I recently finished editing a massive four-gigabyte recording of a bass cover that ran for exactly three minutes and thirty-one seconds. I used CapCut to add the visual effects, transitions, and color grading I wanted. Unfortunately, the finished export somehow contained five seconds of dead, black footage right in the middle of the performance.
I could have imported the finished file into CapCut, created another timeline, and cut the black section there. However, that would have meant re-encoding the entire 4GB video and waiting another twenty minutes to fix five seconds of footage. Instead, I opened LosslessCut, one of the better video-cutting apps available for Windows. For compatible files, it copies the original media streams without re-encoding the footage. It is not really an editor so much as a pair of scissors.
Installation is simpler than the interface suggests
It looks busier than it actually is
LosslessCut is free and open-source, with official downloads available through its GitHub repository and website. On Windows, I downloaded the 7z archive, extracted it, and launched the app without going through an installer. macOS users can choose the free DMG, while Linux builds are available as AppImage and tar archives.
The developer also offers paid versions on the Microsoft Store and Mac App Store. Paying for free software may seem unnecessary, but the store editions provide automatic stable updates, avoid some of the security warnings that can accompany direct downloads, and offer a straightforward way to support the developer. The core features are largely the same, although store releases may lag the latest GitHub build and may include a few platform-specific restrictions. The free GitHub release worked perfectly on my Windows PC.
When you open LosslessCut, the interface can initially look intimidating. It includes a large media preview, playback controls, a timeline, a segment list, track panels, and an output-format selector. The layout looks busy because LosslessCut exposes many of the stream, track, container, and metadata operations performed by FFmpeg, a tool capable of handling many jobs people otherwise pay for, which works underneath it.
The good news is that the basic editing workflow relies only on a small number of controls. Once you know where they are, the busy interface becomes far less daunting.
Mark the good parts, drop the bad, and hit export
I kept the bass, but lost the accidental blackout
LosslessCut shines when you need to mark the useful parts and export them. You drag a video into the window or open it through the File menu, and it functions much like a standard video player.
LosslessCut does not modify the original file. It creates a new output based on your highlighted sections and saves it to the source folder unless you change the working directory.
To tackle the dead space in the middle of the performance, I used the multi-segment workflow. I scrubbed to the beginning of the video and pressed the I key to set the start of my first segment. I played the video and pressed O to set its end immediately before the screen went black. I then pressed the + button to add another segment, skipped over the dead footage, and placed its start point as soon as the picture returned. Finally, I set the endpoint at the three-minute and thirty-one-second mark.
You can also mark the footage you want to remove instead of selecting everything you want to keep. The yin-yang-shaped invert control, available through the advanced controls, reverses the selection so LosslessCut skips the marked segment and exports the footage on either side. This can be quicker when the unwanted section is much shorter than the rest of the video.
With my two useful segments marked, I opened the export options and changed the mode to Merge cuts. This instructed the software to stitch the two segments together rather than export them as separate files. I clicked Export+merge, and the process finished far sooner than reopening the video in CapCut and rendering the entire project again would have.
The resulting file was only slightly smaller, which made sense because LosslessCut had removed a few seconds rather than reduced the video’s file size through compression. Its resolution, effects, transitions, color treatment, and audio remained as they had appeared in the CapCut export.
More importantly, the black interruption was gone. The two sides joined cleanly, and the bass remained synchronized with the picture when I played the result in VLC.
LosslessCut manages this by stream copying. Instead of decoding and recompressing every frame, it takes the already-compressed video and audio packets it needs, places them into a new container, and leaves the unwanted material behind. It may rewrite timestamps, container information, and some metadata, so the output is not a bit-for-bit duplicate. The retained media avoids another lossy encoding pass.
The shortcut still has a keyframe-shaped catch
Your perfect cut may have other plans
LosslessCut’s speed comes with an important trade-off rooted in how compressed video works. Most formats do not store every frame as a complete image. Instead, they use keyframes containing a full picture, followed by frames that record only what changed. Because LosslessCut copies these existing streams rather than rebuilding them, the start of each retained segment typically needs to fall on an existing keyframe.
As a result, LosslessCut may slightly shift the start of a segment to the previous keyframe, leaving a small amount of unwanted footage at the edge. How noticeable this becomes depends on how far apart the keyframes were placed when the source video was encoded. Moving through the preview one frame at a time may look precise, but it does not guarantee a frame-perfect lossless result. The timeline’s keyframe display and jump controls make it easier to choose boundaries that behave predictably.
When a cut genuinely needs frame-level accuracy, LosslessCut offers an experimental Smart Cut feature. It attempts to create a precise boundary by re-encoding only the necessary region around the cut while stream-copying the remainder. This means the boundary itself is no longer produced entirely through lossless stream copying, and the feature can cause glitches, synchronization issues, or visible differences with some formats. I treat Smart Cut as a fallback rather than the default, since standard Keyframe Cut mode remains faster and more reliable.
Other practical limitations come with stream copying. Clips with different video codecs and containers, resolutions, frame rates, or other encoding parameters may not merge cleanly. Some files also contain audio, subtitle, or data tracks that LosslessCut cannot trim correctly, which can result in an incorrect output duration or unexpected blank sections. Disabling unnecessary tracks or choosing another compatible container, such as MP4, MOV, or MKV, can sometimes help.
However, the best option depends on the streams inside the original file. These restrictions are the reason LosslessCut is fast. Once a project requires transitions, effects, extensive rearrangement, or clips from several unrelated sources, I return to a conventional timeline editor. DaVinci Resolve can replace Premiere Pro for that kind of full-scale editing, but both are excessive when the footage merely needs a trim.
This is really the editor for avoiding actual editing
I would not use LosslessCut to build a polished video, but that is exactly why it stays useful. When all I need is to remove dead time from a bass cover, extract a single section, or combine compatible clips, it avoids the project setup and full render demanded by a larger editor such as DaVinci Resolve or Premiere Pro. Keyframe limitations occasionally require a small adjustment or another tool, but LosslessCut is the video editor I use when the footage is already fine, and I merely need less of it.