Setup & Use

Video Duplicate Finder – Wiki (Setup & Use)

Overview

This wiki is intended for new users and explains how to set up Video Duplicate Finder, perform the first scan, and adjust typical settings appropriately.

---

Setup

1) Install FFmpeg & FFprobe

Video Duplicate Finder requires ffmpeg and ffprobe. If they are missing, the program will attempt to download them before the first scan and, depending on the OS, issue instructions on what to do. If this fails, refer to the project's readme for instructions on how to install them manually.

2) Choose a database format (first launch only)

On first launch, you will be asked to choose a graybyte database format:

• 16×16 – Fastest, but more false positives. The pHash algorithm is not available with this format.
• 32×32 (recommended) – Slightly slower on first scan, fewer false positives, and pHash is available.

This choice is permanent. You cannot switch formats later without deleting the entire database and starting over.

3) Start the program

• Windows: Start VDF GUI.
• Linux/macOS: Change to the VDF directory and execute ./VDF.GUI (or dotnet VDF.GUI.dll).

---

Important performance notes

The first run is the slowest

The first scan is often the slowest because it involves a lot of I/O (reading files). Subsequent scans are faster because most information is already cached in the database.

Native FFmpeg Bindings

The native FFmpeg binding brings enormous performance improvements, but requires a specific FFmpeg version. You must use the latest shared and stable FFmpeg release. Using a development/master build will cause VDF to crash.

---

Understanding settings & using them effectively

"Max Parallel"

This setting controls how many files are processed simultaneously (default: 1).

• Local SSD/NVMe: Higher values may help (e.g., 4–8) because the disk can handle many parallel read accesses.
• Network paths (SMB/NAS): Too much parallelism can be slower because the network becomes a bottleneck. A low fixed value is often better than a high one here.

Too many false positives (incorrect duplicates)

• Increase thumbnail count: More thumbnails improve accuracy because more frames are used in the comparison.
• Raise the similarity threshold: A higher percentage means files must look more alike to be flagged as duplicates.

Duplicates are not found

• Slightly lower the similarity threshold: A lower percentage increases sensitivity and finds more similar files, at the cost of potentially more false positives.
• Enable logging in the settings and check whether files are being excluded and why.

---

Tip: Try both algorithms

The two algorithms work differently and complement each other:

• Visual similarity (default) – Faster. Compares frames pixel-by-pixel. Works well for re-encoded videos, watermarks, and most common duplicate cases.
• pHash – Slower, but more robust to contrast changes, resizing, and heavy re-encoding. Only available with the 32×32 database format.

Recommendation: Run a normal scan first, then run a second scan with pHash enabled to catch duplicates the first pass may have missed.