I am kind of in the other direction. I memorize in groups of 8 letters, so I usually end up with one of the following:Originally Posted by Mike Hughey
Person Action Adjective Object
Adjective Person Action Object
Person Adverb Action Object
Adjective Person Adverb Action
This allows a little more wiggle room with the same kind of recall.
Side Note: I prefer the term character over person
Side Note 2: I also use MoP for MP =D
I have just recently started using a Letterpair PAO-ish type system where I have 26 groups of 26 characters that each have An accesory and action that I link to have 10 letters in a loci.for example, Ash ketchum in piccolo's cape uses Avada Kedavra on a wartortle wearing Gandalf's hat and holding his stick would fit in a loci.
I also like this for numbers since they are grouped in 10's which makes it easier to find say the 326th letter of pi in a roman room.
Another trick I figured out for cards is that the first four parts of the image tell me the number and color of the cards while i use an audio loop to remind me wheather it is clubs/spades hearts/diamonds. I then translate the audio into the last image in the loci. This gives me 8 cards in a loci and an easy compromise to the Ben system which would have required me to quadruple my image list. I have never heard of anyone else using a system like this for cards, but it was a convient way for me to make my blindsoving system work better than a single card PAO.
that seems like a pretty good PAO thing. For cards, that seems pretty hard though... how can you keep track of club/spade/heart/diamond for 52 different cards? I really like normal PAO for cards. I end up getting 3 per loci, which is a fairly good amount. 8 cards in a loci seems awfully crammed like you may mix up images.
Maybe I wasn't too clear,I only keep track of 8 at a time. It is a bit more thought intensive than single letter PAO, but it is stores more. A-M are the reds and N-Z are the blacks,Club/Spade Heart/Diamond act like binary I use temperary audio loop to keep the 8 digit binaryish string in my head until I convert it to be the last image. I also don't need to add new images to my letter pair system since 2^8=256.
This system did take alot of effort to make but not as much as the millenium PAO or the Ben System would. The audioloop is very weak and a short term tool that really only helps me translate the last image more quickly. I have mostly worked on finishing and practacing this over the last few weeks so I am still getting more fluent in the system.
Also, Do others use this to practice memo http://www.dave-reed.com/Nifty/randSeq.html I have found it usefull.
Last edited by A Leman; 10-05-2012 at 11:48 AM. Reason: fix the link
ahhh okay.
You mean like this? http://www.speedsolving.com/forum/sh...Stories-ThreadAlso, Do others use this to practice memo http://www.dave-reed.com/Nifty/randSeq.html I have found it usefull.
I just came up with one that works surprisingly well while trying to learn blindfold solving. You just use multiple sentences of this for both cycles:
(first colors)s who eat (second colors)s get (third colors)s for christmas. Just use this cycle over and over until you finish with the edges. Then use this cycle for the corners. It is easy to memorize and you can track how many of each there are without counting so you can find parity. Just note flipped edges and twisted corners. If you use positions instead of colors, it also works. If you don't have a multiple of three, just think (second to last colors) eat (last colors). If there is one more than a multiple of three, say that they get (second to last colors) and (last colors) for christmas. This is very easy to memorize.
This is like a more restricted version of PAO, and would be more efficient if you used a letter scheme and had each word contain two letters.
What not to do
I want to talk about this 26 almost random lists of 26 not particularly ordered people/characters (from books, movies, TV, manga) system. I thought it would work since people say that associations to pegs only matter when you start and that you need to focus on using strong images.
I obviously misinterpreted this advice.
DO NOT MAKE A RANDOM LIST OF 676 CHARACTERS AND TRY TO USE IT FOR MNEMONICS!!!
I was spending every night drilling the list and tapping the stickers to memorize the list for over a month and spent further months practicing it. The whole time I was convincing myself that I would get a better result than letter pairs in the long run. This is probably the BIGGEST cubing mistake I have ever made. Don't make associations were there are none. The craziest thing is that this system actually worked until I stopped Bld for a while to learn some speedcubing algs. It took about 3 weeks without practice for most of the associations to disappear and the rest were very difficult. Furthermore, I found this out right before a competition and did terribly (4x4: DNF DNS DNS, 3x3: DNF DNF DNF).
