After dipping into this system, I have to admit it seems a tad too complicated. The ## pairs are quite doable but together with the rest, where some have the 3rd digit of the image defining the suit and some having the 1st, it can get rough. I mean, maybe with a lot of practice, but for me as a non-competitive mnemonics utilizer, it’s a bit too much.
That said, I’d still be interested to hear if anyone have had more success with this or a similar system where less than 1000 images are used.
There are plenty of people who have had major success, including record elite level times, with a simple single-card system, where each card is represented by a single intentional element either on its own or compounded within an image or scene, like in a PO (person/object), or PA (person/action), or PAO (person/action/object) system. This requires at most 156 unique associations to learn (3 per card with a PAO system.) These associations can be trained to fluency much quicker than with a 2-card system where there are exponentially more associations to learn. This is the big upside of a single card system. Fast onboarding and ready for use fairly quickly. The “downside” of the single card system is that you’ll have to read and intentionally visualize 52 total elements to get through a full deck. if you can do this at about one second per card, you can memorize a deck in just under a minute. Your speed limit is mostly affected by how quickly you can see the card and “read” it as its element and then combine them as needed into a scene.
Enter the two-card system.
If you can figure out a way to encode two cards in a single intentional element, then in theory given the same recognition fluency, you can cut your times in half because you only need to read and encode 26 elements. If you can read and visualize each pair in about a second, you can get through a deck in under 30 seconds. One major tradeoff, as you’ve surely realized, is the size of the system and the time and effort to train it to the same fluency level.
Single card systems can be SIMPLE as you’re only looking to compress 2 pieces of information into each element, (card value and card suit.)
The challenge with creating any 2-card system is that there has to be a way to indicate four unique pieces of information (card 1 value, card 1 suit, card 2 value, card 2 suit) in a SINGLE mnemonic ELEMENT. Not just a single COMPOUND IMAGE or SCENE that contains multiple intentional elements, but a SINGLE INTENTIONALLY ENCODED ELEMENT. (Those terms in caps are important when it comes to analyzing how effective a system is. More on them later.)
As someone who has created a few two-card systems, I can speak a little bit to the process and the obstacles and limitations.
First is the number of possible pairs. There are 2652 possible pairs that can be made from a single deck of cards. (52 options for the first card, 51 options for the second. 52 x 51 = 2652.) This alone is a big challenge. Each one of these 2652 elements needs to be unique and memorable so that you can easily visualize each one and not confuse any with any of the others when trying to recall the deck.
Then there is the problem of values (there are 13 of them) and suit combinations (there are 16 possible two-suit combinations.) Here is where the problem of consistent structure comes in.
If you want to map your card pairs onto a number system, there are big difficulties which usually end up forcing these “alternate reading structures” in order to differentiate pairs. It’s easy to map the 10, Ace, 2, 3, 4, 5, 6, 7, 8, 9 values to single digit numbers 0-9, but what do you do about the Jack, Queen, and King? This is where traditional number mapping starts to break down. You either need to have those values “share” associations with other values, or you need to assign them an indicator outside of those numbers so that they don’t get misread.
Similarly with the suit combinations, you can assign ten of the suit combinations to numbers, but what do you do with the other six?
This is the problem people have tried to solve by proposing different reading or phonetic structures depending on which value or suit appears first or second in a pair. It makes it a bit clunky to execute, but it’s kind of a product of the properties of a deck. If there were only 10 values, and 10 suit combinations, you could easily assign each one a simple number and be done with it. But because of these “extra” properties, sometimes you’re forced to adjust the structure depending on when those show up.
In order to make the reading structure consistent, you either need to assign a combination of numerical and extra-numerical associations to complete a full 2652 element system, or start dividing the total used in half via a “block” system.
