# Louis Verhaard's eii

> Louis Verhaard's eii won the only prize the puzzle ever paid, but it shipped as a Windows binary with no source. What his method is, why the original will not run here, and where a faithful reimplementation of it lives.

- Canonical page (with interactive figures/demos): https://eternity2.dev/research/lab/experiments/louis-verhaard/
- Updated: 2026-07-15

---
This section is about [Louis Verhaard's](/research/people/louis-verhaard) eii,
the solver that found the 467 and won the puzzle's only prize. Unlike McGavin
and Blackwood, he never released source, so his own program cannot be run here.
This page records his method and that gap; the runnable engine is a
[reimplementation](/research/lab/experiments/louis-verhaard/verhaard-reimpl),
which lives here beside the method it rebuilds, bylined to Raphaël Anjou because
the code is his even though the idea is Verhaard's.

## Pages in this section

- [Verhaard's eii: the solver that won the only prize](https://eternity2.dev/research/lab/experiments/louis-verhaard/eii) — The engine behind the 467, the only Eternity II score ever paid, read from Louis Verhaard's own mailing-list posts: forward pruning, comb-search fill orders, depth-gated edge slipping, a Markov-tuned slip schedule. And why his own source-less Win32 binary cannot be built or run on this machine at all.
- [Verhaard reimplementation](https://eternity2.dev/research/lab/experiments/louis-verhaard/verhaard-reimpl) — A from-scratch reimplementation of Louis Verhaard's eii method, since his own binary ships no source and will not run here. Set-composition swap-annealing under the 2×2-tiling metric; on the real five-clue puzzle it reaches 438 of 480, single core.