Big Cube BLD Discussion

[tseitsei]

Wow, that is much more free form than I expected. It seems the idea evolves eventually into having each letter pair have multiple choices, each of a different part of speech.

I think empirically that the sentences method is much faster than letter pair images or PAO, but I am baffled as to how sentences is faster than a letter pair PAO. To be clear, I am not questioning whether sentences is faster, I am questioning how is it faster? Do other memory experts use free form methods like this? I was under the impression that the memory sports people all use very rigid and structured systems by comparison.

I'm just curious. I am very motivated to learn a letter pair PAO system because of Ian Winokur, but I am trying to entertain the idea of sentences before I do so.

I use something quite similar to Ollies sentence memo for 4bld and 4x4 part of 5bld.

It is very free form but that's what makes it so good. I can store very many targets to one sentence AND I can choose my words so that the sentence is easy to remember also.

I think that the reason sentence memo is so fast is because unlike images I don't really visualize them nearly as vividly as I do while using images for MBLD which obviously saves time.

And the reason why I think memory experts don't use this method is that they usually memo much more information than 4bld or 5bld and this sentence memo is fast but not that suitable to memo big amounts of information. Kind of a middle ground between images/journey method and just pure audio memo... That's why I use more of image/journey type memo for MBLD and even for +centers and midges on 5bld.

 

[moralsh]

didn't know where to post this, it's neither an accomplishment nor a failure

4BLD Multi 1/2 in 47 minutes, the DNF only off by 2 edges. I will definitely try this again.

 

[ender9994]

didn't know where to post this, it's neither an accomplishment nor a failure

4BLD Multi 1/2 in 47 minutes, the DNF only off by 2 edges. I will definitely try this again.

Getting that close may not count as a complete success, but it is definitely an accomplishment in my book!

 

[A Leman]

Wow, that is much more free form than I expected. It seems the idea evolves eventually into having each letter pair have multiple choices, each of a different part of speech.

I think empirically that the sentences method is much faster than letter pair images or PAO, but I am baffled as to how sentences is faster than a letter pair PAO. To be clear, I am not questioning whether sentences is faster, I am questioning how is it faster? Do other memory experts use free form methods like this? I was under the impression that the memory sports people all use very rigid and structured systems by comparison.

I'm just curious. I am very motivated to learn a letter pair PAO system because of Ian Winokur, but I am trying to entertain the idea of sentences before I do so.

I decided it would be better for myself to stick to one method and master it instead of learning many different methods. I have experimented/played with sentences,linking, and other no loci methods; but when I memo, I always use images at locations for everything except audio corners. I think it's better to get good at one method than mediocre at a bunch of methods. If I picked sentences instead, then they would have gotten all of the focus.

For memory sport, recall timing is kinda lax and having a rigid system makes it easier to force back an image after you forget it. I usually go through my list when I forget something and that's all it takes to remember. I also think images are more suited for the length of the events in memory sport. It would hardly be worth learning sentences to just use it for speed cards. For names and faces and historic dates, linking is popular. I also know some people can do a deck of cards with just linking, but not as fast as using locations. In memory sport, the main changes are the size of the system and the images/loci which each have fast representatives ie Wang Feng=simple 2 digit 2/loci, Jonas von Essen=PAO, Johannas Mallow= 1000 image 2/loci, Ben Pridmore=Ben system 3/loci, Simon Reinhard=10k system. All of them are in the top 10 using different methods but they stick to what they use.

If you want to learn PAO for letter pairs, then just make a list of people and associate items and actions to them. This will make them easier to remember. It's easier for 2/loci as I use it. For example, Gandalf and his staff, Frodo and Sting(his sword), Ash and a pokeball, muhammad ali punching with gloves, goku and kamehameha. When that doesn't work, then I find another word like Oliver/ olives, obelisk(a monster)/obelisk(a stone pillar), Rose(person),rose(flower) where they are the same,or use associations I already knew like mad eye and Mud. I don't completely stick to PA though because some adjectives are very ingrained into my system.

 

[Berd]

What's PAO?

 

[moralsh]

Person Action Object? to remember CS TO RD you use "Chris Turns On RaDio" is very easy as you will picture Chris doing so

Edit: Ender9994 thanks! it felt like it

 

[Berd]

Person Action Object? to remember CS TO RD you use "Chris Turns On RaDio" is very easy as you will picture Chris doing so

Edit: Ender9994 thanks! it felt like it

Awesome!

