Work

The 697 × 611 rectangle needs exactly fourteen squares

Year: 2026
Type: preprint
Abstract: Let f(m, n) be the least number of integer-sided squares that tile an m × n rectangle. Answering a MathOverflow question, Ed Pegg Jr reported in 2017 that 697 × 611 was the one rectangle with sides at most 760 that might violate the conjecture that scaling a rectangle never lowers f: its best known tiling used 17 squares, while 1394 × 1222 could be tiled with 16. We show that f(697, 611) = 14, and indeed f(697t, 611t) = 14 for every t ≥ 1. The fourteen-square tiling was posted by the author on MathOverflow in September 2026; the lower bound is new. It comes from a complete enumeration of squared rectangles with at most 13 squares through the electrical networks of Brooks, Smith, Stone and Tutte, generated as quadrangulations by plantri, which finds no tiling of any rectangle similar to 697 × 611. The currents are verified in exact integer arithmetic; extended to 14 squares, the enumeration reproduces every value of OEIS A219158 that is at most 14 (65,231 rectangles), and a planted error is caught. This is a computer-certified result, without a Lean formalisation. This record is identical to the folder squared-rectangle-697x611 at tag squared-rectangle-697x611-v2 of https://github.com/jvvk/mathematics. Version 2 of the preprint; version 1 (9 October 2026) was released on GitHub only.
Language: English
Open Access status: green
OpenAlex ID: W7236727406
Cites: 1
Cited by:
Related to: 9