The Hidden Physics of Data Types (With Benchmarks) скачать в хорошем качестве

The Hidden Physics of Data Types (With Benchmarks)

Most developers default to using standard 64-bit integers for everything. But under the hood, this one-line decision can quietly bottleneck your CPU. In this video, we dive into the low-level physics of data types. We explore how modern CPUs handle division, why the ALU struggles with 64-bit math, the exact assembly instructions (DIV vs IDIV) running your code, and why uint32 is the undisputed champion of performance. Plus, we look at a massive exception for Apple Silicon (ARM) users, and how the V8 Engine secretly optimizes JavaScript code so you don't have to worry about it. Whether you are writing Go, C, or Node.js, understanding what the hardware is actually doing separates good developers from great ones. Chapters: 0:00 - Intro 0:31 - Signed vs Unsigned 0:54 - 16 bit Range of signed vs unsigned 2:52 - Why CPU can't work on RAM 3:24 - CPU & Register architecture 4:18 - How operations work on registers 4:58 - Live Benchmark (Go language) 8:12 - CPU clock cycles 9:28 - Division in depth (DIV vs IDIV Assembly) 11:08 - Practical Benchmark (int64 vs uint32) 13:19 - Effects of choosing data type 13:40 - How JavaScript & The V8 Engine Cheat 14:10 - Closing Connect with me: Twitter/X: https://x.com/XronTrix10 GitHub: https://github.com/XronTrix10 #softwareengineering #computerscience #programming #golang #nodejs #cpu #architecture #register #coding #benchmark #cprogramming