SQLite's own performance notes (fasterthanfs.html) and PostgreSQL's 8KB default block size exist because of exactly this amortization math — a bigger read spreads a fixed seek cost over more bytes. MySQL InnoDB's 16KB page size makes the same tradeoff, which is why none of these engines default to 512-byte pages despite the extra memory a larger page wastes on small rows.