https://www.youngju.dev/blog 천천히 올바르게. AI Researcher & DevOps Engineer Youngju's tech blog. GPU/CUDA, LLM, MLOps, Kubernetes AI workloads, distributed training, and data engineering. ko fjvbn2003@gmail.com (Youngju Kim) fjvbn2003@gmail.com (Youngju Kim) Sat, 20 Jun 2026 00:00:00 GMT https://www.youngju.dev/blog/2026-06-20-floating-point-explained.en https://www.youngju.dev/blog/2026-06-20-floating-point-explained.en In almost every language, 0.1 + 0.2 produces something that is not 0.3. It is not a bug; it is the designed behavior of IEEE 754. This post unpacks the truth about floating point from the ground up: how binary floating point stores numbers, why decimal fractions cannot be exact in base 2, how to compare with epsilons and handle money, what NaN, infinity, and negative zero really are, and how errors accumulate. Sat, 20 Jun 2026 00:00:00 GMT fjvbn2003@gmail.com (Youngju Kim) floating-pointieee754fundamentalsnumerics https://www.youngju.dev/blog/2026-06-20-floating-point-explained.ja https://www.youngju.dev/blog/2026-06-20-floating-point-explained.ja ほとんどすべての言語で、0.1 + 0.2 は 0.3 ではない値を返します。バグではなく、IEEE 754 の設計どおりの結果です。2進浮動小数点がどう数を格納するのか、なぜ10進小数を2進で正確に表せないのか、イプシロン比較とお金の計算はどうすべきか、NaN・無限大・負のゼロの正体、そして累積誤差まで、浮動小数点の真実を土台から解きほぐします。 Sat, 20 Jun 2026 00:00:00 GMT fjvbn2003@gmail.com (Youngju Kim) floating-pointieee754fundamentalsnumerics https://www.youngju.dev/blog/2026-06-20-floating-point-explained https://www.youngju.dev/blog/2026-06-20-floating-point-explained 거의 모든 언어에서 0.1 + 0.2가 0.3이 아닌 값을 냅니다. 버그가 아니라 IEEE 754의 설계 결과입니다. 2진 부동소수점이 어떻게 수를 담는지, 왜 10진 소수를 정확히 표현할 수 없는지, 엡실론 비교와 돈 계산은 어떻게 해야 하는지, NaN·무한대·음수 0의 정체와 누적 오차까지, 부동소수점의 진실을 밑바닥부터 풀어냅니다. Sat, 20 Jun 2026 00:00:00 GMT fjvbn2003@gmail.com (Youngju Kim) floating-pointieee754fundamentalsnumerics