces at a time is just way too much more efficient; it was too slow for me to do one piece at a time. I might not feel that way if I weren't so slow, though.

 

[MatsBergsten]

I use a single specific set of commutators with lots of setup moves. It's not that bad.

For a while I used Turbo corners for 3x3, I think this must be rather the same.

Do you have a list of the commutators you use for that?

I aim to be lazy so as to use a subset of my existing memory letter pairs but use a letter pair to denote a "cubie" (whats that for a Megaminx?
megie?). Even if that means a waste of letters and long memory lists it seems simple, I am not concerned with speed (at least not to begin with.

Note that the second piece on the top layer can go anywhere on that layer; the algs are the same. So that helps quite a bit in getting a piece up there quickly.

Not that I understand that but I suppose I will when I see the algs.

Finally I am more daunted now with your attempts at a Gigaminx. Phew!

 

[A Leman]

For a while I used Turbo corners for 3x3, I think this must be rather the same.

Do you have a list of the commutators you use for that?

I aim to be lazy so as to use a subset of my existing memory letter pairs but use a letter pair to denote a "cubie" (whats that for a Megaminx?
megie?). Even if that means a waste of letters and long memory lists it seems simple, I am not concerned with speed (at least not to begin with.



Not that I understand that but I suppose I will when I see the algs.

Finally I am more daunted now with your attempts at a Gigaminx. Phew!

Everyone uses one image/target. It would take 3600 images otherwise. A possible improvement would be to practice some combinations or learn something like PAO. Here is a link to how I lettered it for my attempts https://www.speedsolving.com/forum/showthread.php?27353-One-Answer-Blindsolving-Question-Thread/page399&highlight=megaminx

As a general advice, I find some good stuff by using the "search thread" tool on the top of the thread and searching "megabld" or "megaminx". I usually have more luck finding things with the thread search in the OAQT's than with the website search.

 

[JasonDL13]

At what point would you advise people to move from 2MBLD to 3MBLD? How many 2/2 solves?

 

[Goosly]

At what point would you advise people to move from 2MBLD to 3MBLD? How many 2/2 solves?

One. And after one 3/3, move to 4 cubes.

 

[CyanSandwich]

At what point would you advise people to move from 2MBLD to 3MBLD? How many 2/2 solves?

Yeah I'd say one at most is all you need. I went from 3/3 straight to 10/10.

 

[JasonDL13]

So even if I'm struggling to get 2/2, but I still have gotten 2/2 (Only two successes) - You would still recommend moving to 3MBLD?

 

[CyanSandwich]

So even if I'm struggling to get 2/2, but I still have gotten 2/2 (Only two successes) - You would still recommend moving to 3MBLD?

Couldn't hurt. If you keep getting 0/3 or 1/3 though I would go back to 2. Try to film most of your MBLDs if you can so you can catch mistakes.

 

[ollicubes]

So even if I'm struggling to get 2/2, but I still have gotten 2/2 (Only two successes) - You would still recommend moving to 3MBLD?

You can also try 5 cubes multi just for fun. Then when you get back to 2cubes multi you are like "Ouh, just two cubes. This is easy."

 

[deadmanlsh]

Which are the best buffers to use for 3-style? I think UB/UBL would be really good for resolving parity, since all you need to do is solve the edge to UL and then solve the last corner like OP, since audio edges/image corners seems to be the best memorisation method for 3BLD. I use UF and URB. I come from a corners-first kind of solving style, so I do extra cycles just to set up the cube to a J Perm or use weird set-ups to PLLs for easier cases for the last edge. Any tips/thoughts?

 

[tseitsei]

I would say that you should try as many cubes as you "can" do. And by can I mean that you can memo(so well that you dont usually forget the memo) and execute the amount of cubes you try within the time you have in use.

So if you can memo 5 cubes in, lets say 30 mins, then that leaves you 20 mins for execution and you should try 5 cubes. And so on and so on...

 

[deadmanlsh]

So even if I'm struggling to get 2/2, but I still have gotten 2/2 (Only two successes) - You would still recommend moving to 3MBLD?

You can try as many cubes as you think you can (be generous in your estimate here), then decrease the number of cubes until you are comfortable (but not too comfortable) enough to get a decent success rate. If that number happens to be 2, stick with 2 for a while, then repeat the above process again. If that number happens to be 3, or even higher, stick with that number and repeat the above process as well after a while. This way you can keep improving without the task being too easy to improve or too difficult to complete.

Or you can just push yourself as hard as you can if you can handle the constant failure without losing motivation, which can be disastrous to improving at all. You'll make it there either way if you continue pushing on without giving up.

 

[JasonDL13]

-snip-

Thanks

 

[h2f]

How to solve parity with commutators on corners and M2 for edges? I solve corners first.

 

[deadmanlsh]

How to solve parity with commutators on corners and M2 for edges? I solve corners first.

There are many possible ways, but the simplest would be this.

If you use the parity algorithm that swaps UB and UL (for Old Pochmann for those coming from M2/OP, with UBL as buffer), then you can solve the last corner as you would with OP.

Otherwise, you can do an extra cycle and solve to a convenient state; e.g. if your last target was FRU, you can do (UBL FRU RFD), an A9, which allows you to always end with the truncated Y-Perm if you have parity. There is a lot of flexibility in choosing that "convenient state", which can help a lot if you don't want to hold on to the last piece in your memory.

 

[h2f]

There are many possible ways, but the simplest would be this.

If you use the parity algorithm that swaps UB and UL (for Old Pochmann for those coming from M2/OP, with UBL as buffer), then you can solve the last corner as you would with OP.

Yeah, I thought about that way of solving parity. Thanks.

 

[Myachii]

For Multi-BLD, is there a minimum amount of cubes you can have? (obviously more than 1)