A SIMD library that provides an intuitive and readable interface to 256-bit AVX and AVX2 SIMD instructions using low-cost abstractions.

SHA-256 / Blake3 performance benchmark on different languages and algorithms

An experiment in optimizing a string counter.

Mandelbrot set SIMD optimization

Optimizing matrix multiplication using parallelism and SIMD (AVX2, CUDA)

Уравнение теплопроводности в трёхмере

AVX2 version of Luhn algorithm

AVX sorting experiment

This C++ program is a demonstration of array vectorization techniques utilized in the AVX2 SIMD Assembly library, being run with C++ arrays through the vector class library created by Agner Fog. An ASM version of the same process has been implemented for comparison.

Mpbenchmark is an application benchmark for evaluating the efficiency of multiprocessor support for parallel processing. Mpbenchmark is a revised version of JetBench written in C++20.

Tiny optimized math framework game oriented

row oriented to column oriented

A hash table optimization project - using assembly and several techniques in order to improve its performance.

SIMD accelerate xoshiro128, generate 256 bits 'number' at once

⚡️⚡️⚡️Blazing fast correlation functions on the CPU.

Simple image editor made using Windows Forms. It includes three versions of functions: C#, C++ and ASM and allow to compare their processing times.

AVX Accelerated BigUInt Arithmetic Implements

Battleships opponent and compute experiments, with AVX2 / AVX-512

N-body AVX2/FMA3 test