Research section Research projects and references

Research: CORE22-BENCHMARKING Folding Project #17104

Project #17104 overview

Project Summary AI Beta

This project tests a new computer program called core22 on different hardware. It uses small tasks to find problems and help make core22 work better on all types of computers.
Automated summary; simplified and may not be fully accurate.

Project team

Manager(s)
Dr. John Chodera
Institution
Memorial Sloan Kettering Cancer Center

Work unit

Atoms
50,000
Core
OPENMM_22
Status
Public
Source material

Official Project Description

This project contains a variety of benchmarking systems (with adjusted numbers of steps per system to give equal time/WU on an RTX 2080) to test and benchmark core22.

These WUs are very short and just used to detect issues with various hardware to ensure we can continue to improve core22 on all hardware.

Performance data

Hardware Performance for Project 17104

Compare community-sampled Folding@Home output for the GPUs and CPUs processing this project.

Data as of Sunday, 02 August 2026 22:00:14

GPU PPD Averages

Rank
Project
Model Name
Folding@Home Identifier
Make
Brand
GPU
Model
PPD
Average
Points WU
Average
WUs Day
Average
WU Time
Average
1 Iris(R) Graphics 540
SKL GT3 [Iris(R) Graphics 540]
Intel SKL GT3 23,812 3,204 7.43 3 hrs 14 mins
2 HD Graphics 630
KBL GT2 [HD Graphics 630]
Intel KBL GT2 4,314 1,109 3.89 6 hrs 10 mins