FrameTool is a macOS application for analyzing whether a video contains dropped (duplicated) frames. It’s primarily intended for evaluating game performance via recorded output but can be applied to other scenarios as well (i. e. Dropped frames in recordings from camera).
|
|
FrameTool is designed to identify dropped or duplicated frames in a video file. This is often used to analyze gaming performance based on captured footage.
The app calculates:
- Average FPS
- Minimum FPS
- Maximum FPS
- Exports csv Table with every frame's display time
- Has options for additional exports (like Graphs or Statistics)
These calculations are based solely on the frame time that was captured, and do not reflect internal game engine behavior. As such, FrameTool's readings do not directly compare to tools like MSI Afterburner or Fraps.
Installation of the app:
- Open DMG
- Drag the app to the application folder
Using FrameTool is simple:
- Launch the app
- Set up the output location (defaults to user/Documents)
- Set up additional exports you would like (CSV statistics or Graph Output) in the settings menu
- Drop a video file to a drag and drop section
- Press "Run Analysis"
- Wait for the analysis to complete
- Once done, you can drag and drop another video to analyze
A test sample video is provided within the .dmg installer to demonstrate how the tool works.
⚠️ Note: Analysis can take a significant amount of time, especially for videos longer than 2 minutes or on lower-end machines. Outputing "Animated Overlay" significantly increases the analysis time even further.
⚠️ Important: The analyzed video must be free of screen tearing. Any tearing will cause the app to falsely detect new frames. Poor bitrate of the video might also result in falsely detecting new frames, it is not recommended to process the videos recorded on Intel Arc by its own codec (AMD and Nvidia is fine as long as its reasonable bitrate). Additionally, the app does not report FPS values higher than the framerate the video was recorded at.
⚠️ Note: Tearing Detection algorithm is available in v0.2.X, however, it is experimental and known to produce both false negative and positive results. You can use this mode for entertainment, but it is in the early development stage. To acccess this mode hold ⌥ (Option Button) in the settings menu.
- Tested on Apple Silicon Macs only, however the v0.1.0 and newer may run on Intel-based Macs
- 8GB RAM recommended for Single threading processing
- 16GB RAM recommended for Multithreading accelleration for 4K videos under 1 minute long
- 24GB RAM recommended for Multithreading accelleration for 4K videos over 1 minute long
- macOS Sequoia 15.1 or newer for v0.1.0 and newer
- Video frametime analysis with min, max and avg FPS output
- Generation of csv files with times of every frame
- Detailed statistics csv output
- Graphical visualization of frame time and fps, that can either be rendered as image, interactive HTML graph or Animated overlay on top of the provided video source
- Processing time tracker
- Two custom design themes with vocaloids (Hatsune Miku and Megurine Luka)
- Multiple bugfixes
- Addition of customizable graph output designs
- Animated Overlay render optimization
- Better statistics output options
- Ability to run multiple graph outputs for a single analysis
- Windows port with different UI framework
- More customization and settings with a better layout
The .dmg installer includes:
FrameTool.app- A test video (
SampleForTesting.mp4) - An alias to the Applications folder for easy installation
This project is licensed under a custom license:
- You may use, modify, and redistribute the software (with attribution)
- You may not sell this software or any modified version
- Commercial use as part of a larger product is permitted (as long as FrameTool itself is not sold directly)
See LICENSE.md for full details.
- Built with assistance from ChatGPT (OpenAI)
- Packaged using xCode, UI is designed using SwiftUI
- Developed in Swift, using:
AVFoundationCoreImageAccelerateAppKitCoreText
- Developed and tested in xCode on macOS
- Uses arts from piapro by nibeさん and なっつみかんさん
- Uses music from piapro by 厚寝巻 and Tsubaki_Kun