Skip to content
This repository was archived by the owner on Nov 16, 2025. It is now read-only.

Releases: error-wtf/CALCULATION_OF_NUMBER_PI

CALCULATION_OF_NUMBER_PI-CUDA-Version1.0

30 Oct 21:04
fb38f87

Choose a tag to compare

NVIDIA GRAPHICS CARD NEEDED AND CUDA TOOLS INSTALLED !!!

Linux:

sudo apt update
sudo apt install -y build-essential cmake libboost-dev nvidia-cuda-toolkit
chmod a+x pi-error-wtf-v1.0-Linux.deb
dpkg -i pi-error-wtf-v1.0-Linux.deb

Then execute it with:

pi-error-wtf

Example:

--------------------------------
pi-error-wtf - A Pi Computation Suite
Copyright © 2025 Carmen Wrede und Lino Casu
--------------------------------
System Resource Overview
--------------------------------
Free RAM: 33.86 GB
Free Disk Space: 471.93 GB
Total Swap Space: 0.00 GB
Free GPU VRAM: 13382.00 MB
--------------------------------
Maximum recommended digits based on resources: 2545176453
Note: For very large calculations, RAM (70% free) and Disk (90% free minus swap) are the primary limits.

Enter the output file path (default: pi.txt):
Enter the number of digits to compute (2-2545176453): 1000001

Using algorithm: chudnovsky-gpu
Computing 1000001 digits...
Output will be saved to: pi.txt

Wrote π with 1000001 digits to pi.txt using chudnovsky-gpu algorithm.

Windows

Just open the EXE-File

minimum PC-Requirements:

Testet on:

64gb ram
NVIDIA Geforce RTX 4060 TI
(YOU WILL NEED A NVIDIA CARD AND CUDA TOOLS INSTALLED)
i9 cpu

License

ANTI-CAPITALIST SOFTWARE LICENSE (v 1.4) 2024 © Lino Casu