If you divide once and create a “2-block” system where every element represents two possible card pairs, you still have over 1000 associations to make (1326 to be exact) and will still need to find some way to encode the 326 “additional” elements beyond the 1000 that you can map to numbers. Additionally, you’ll need to find a way to indicate which of the two possible pairs is encoded with your element, by adding an additional INTENTIONALLY ENCODED attribute to your image or scene.
If you divide again, you end up with 663 elements, each representing four possible pairs. This easily “fits” into a 3-digit number system, but the tradeoff (and it is a big one) is that you need to encode TWO additional intentional attributes into each image in order to indicate which of the four possible pairs it represents.
Lets take a quick detour to look at what I call INTENTIONAL ELEMENTS.
I consider a MNEMONIC ELEMENT to be a single attribute that contains encoded information. The fundamental particles, the quarks of memorizing. These elements can be in the form of a person, an action, an object, a color, a feeling, a costume, a mode of transportation… whatever you might think of that can somehow represent some piece of data. A mnemonic IMAGE can sometimes be made up of multiple elements, like if a color represents the card value and an object represents the suit, then something like a “BLUE PICKLE” might represent the 2 of Hearts. Now you might imagine a blue pickle and think “ok, thats a single thing. A blue pickle. I’m picturing this one blue pickle. I’ve got a really simple picture in my mind and I’ve successfully encoded this card identity into a single image!” but in reality you’ve had to recognize and INTENTIONALLY consider and create two different elements to form your image. BLUE is one “intentional element”, PICKLE is the other.
Compare this to a system where “PICKLE” fully represents the 2 of Hearts, maybe through the sounds that the word uses or by a connection to a two-digit number that can represent value and suit. This is “better” because it only requires recognizing a single intentional element. This will typically take less time and less brainpower than recognizing and converting both BLUE and PICKLE elements. This is what happens with a single-card system. One intentional element per card. PICKLE is 2 of Hearts, DOG is 3 of Diamonds… etc.
Now compare this to an approach where “PICKLE” can represent two possible PAIRS of cards via a two-block system. Is there an advantage here in terms of data compression and INTENTIONAL element usage vs. the single-card system? Well, yes, but maybe not as much as you may think… In order to indicate which pair out of the two possibilities the pickle represents, an additional INTENTIONAL element needs to be added to what you create in your mind. This is some kind of marker or property or position or whatever it is, that needs to be correctly identified, visualized, and integrated into your image or scene. If you need to add an indicator for each pair that you come across, then it is not really any better than a single-card system. You have two intentional elements required to identify the two cards in a pair. Now, there are techniques like variable image stacking where you only need to intentionally indicate a marker at the end of a compound scene, so your data compression rate can vary, and be better than a simple 1:1. If you combine 2 pairs/elements and then apply the block indication, you’re achieving a 4:3 card to intentional element compression ratio. 3 pairs per scene can get you 6:4. But, there is a progressively greater tradeoff that occurs here as well. The larger your scene gets before needing to indicate a block, the better your data compression is but the tougher it gets to hold onto that scene accurately for recall. If you “lucked out” and had all 26 pairs fall into the primary block, you wouldn’t need to add any intentional block indicator, but you’d have to stack all 26 in the same scene, which is exceptionally difficult. (This is assuming variable image stacking is the block-indication method.)
Now, a “true” two-card system where there is a single, unambiguous, unique element for each possible pair of cards gives you a distinct advantage in that PICKLE now represents say the 2 of Hearts AND the 3 of Diamonds. No extra intentional indication needed. If I recall a pickle, I can decode it into two cards. With this structure, I can encode a full deck with 26 elements and no extra intentional stuff needed. But I have to learn 2652 unique associations and practice them until they are fluent. This is usually more than most want to tackle.
