miniblog.

← Back to all posts
Nov 11, 2017 at 00:38
C++ compiler beating hand written assembly: https://stackoverflow.com/questions/40354978/why-is-this-c-code-faster-than-my-hand-written-assembly-for-testing-the-collat If you haven't measured the cycle time of the different instructions, the optimiser can often do better!
performance - Why does C++ code for testing the Collatz conjecture run faster than hand-written assembly? - Stack Overflow
performance - Why does C++ code for testing the Collatz conjecture run faster than hand-written assembly? - Stack Overflow

Related Posts

May 17, 2025 at 11:05
5
I've written difftastic packaging instructions: https://difftastic.wilfred.me.uk/packaging_difftastic.html The different distros have taken different approaches, so I'm trying to help with common gotchas — don't forget the man page! Feedback welcome, especially if you've ever packaged something 🙂
Packaging Difftastic - Difftastic Manual
The manual for difftastic, the structural diff tool
Dec 10, 2022 at 22:45
Computing optimal 8501 instructions for rotations using an SMT solver and Racket with Rosette:
Synthesizing optimal 8051 code — whitequark's lab notebook
Sep 19, 2022 at 06:20
A decent part of VMs being faster than AST interpreters is just memory layout AIUI. VM instructions are largely flat arrays, so there's less pointer chasing.