[frading]

Yes, minimizing clutter was a priority, and still is. I've tried to avoid those arrows for a long time, but the pawn movements wasn't clear enough without them. And I agree it's not the optimal display.

Displaying them on some tiles only is something I have not thought of. My first reaction is that it may feel harder to read, as tiles with an arrow may look like they would act differently than those without? We may induce the idea that it has a different purpose. So for now, I'd prefer a solution that would look the same on all tiles, I think that's much easier to parse. But then I should probably give this a try before rejecting it.

You do have a good point regarding the wood texture, I've tried other textures, but couldn't find a better one yet. But I'm sure that's possible. Nevertheless for now, you can also display different skins: https://imgur.com/a/zitKlpz

And thanks a lot for mentioning the four colour theorem. I'm looking at its wikipedia page now. I'm sure I came across it many times before, but I did not think about it when tackling this part of the problem. My implementation is here is anyone can spot my mistake: https://github.com/polygonjs/polygonjs/blob/master/src/engin...

[wizzwizz4]

The five-colour theorem was first proven in the late 19th century. The known proof of the four-colour theorem is non-surveyable. There's probably no small mistake in your implementation, and the solution you should go with is unlikely to be a fully-general algorithm. (I'd suggest starting from the border on the outside, greedily filling in as many choices as are forced (up to isomorphism), then randomly choosing a candidate breadth-first or depth-first until you've found a four-colour solution.)

[frading]

actually, I realise now that the implementation I shared in my previous comment is a generic one. At some point, I decided to make a variation of it, specialised for this board. The reason wasn't the number of colors, though, it was to have better control of the colors of the bottom and top rows.

And I also realise that, contrary to what I said above, I do not set the colors on the left and right sides. So that will be an easy modification.

After that, I do go breadth first. But I do so by looking at all neighbours of the current tile, simply by taking them in the order given by my internal graph. Instead I should just take those that are connected to another tile that has been already visited.

So thank you for those advice, that's super helpful, I really appreciate. I had to look up "non-surveyable" as well, I'm learning a lot here. That said, I'm having a hard time understanding "as are forced (up to isomorphism)", is that something you could clarify?