After all this… consider the “double-2-block” style system where only 663 associations are needed, but you need to do a double intentional indication to identify which of the four possible pairs an element represents. You can probably see the downside. Even though at a basic level, you can represent all 2652 pairs with “only” 663 elements, the actual number of intentional things you need to precisely include in a mnemonic image in order to correctly encode the actual pairs involved is going to be more than that. So ultimately you’ve increased the workload in application in order decrease the frontload of learning fewer associations. I don’t think this is a trade worth making…
So, back from that long detour…
I think it is very unlikely that competitive times can be achieved practically with a double-2-block style system, due to the additional encoding of intentional markers that is needed. I know that there are folks who effectively use a standard 2-block system who can hit 15 seconds for a deck. That extra indicator element has become so fluent it has basically become and INCIDENTAL element that is subconsciously and automatically added during memorization, but this is the result of LOTS and LOTS of drilling and practice, such as you’d expect from any elite-level athlete. I also know there are world champions and record holders who use a “true” 2-card system, and some who use a “basic” single-card PO or PAO. The common factor for all of them is massive dedication to practice and improvement.
It has been said many times on the forum, but it’s true: the system that you use is less important than the dedication you have to learning and implementing it.
You can have a “perfect” 2-card system where every pair is perfectly mapped to a single syllable and each one reads intuitively and the structure is consistent, and you can get absolutely crushed by someone who has drilled their single-card PO to instinctive reflexive speed if you don’t spend the time needed building your fluency to that same level. It will always be about effort. The larger systems require more effort up front to build and learn in order to gain a benefit of faster times due to fewer intentional elements needed for a deck. The smaller systems require less learning time but more effort to build up more speed in order to encode the larger number of elements needed quicker… So it’s up to the individual on what their preference is for how they want to approach that workload.
There are all kinds of card and number systems out there and it can be fun to explore the possibilities. If you like tinkering and can approach system building as a mental exercise or brain game almost like some people enjoy sudoku or crossword puzzles then it can be really enjoyable, even if you don’t actually end up using them.
So… ALLLL of that said… Not sure if this was what you were looking for, but if you’ve made it through this war-and-peace length post, hopefully you took something away from my brain dump!
Cheers for a good general, Wikipedia entry worthy write-up! I was indeed referring to “double-2-block systems” in my question. So basically the following part…
I think it is very unlikely that competitive times can be achieved practically with a double-2-block style system, due to the additional encoding of intentional markers that is needed.
… confirms the hunches I had about my wonderings.
There are all kinds of card and number systems out there and it can be fun to explore the possibilities.
Fully agree.
Not sure if this was what you were looking for, but if you’ve made it through this war-and-peace length post, hopefully you took something away from my brain dump!
Not all of it was, but as I’m no stranger to literature classics, this wasn’t at all a bad wall of text. In fact, on the contrary!
I havent read anyone describing the 4 block system being achieved by four different loci strategies:
Linking / Stop and move to next / Stop and move to next +1 and next +2
Would this work or am I completely missing something? This doesnt seem to involve any change to the images, just change in the locations.
So as it relates to cards - Phonetics could involve converting all black cards to a to z, and converting all red cards to A to Z.
All lower case a - z is keep linking, all capitals A - Z is encode and move to next, capital and lower case (Az) is encode 1 image and then skip 1 location, and lowercase and capital (aZ) is encode and skip 2 locations
You have 26 x 26 images = 676
Say you draw ace of clubs and king of spades - and this converts to az. You use your az image, and keep linking. You then draw a 2 of diamonds and queen of hearts (BY). You add your by image, and then move to next location.
You are now up to second location. Your third card pair: 3 of clubs, and jack of hearts (cX). You add your cx image and then skip forward two locations.
When extended to 3 letters (a, b, c) there is for someone perhaps who has the potential to improvise words across a 17576 set of possibilities. Improvisation seems a feature of the way the 4 digit system is used by Simon (consonant, vowel, consonant, consonant), with some of the final consonant letters potentially involving some level of category improvisation.
Two letter images seem like they would be quite useful for other purposes as well. I am not sure if all alphabets could use this (such as languages with alphabets less than 26 characters).