The triangular table view is fascinating. It looks like the periodic table. I wonder if there are number-theoretic lemmas (or at least conjectures?) about what "family" the optimal packing for a given number falls into (like diamond, diagonal strip, two blobs, etc). I didn't see anything when skimming the survey paper linked at the bottom of the site, but I'm sure there's a lot more literature here.
Brief search reveals prior academic research related to sexual orientation and dating apps. Doesn't appear to have done anything in maths before.
But then, why were they the first to issue the correct series of prompts to produce these results?
This would lend credence to the efficacy of using LLMs as tools. If mathematicians in the packing field had used the tool before this liberal arts student, they'd have their names on the record page.
if, like me, you're a non-native english and speaker don't immediately understand what this is about: the page shows for each `n` what's the minimum `s` such that `n` squares with side of length 1 fit in a square with side of length `s`.
what I'm curious about though is what a proof for something like this looks like. and why does it need a proof? not to mention the randomness of some of the `n`s. Math is most of the time beatiful and whenever I see something like `n=11` I think "it looks wrong so it must be wrong" yet it has a proof.
Do you want the graphs with 300 squares to be bigger than your screen, or do you want the graph with 1 square to be 30x30 px for no reason? They're just zoomed.
That's what I mean, I can't imagine why anyone would argue the same thing of graphs in general so I'm curious what the difference makes it so they find it so odd in this specific case.
Both appear to be in use, the author succesfully communicated what they intended to communicate, and the audience (you) succesfully recieved the communication.
The 4 packing takes up 100% of its square; it's trivially optimal. The 6 packing only takes up 2/3 of it, so it's not necessarily obvious that you can't do better.
for me it is obvious. If I am reading s=3 as multiplier of side of smaller square to side of bigger square, which means that bigger square side is 3 times the side of smaller one, than it is obvious that it should be poossible to squeeze 9 small squares into bigger square. It is children puzzle after all. What is not obvious here?
Because it could be possible, as we just saw with 5, that by some rotation of some number of cubes, you could fit six unit cubes in to box smaller than area 9. Since we just saw the unusual result in 5, it is worth verifying.
i believe 4, 9, 16, 25 etc are just subdivisions of the unit square (they're perfect squares)
but the text also says "For the $n ≤ 324$ not pictured, the trivial packing (with no tilted squares) is the best known packing." applying 'trivial' to numbers that aren't perfect squares so iunno
It's two different trivial things. For each, it's just the case which doesn't require doing anything special.
One is trivial proofs, which are where 100% is covered. This doesn't really leave much to prove in terms of whether or not more area can be covered by a different layout.
The other is trivial packings, the very simple type without any tilting or need of gaps between squares. Trivial packings are only sometimes optimal. Of optimal trivial packings, only some can be shown optimal with an aforementioned trivial proof.
https://thejenkinscomic.wordpress.com/2024/12/01/brady-bunch...