If this stops anyone else from making such a list, I will consider this post a success because
CE RH EB JF US
should never mean "Ash Ketchum in piccolo's cape uses Avada Kedavra on a Wartortle wearing Gandalf's hat"
I thought this would solve similar letter confusion and rare letter pair difficulty, but I created a whole host of problems instead.
I was stupid and tried to brute force mnemonics. Don't be like I was. I am back to using phonetic letter pair associations and I am much better off.
I tried something fun yesterday, and I considered it a failure for what I was trying to achieve but I also found it very motivating to improve my memory method.
My girlfriend had a friend from college visiting town this weekend and the three of us spent a lot of time together just hanging out. At one point we went to WonderWorks in Orlando, which is sort of a science museum/learning center geared towards kids and adults. One of the games was a giant version of the memory game Simon. It had 5 buttons, each showing a different color. At one point Katie and her friend were playing a two player game elsewhere so I decided to give the Simon game a try using my memory images and journeys.
The record for the Simon game (which the game displays at all times) was 25, and I figured that would be 13 images in order to beat and set a new record of 26. 13 images would fit into 4 full journey locations each containing 3 images of 2 letters, plus another letter pair image in the start of a fifth location. This is less information to remember than a 4x4x4 cube, so I figured it would be a cinch to beat the record of 25 without much effort.
I was very wrong. I made it to 15 in two tries, and after that we had to leave because I was really there to hang out with Katie and her friend, not to play memory games
I found that my problem was not with the encoding. I could encode as quickly as the machine showed me new colors. Basically I took the first letter of every color and set it as a letter from my letter pairs. (O)range, (B)lue, etc. If I got two of the same color in a row, so (B)lue (B)lue then I would just picture an exaggeratedly large version of my single letter image for B which is Orlando (B)loom. So the memory method was sound for the most part, albeit a little different than what I am used to.
I found that the problem came during recall. The machine did not allow you much time to pause when playing back the sequence after it had shown you a new color added onto your previous sequence. I sometimes would find that I would have a memory recall delay for an image, and as I tried to use my normal techniques to decode what was there the machine would tell me that I timed out and lost. This happened to me twice, and is how I stopped both of my attempts. As I said before, the furthest I made it to was 15 colors in a sequence. This is 7 images and a single letter image after that, which only takes 2.5 locations, about as much to remember as a 3x3x3 cube.
This was a huge wake-up call for me, as it showed me that under intense pressure my failing is not encoding, but decoding. Does anyone have any ideas on how to improve your ability to decode when memorizing quickly, to avoid the recall delays? I know that is a HUGE question, and is the kind of thing we are all searching for. I am open to ideas, though, if anyone has any.
Last edited by cmhardw; 01-06-2013 at 09:35 AM.
My 4bld started as a very formulaic memo, with letters made into sentences, placed into locations in rooms. My coding and decoding times improved when I decided not to be so stuck in my set ways of memorising things. After all, some cycles are so easy to remember visually that coding into roman rooms is very inefficient. I stuck to memorising edges 4 pieces at a time into each location, but I started to occasionally memo 4 pieces visually instead of with letters, when the targets seemed easy to do so. My centres were always fairly flexible in terms of number of targets to memo in each locations, but again I tended to start using more visual and sometimes numbers as well (following A=1 B=2 etc). The numbers were particularly useful when I got 3 targets (or more) in a row that would clearly make a nice number sequence (e.g. constant difference). For example, BGLQ isn't so pretty to memorise with letters, but for me it was obvious from their position on the cube (before even thinking of the letters) that they would make a simple number sequence 2-7-12-17.
This memo (that I called 'freestyle memo') was very easy for me to recall as I still had set locations in each room, but some locations I would have remembered that I memorised the 4 pieces visually, or 4 pieces with numbers. And if I had memorised e.g. 4 centres with numbers, 5 centres visually and 4 edges visually, it was very easy to remember what the memo was because there weren't many cases of it. This also reduced the number of letters/sentences I had used and therefore those were also easier to recall.
I encourage 'fast' people to experiment with a more freestyle type of memo because that accounts for much of what got me from a 2 minute memo to 1:20
Posting Permissions
Reply With Quote
Bookmarks