 

[rybaby]

I decided it would be better for myself to stick to one method and master it instead of learning many different methods. I have experimented/played with sentences,linking, and other no loci methods; but when I memo, I always use images at locations for everything except audio corners. I think it's better to get good at one method than mediocre at a bunch of methods. If I picked sentences instead, then they would have gotten all of the focus.

For memory sport, recall timing is kinda lax and having a rigid system makes it easier to force back an image after you forget it. I usually go through my list when I forget something and that's all it takes to remember. I also think images are more suited for the length of the events in memory sport. It would hardly be worth learning sentences to just use it for speed cards. For names and faces and historic dates, linking is popular. I also know some people can do a deck of cards with just linking, but not as fast as using locations. In memory sport, the main changes are the size of the system and the images/loci which each have fast representatives ie Wang Feng=simple 2 digit 2/loci, Jonas von Essen=PAO, Johannas Mallow= 1000 image 2/loci, Ben Pridmore=Ben system 3/loci, Simon Reinhard=10k system. All of them are in the top 10 using different methods but they stick to what they use.

If you want to learn PAO for letter pairs, then just make a list of people and associate items and actions to them. This will make them easier to remember. It's easier for 2/loci as I use it. For example, Gandalf and his staff, Frodo and Sting(his sword), Ash and a pokeball, muhammad ali punching with gloves, goku and kamehameha. When that doesn't work, then I find another word like Oliver/ olives, obelisk(a monster)/obelisk(a stone pillar), Rose(person),rose(flower) where they are the same,or use associations I already knew like mad eye and Mud. I don't completely stick to PA though because some adjectives are very ingrained into my system.

Interesting how each loci can take several pairs at once. Whenever I tried to use loci, I just put one pair in each location (10 locations in my room worked for centers). Maybe it's because I'm used to linking, where you just go from one object to another. Plus I learned linking first (from Harry Lorayne's "The Memory Book" -- it doesn't teach loci), so it's a habit. I am also getting more into memory sports, and sentences don't seem too useful for things like speed cards or numbers. So probably the method I will practice most will be linking since that's what I use when I try out memory events (I'm only a beginner though). But I could always just separate 4bld from memory sports and use different techniques.

 

[CyanSandwich]

Can someone please give me an alg for solving the M-slice(M2'ing the centers and UF/DB) and swapping UBL-UBR for 5BLD?

Right now I do:
U' F2 U M2 U2 (Rw2 Uw2 F2 r2 F2 Uw2 Rw2) U F2 U and then
R' U L' U2 R U' R' U2 R L U' (L-perm)

So I would prefer something center-safe and shorter.

 

[JemFish]

In Noah's 4BLD tutorial he says that it's best to execute centres first, but I want to use loci/visual memory for wings and centres, and then short-term audio for corners, executing corners first. Can someone here help me find algs for this, or suggest how I could get this to work?

 

[CyanSandwich]

In Noah's 4BLD tutorial he says that it's best to execute centres first, but I want to use loci/visual memory for wings and centres, and then short-term audio for corners, executing corners first. Can someone here help me find algs for this, or suggest how I could get this to work?

Each y-perm does a U' to the centers. So if you have corner targets divisible by 4 you're fine. Otherwise, turn U so that your centers are where you memorized them, then solve centers and undo the setup move afterwards.

Example: You've solved 7 corners, your centers need a U' to be where they were. Do a U', solve the centers, do a U, solve wings.

Also don't want this to get lost.

Can someone please give me an alg for solving the M-slice(M2'ing the centers and UF/DB) and swapping UBL-UBR for 5BLD?

Right now I do:
U' F2 U M2 U2 (Rw2 Uw2 F2 r2 F2 Uw2 Rw2) U F2 U and then
R' U L' U2 R U' R' U2 R L U' (L-perm)

So I would prefer something center-safe and shorter.

 

[A Leman]

Can someone please give me an alg for solving the M-slice(M2'ing the centers and UF/DB) and swapping UBL-UBR for 5BLD?

Right now I do:
U' F2 U M2 U2 (Rw2 Uw2 F2 r2 F2 Uw2 Rw2) U F2 U and then
R' U L' U2 R U' R' U2 R L U' (L-perm)

So I would prefer something center-safe and shorter.

I don't know if the parity could be done completely center safe. If I used M2 and your buffers then I'd do M2U2MU2M' Rw2F2 U2r2U2 F2Rw2 T-perm, but since you use The UF buffer, I can give better advice. Every UF Commutator case that has UB in it is very very fast and easy to learn so instead of doing the last M2 target that misaligned the M-slice, you should do a 3 cycle that solves the last target and puts the buffer midge in UB to setup to the T-perm.

