Talk:Galactic algorithm

Multiplication algorithm

I think this is WP:SYNTH since in no place does that paper claim that $n_{0}=2^{1729^{1}2}\approxeq 10^{2.148\cdot 10^{38}}$ is the lowest such number, and indeed claims that d = 8, with a substantially smaller value of $d^{12}$ is sufficient. We should instead be vaguer, perhaps saying it needs to be bigger than can be written in the observable universe.--Jasper Deng (talk) 00:30, 14 August 2019 (UTC)[reply]

This is reasonable. I replaced this with a less spectacular but better supported bound, based on the fact that the algorithm is built upon a 1729 dimensional transform. Since each dimension must have a size of > 1, this requires data of at least 2¹⁷²⁹ bits even to fill the array. LouScheffer (talk) 00:18, 15 August 2019 (UTC)[reply]

Citation?

"Computer sizes may catch up to the crossover point, so that a previously impractical algorithm becomes practical."

This makes no sense and isn't cited. Is it correct? Galactic algorithm are impractically because they involve data that isn't in practice used? Increased 'Computer sizes' won't change that. — Preceding unsigned comment added by 61.68.214.83 (talk) 11:02, 5 October 2019 (UTC)[reply]