Skip to content
View IAA03's full-sized avatar
📈
Optimizing
📈
Optimizing

Block or report IAA03

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
IAA03/README.md

Hi there 👋

I am Abdulrahman Alharbi , An Information System(Data Science track) graduate


i do enjoy low level language's power and control,and High level languages ease of use as such you might see my c++ code written in c style c++. Currently i am working on several projects :

  • Vectra : a Regex replacement project,goal is to be Readable, Fast, and Lightweight. Currently written in C.
  • Bridge : a CPU Rasterization single header, it mimicks the normal drawing pipe line via lambdas and per draw function context object.
  • Fast-Math(Atan2 only for now) : using optimization techniques such as Branchless Programming , Single Instruction Multiple Data(SIMD) and Instruction Level Parallelism(ILP), we achieved 0.23 nanosecond per atan2 Pair in SIMD 8 floats while still maintaining IEEE 754 compliance in the accurate version (which has max error of 2.57e-05 radians and average of 1.06e-05 radians and it still zoom into zero at octants edges).
  • Vitriol Rephrase : Goal is to help online platforms to hide toxic speech and reinterpret useful Vitriol(toxic) criticism into neutral clear criticism.

now a Question one might ask , "Why? why all of those? you should make a single project one at a time."

  • answer is : to avoid burnout ,by having diverse projects i don't tire as fast and the joy of programming still persist ,and by having those different projects, i look at problems from more than one perspective.
  • Plans and Logic(private documentation) are pre-written , only application(language specific implementations) remains which is the thing that takes time to do right.

Pinned Loading

  1. IAA03-fast-atan2 IAA03-fast-atan2 Public

    IAA03_fast_math is a single-header math kernel(Atan2 only for now) designed to eliminate the "Trigonometry Tax" in high-throughput systems (Physics Engines, Audio DSP, and ML Pre-processing). Branc…

    C++ 2