Brainstorming a Full 10,000-image Mnemonic System
This post is a brainstorm about how a 10,000-image mnemonic system might look. I’ve created a script that displays all of the assignments. (The link is below.) It’s based on my existing systems, so it wouldn’t be difficult to expand, except for the unimaginable time commitment of creating and maintaining 10,000+ images – something that I will likely never have time for. I built the script on request, and decided to post the explanation here in the blog.
If you want to skip ahead to the actual system, click here. It’s split into different pages with up to 1,000 images each.
The assignments would follow my existing ones for the most part. Instead of the consonant-vowel-consonant of the Ben System, it would be consonant-vowel-consonant-vowel.
- 0 — s/z
- 1 — t (not d)
- 2 — n
- 3 — m
- 4 — R
- 5 — L
- 6 — b
- 7 — k (not g)
- 8 — v/f
- 9 — p
The system seems very elegant to me. I wish that I could use it. The face cards add 144 extra images* that don’t fit neatly into the 10,000, and I think this is it’s largest flaw (other than its size). These are separate from the main 10,000 images in order to keep the 10,000 images perfectly regular.
The assignments that I came up with match my current one-card images:
- Jack — j/ch/sh/g (g as in George)
- Queen — g as in green
- King — d
Card values are consonants and card suits are vowels. For more brainstorming on cards and sounds, see my previous posts, Thoughts on Expanding to a 10,000 Image Mnemonic System and More Thoughts on a 10,000-Image Memory System.
(* 12 face cards multiplied by 12 face cards)
The vowels are a rearrangement of the Ben System vowels:
- 0 — o as in bow because 0 looks like o.
- 1 — i as the ee in eel. 1 looks like an i.
- 2 — u as the vowel sound in two.
- 3 — aa as the vowel in cat.
- 4 — a as in father. 4 looks like A.
- 5 — ai as the vowel in five.
- 6 — ih as the vowel in six.
- 7 — e as the vowel in seven.
- 8 — ei as the vowel in eight.
- 9 — uh as the u in lullaby.
Card suits are mapped to the digits:
- Spades — 0 because of the s.
- Hearts — 4 because of the R in hearts
- Diamonds — 1 because of the d sound, which is close to t.
- Clubs — 7 because of the k sound
Some four-digit numbers in consonant-vowel-consonant-vowel format:
- 0000 = SO-SO
- 0001 = SO-SI
- 0010 = SO-TO
- 4279 = RU-KUH
- 5637 = LIH-ME
- 9365 = PAA-BAI
I already pronounce numbers like this in my two-digit system. The difference here is that a four-digit number would be a single image rather than two images.
Image © ESO and used under CC BY 4.0 License
It seems easiest to convert each decimal digit to its equivalent binary.
- 0 = 000
- 1 = 001
- 2 = 010
- 3 = 011
- 4 = 100
- 5 = 101
- 6 = 110
- 7 = 111
So the image for 4507 would match 100-101-000-111.
Since there are 30 binary digits per row in competitions, it might be easier to memorize them as 4x3 binary grids. 100-101-000-111 could also be chunked like this:
100 101 000 111
10 binary grids would fit in one competition row.
See the binary memorization systems page for more information about binary grids.
It would probably be easiest to find images by creating categories. The first two digits could represent a category. Examples:
- 00 — SO — “sew” — 100 images related to sewing, cloth, fabric
- 01 — SI — “sea” — 100 images related to the sea
- 02 — SU — “zoo” — 100 images related to animals (“zoo” is animal in Greek)
- 03 — SAA — “satelite” — 100 images related to space vehicles, telescopes, rovers, etc.
- 04 — SA — “saw” — 100 carpentry tools
It’s such a large system that I probably would never attempt it. One way that it might be doable in stages is if one used the images from 0000-0999 as a three-digit system at first. Then it could be gradually expanded from there.
For example, the 0456 image could be used for 456 until all 10,000 images were memorized.
The script is here. This 10,000-image idea is just an untested brainstorm, but maybe someone can extract a few useful ideas from it. As far as I know, Simon Reinhard is the only person who has a working 10,000-image system, and I don’t know much about how it works.
More Thoughts on a 10,000-Image Memory System - Art of Memory 10 Nov 2021
[…] to convert my current system into 10,000+ images if I were to do it. (Update: see the newer post on brainstorming a 10,000 image system which includes the structure of all 10,144 […]