TL;DR

Matt Parker made a video about the maths behind Dobble.

It was interesting to see the video and be reminded of something I wrote about three years ago: Some Maths for Dobble.

There were a few things that left me… dissatisfied. As an example, the whole process where you make a circle of cards and use a specific sequence to generate the cards seems a bit as coming out of thin air. mean… why 57 cards? Why those displacements? Matt’s explanation about why it works is fantastic, though.

Dissatisfaction is an engine, though. There are a couple of loose ends that will be interesting to follow.

I was curious about generating the cards with 102 pictures on, leveraging $PG(2, 101)$. Using pg2 it took me about 70 minutes to generate all the arrangements… I’m not impressed by my brute-force algorithm to find all orthogonal sequences… 🙄