The super position 2 card system


  • The number value combination of 2 cards is translated as 1 object (13 * 13 = 169 different objects);
  • the 2 suits of each combination are translated as 2 sub locations of 2 consecutive locations at which the object is placed; so every object is placed twice;
  • Object X is placed at location X and (X+1); therefore the total number of locations (each with 4 sub locations representing the 4 suits) for 1 deck of cards is: (52/2= 26) + 1 = 27 (= 27 * 4 = 108 sub locations).


value translation (keep in mind that I’m Dutch so you may need to adjust to your own language):

A = i (A is basically 1, which I traditionally translate as i because of shape)
J = j
Q = kw (queen is phonetically written in Dutch as kwien)
K = k
2 = n
3 = m
4 = a (I normally use 4 = k in number system, but I also have K = k)
5 = s
6 = B
7 = t
8 = f
9 = g
10 = l

first 4 locations (and sub locations) of memory palace:

1 = front door of house:

  • :heart: = on top of door;
  • :diamonds: = under the door;
  • :spades: = in front of door;
  • :clubs: = behind the door.

2 = bed in bedroom:

  • :heart: = in the middle of the bed;
  • :diamonds: = under the bed;
  • :spades: = against the head board of the bed;
  • :clubs: = on the edge of the bed.

3 = couch in living room:

  • :heart: = on top of couch;
  • :diamonds: = on the middle of the couch;
  • :spades: = under the couch;
  • :clubs: = on the side arm support of the couch.

4 = desk in living room:

  • :heart: = desk chair
  • :diamonds: = under the desk;
  • :spades: = in the middle of desk top;
  • :clubs: = on the edge of desk top.

3 2-card combinations;

:spades: Q :clubs: K = a duck (kwak = one of the (Dutch) cousins of Donald Duck) in front of door and on the edge of the bed.

:heart:A :clubs:9 = a toilet brush (ig = an expressions of disgust) in the middle of the bed and on the side arm support of the couch.

:spades:7 :diamonds:3 = a rabbit (tam = Dutch for domesticated) under the couch and under the desk.


what difference? u just simply put one card on two loci?

I don’t understand this question. Are you implying that this system already exists or that it doesn’t make a difference in terms of performance compared to other systems (that take a simular amount of training)?

u just simply put one card on two loci?

No, I don’t; I put one object that represents the value of 2 cards in 2 locations, each of which representing the suit of 1 card.

Perhaps I should elobarate somewhat on the reasoning behind this system.

2 locations for 1 object? Yes and no.

I once had to replace a broken toilet seat and to make a long story short, I ended up being stuck between the bottom left side of the toilet and and the bottom right side of the washing machine. Was I really in 2 places at once and therefore braking the natural laws of this universe?

In a simular fashion it would almost always be possible to place an object at 2 consecutive locations, without this being 2 independent memory tasks and without having to use twice the amount of time it takes for a single location placement.

in my examples:
:spades:Q :clubs:K: A duck that jumps from the front of the door (when the door is opened to the inside) to the side of the bed and ends up on the floor with a concussion.
:heart:A :clubs:9: A really long toilet brush that tips over with the stick in the middle of the bed and the dirty end on the arm support of the couch.
:spades:7 :diamonds: 3: a scared rabbit hiding under the couch and under the desk.

tbh, if u go through too much loci, it ll be ended up wasting too much time(*and maybe ghost images too). But i cant refuse thats a nice idea. Perhaps i ll try this system to see its really practical.

You mention 2 possible objections, that I will adress one at the the time.

usage of loci

The total number of “normal” loci is 27, which is 1 more than the Ben system. The basic idea of the sub locations is that the suit tells you if it’s on top, under, in front of or behind in a systematic fashion (the examples I gave in my original post where a bit inconsistent with this systematic approach; my actual memory palace has loci that are a bit difficult to put in words; depending on the location I might also have to divert from this system like with a desk board hanging on the wall as there is no behind the desk).

It is undeniable that for every 2 cards I have to place 1 object in 2 locations. It is my guess that this will take more time than 1 object in 1 location like the Ben system. But keep in mind that no matter how much you or Ben trains, the translation speed of 2 playing cards into 1 object will very likely be slower than that of the combination of number values of 2 cards into 1 object. According to Ben, it took him several months to get his translating up to speed; I am confident that I can learn my 169 images in 1 day. On top of that I would like to argue that placing 1 object in 2 consecutive (sub) locations will likely take less time than 2 objects in 2 locations (1 object per location) as in most 1 card systems (of course you can put multiple cards in 1 location as well; I have never been good at this) because they are not independent memory tasks.

More on loci, possible variations of the system

When I came up with this system I also immediately thought of 2 variations, both of which use the 2 playing cards as 1 object in 1 location approach.

variation 1: reverse memory palace with 169 loci, 26 objects (4 variations each)

In this system the number values of each 2 card combination is translated as 1 location of a 169 loci long memory palace. The first suit is translated as 1 of 4 variations of an object (that is part of a prememorised sequence of 26 objects) and the second suit is translated as a sub location of the location.

variation 2: number value of 2 cards + 1 suit = 1 object, second suit = sublocation

In this system you need 13 x 13 X 4 = 676 objects and 26 locations (each with 4 sub locations).

Ghost images

I don’t really know why you would think that this system is more likely to suffer from ghost images than other systems. As a possible solution to this problem in my system, you might train the suit translation (in sub loci) seperately from the object translations and placement in (a different) memory palace.

Is your system significantly faster than the Ben system and if it is then I will be using it,


Just to make sure I 'm not sending out the wrong message: I’m not claiming that my system (as a whole) is faster than the Ben system. Having said that, it seems obvious to me that with a relatively short amount of practice like a couple of days, my system or any simple system for that matter, will likely be faster than the Ben system, simply because of the huge training investment that is required to just be able to use the Ben system at all.

Learning the 169 objects representing all 2 card value combinations is very easy (see: Playing with unusual number/card translations). So the real question is, how fast can a person put those objects at 2 consecutive sub locations (or connect the 2 sub locations with that object).