I'll outline those commutator cases for UF-?-UB so you can learn them in a few minutes.
1.Every good edge case(zz definition) can be solved by setting up to M'U2MU2 or MU2M'U2. This covers the UR,UL,FR,FL,DR,DL,BR,BL,DF,DB targets
2. Setup to the MUM'U2MUM' OLL or it's inverse. That covers the RU,LU,RF,LF,RB,LB cases
3. For RD and LD, I use [U2,MDM'] and [U2,MD'M']
4. For FD and BD, Setup using D to step 3.

So 22 cases is actually only 3 algs.

I'd do UR,UL,FD,BD differently on 3bld but the cases above are rotationless, don't lock up much, and fast. (UR and UL would be the standard <RU> U-perm that doesn't work for 5BLD and FD,BD are setups to M'U2MU2 with M moves ie. UMUMU2M'UM'U and U'M'UM'U2MUMU They lockup too much on big cubes)

In Noah's 4BLD tutorial he says that it's best to execute centres first, but I want to use loci/visual memory for wings and centres, and then short-term audio for corners, executing corners first. Can someone here help me find algs for this, or suggest how I could get this to work?

I do it this way. I memo Centers->Wings->Corners and then execute Corners->Centers->Wings with 3 cycles. I do an audio loop for corners and then go through my images on a route. There are a couple reasons why Noah's method is also very good. 1. He can audio loop his centers. It takes practice but centers are usually ~8-10 syllables 2. There are things that are not center safe so solving them first usually solves that problem. Corner parity is one of those things.

If I get Corner parity, then put the swapped corners at UBR-ULB and cross my legs. After I do centers, I check if my legs are crossed and do the corner parity Rw2 F2 U2r2U2 F2 Rw2 y T-perm. Since parity is a fairly long and brain dead sequence of moves, I usually use that time to think ahead to my wing images.

 

[CyanSandwich]

I actually use the DF buffer. What I meant was just normal OP/M2 parity for 5BLD (swaps UF/DB because of the M2 off).

But if there's not really any shorter algs or center-safe ones, I'll stick to what I'm doing. Thank you though!

 

[h2f]

Each y-perm does a U' to the centers. So if you have corner targets divisible by 4 you're fine. Otherwise, turn U so that your centers are where you memorized them, then solve centers and undo the setup move afterwards.

Example: You've solved 7 corners, your centers need a U' to be where they were. Do a U', solve the centers, do a U, solve wings.

Wow. I wasn't aware of that, I didn't noticed that so far. And that's why sometimes I was mistaken in centeres. Thanks a lot, it helps me a lot.

 

[josh42732]

Can anyone help me with 4BLD centers? I can pretty much solve any case other than the D faces. I use Dan's tutorial for centers but I didn't quite understand the part where he solves the d face centers. Everything else (wings, corners) is just fine. I am trying to learn how to solve 4BLD so any other helpful tips would be greatly appreciated.

 

[moralsh]

Which method are you using to solve centers? if you are using U2 is just an alg with D moves to set it up, if you are using conmmutators just set it up to an easy case not on U or D daces taking care of not disturbing your buffer, could you tell us what your buffer is and a case you have problems with?

 

[josh42732]

Which method are you using to solve centers? if you are using U2 is just an alg with D moves to set it up, if you are using conmmutators just set it up to an easy case not on U or D daces taking care of not disturbing your buffer, could you tell us what your buffer is and a case you have problems with?

I used Daniel Sheppard's tutorial for centers. I don't know what method it is called. I think that it is freestyle commmutators with Urb as the buffer. I can not figure out how to do a 3-cycle to the bottom face. Is there something that I'm not quite getting or what?

 

[Ollie]

I used Daniel Sheppard's tutorial for centers. I don't know what method it is called. I think that it is freestyle commmutators with Urb as the buffer. I can not figure out how to do a 3-cycle to the bottom face. Is there something that I'm not quite getting or what?

U r2 U' l' U r2 U' l

If you get a case with a D face center, try setting up to this alg instead.

 

[josh42732]

U r2 U' l' U r2 U' l

If you get a case with a D face center, try setting up to this alg instead.

Thanks, Oliver! What if the D face cycle is the last one and not the first one? Ex. Urb>Rbd>Dlb

 

[Cale S]

Thanks, Oliver! What if the D face cycle is the last one and not the first one? Ex. Urb>Rbd>Dlb

Just do the inverse (l' U r2 U' l U r2 U').