https://macworks.dev/tags/fintech/Recent content in Fintech on MacWorksHugoenhttps://macworks.dev/docs/week/hackernews/hackernews-2026-05-27/Mon, 01 Jan 0001 00:00:00 +0000https://macworks.dev/docs/week/hackernews/hackernews-2026-05-27/<h1 id="hacker-news--2026-05-27">Hacker News — 2026-05-27<a class="anchor" href="#hacker-news--2026-05-27">#</a></h1> <h2 id="top-story">Top Story<a class="anchor" href="#top-story">#</a></h2> <p><strong><a href="https://www.thonking.ai/p/strangely-matrix-multiplications">Matrix Multiplications on GPUs Run Faster When Given &ldquo;Predictable&rdquo; Data</a></strong> Matrix multiplications are supposed to be fully deterministic, executing the same number of operations and memory accesses regardless of the tensor&rsquo;s contents. Yet, initializing matrices with zeros or ones yields measurably faster performance than using normally distributed random data. The culprit is dynamic switching power: predictable data minimizes transistor state flips, reducing power consumption and preventing the GPU&rsquo;s Voltage Regulator Module from aggressively throttling clock frequencies under heavy load.</p>