How to Use This Tool

Upload your video file or drag and drop into the browser

Supports MP4, MOV, AVI, WebM, MKV • Maximum 2GB file size

Preview video and identify the segment you want to keep

Use frame-by-frame navigation for precise selection

Set start and end points using the timeline slider or time inputs

Timestamp format: HH:MM:SS.mmm (hours:minutes:seconds.milliseconds)

Choose processing mode: Fast (stream copy) or Quality (re-encode)

Fast = instant • Quality = slow but precise frame cuts

Click "Trim Video" and download your edited file

Processing time: 5-30 seconds depending on duration and mode

How Browser-Based Video Trimming Works

Browser video trimming leverages WebAssembly-compiled FFmpeg running entirely client-side. The process involves three distinct technical operations:

1. Stream Copy (Fast Mode) - No Re-encoding

Fast trimming uses stream copying (codec copying) which extracts video/audio packets without decoding and re-encoding:

FFmpeg Command Equivalent:

ffmpeg -ss 00:01:30 -i input.mp4 -to 00:03:45 -c copy output.mp4

Process breakdown:

• -ss 00:01:30: Seek to start timestamp (90 seconds)
• -to 00:03:45: Stop at end timestamp (225 seconds)
• -c copy: Copy codec streams without re-encoding (preserves quality)

Limitation: Can only cut at keyframes (I-frames), not arbitrary timestamps. If requesting 1:30.500 but nearest keyframe is at 1:30.800, the output starts at 1:30.800.

⚠️ Keyframe Interval Impact

Most videos have keyframes every 2-10 seconds. Cinema/professional footage may have 1-second intervals (more precise cuts). Screen recordings often have 30-second intervals (less precise cuts).

2. Re-encoding (Quality Mode) - Frame-Accurate Trimming

Precise trimming requires full decode-encode pipeline to cut at exact frames:

FFmpeg Re-encode Command:

ffmpeg -i input.mp4 -ss 00:01:30.500 -to 00:03:45.250 -c:v libx264 -preset fast -crf 23 -c:a aac output.mp4

Encoding Pipeline Steps:

1. Demux: Separate video/audio streams from container format (MP4, MOV)
2. Decode: Decompress H.264/H.265 video to raw YUV frames and AAC audio to PCM
3. Trim: Discard frames outside specified timestamp range
4. Encode: Recompress kept frames using H.264 codec (libx264) with specified quality
5. Mux: Package encoded streams back into MP4 container with corrected timestamps

Quality Control Parameters

→CRF (Constant Rate Factor): 0-51 scale where 0 = lossless, 23 = default balanced quality, 51 = lowest quality. Each +6 CRF halves file size.
→Preset: Speed/compression tradeoff. "ultrafast" encodes 10x faster but 30% larger files vs. "slow" preset.
→Bitrate mode: VBR (variable bitrate) adjusts quality per scene. CBR (constant) maintains fixed bitrate for streaming.

3. Timestamp Recalculation and Container Repair

Video containers store Presentation Timestamps (PTS) and Decode Timestamps (DTS) for every frame. Trimming requires timestamp adjustment:

Original file timestamps (90-second trim point):

Frame 2698: PTS = 89.933s, DTS = 89.900s

Frame 2699: PTS = 89.967s, DTS = 89.933s

Frame 2700 (keyframe): PTS = 90.000s, DTS = 89.967s ← CUT HERE

Frame 2701: PTS = 90.033s, DTS = 90.000s

Trimmed file (timestamps reset to zero):

Frame 0 (was 2700): PTS = 0.000s, DTS = 0.000s

Frame 1 (was 2701): PTS = 0.033s, DTS = 0.033s

Frame 2 (was 2702): PTS = 0.067s, DTS = 0.067s

Improper timestamp handling causes A/V desync (audio drift), seeking failures, or incorrect duration metadata. Browser-based FFmpeg automatically performs PTS/DTS normalization to prevent these issues.

Browser Performance Constraints

WebAssembly FFmpeg runs 3-5x slower than native desktop FFmpeg due to JavaScript overhead and single-threaded execution. Realistic processing speeds:

• Stream copy mode: ~200-500 MB/s (near-instant for most files)
• Re-encode 1080p H.264: ~15-30 FPS (2-4x realtime)
• Re-encode 4K H.264: ~5-10 FPS (6-12x realtime)
• Re-encode HEVC/H.265: ~3-8 FPS (15-20x realtime)

Example: 10-minute 1080p video trimmed with re-encoding takes ~2-5 minutes to process in browser.

Fast Mode vs. Quality Mode Comparison

Attribute	Fast Mode (Stream Copy)	Quality Mode (Re-encode)
Processing Speed	5-10 seconds for 1GB file	2-5 minutes for 1GB file
Cut Precision	±2-10 seconds (keyframe-limited)	±0.033 seconds (1 frame at 30fps)
Quality Loss	Zero (lossless copy)	Minimal (lossy re-compression)
File Size Change	Proportional to trimmed duration	May increase or decrease (depends on CRF)
CPU Usage	5-15% (file I/O only)	90-100% (full encode)
Battery Impact (Mobile)	Negligible	High (rapid drain)
Best Use Case	Rough cuts, content removal, quick edits	Precise timing, professional edits, social media

When Fast Mode Produces Bad Results

Stream copy fails at exact cuts when keyframe intervals are large. Common scenarios:

→Screen recordings: OBS/Camtasia often use 10-30 second keyframe intervals for smaller files
→Action cameras (GoPro): High-motion footage benefits from sparse keyframes, creating 5-8 second cut granularity
→Streaming video downloads: Adaptive bitrate segments may have irregular keyframe placement

Solution: Always use Quality mode when trim precision matters (within 1 second accuracy needed).

Supported Formats and Codecs

Video Codecs

H.264 (AVC)	✓ Full support
H.265 (HEVC)	✓ Full support
VP8/VP9	✓ Full support
AV1	⚠ Slow decode
MPEG-4	✓ Full support
ProRes	⚠ Large files

Container Formats

.mp4	✓ Recommended
.mov	✓ Full support
.webm	✓ Full support
.mkv	✓ Full support
.avi	⚠ Legacy format
.flv	⚠ Limited

Audio Track Handling

Trimming preserves all audio tracks from source videos:

Multiple audio tracks: Stereo + 5.1 surround both copied to output
Audio sync: PTS alignment ensures A/V sync maintained across trim points
Codec support: AAC, MP3, Opus, Vorbis, AC3, DTS all supported
Sample rate: 44.1kHz, 48kHz, 96kHz preserved without resampling (Fast mode)

Re-encode mode: Audio transcoded to AAC 128kbps stereo by default. Original multichannel layouts simplified to stereo for browser compatibility.

Advanced Trimming Techniques

Pro Tip: Two-Pass Trim for Maximum Quality Retention

When trimming already-compressed video (YouTube downloads, social media exports), avoid quality degradation:

Standard approach (quality loss):

Original (H.264 CRF 23) → Re-encode (H.264 CRF 23) = Generational loss from double compression

Better approach (minimal loss):

1. Use Fast mode (stream copy) to get rough cut with ±5 second precision
2. Only use Quality mode if Fast mode's keyframe alignment is unacceptable
3. When re-encoding is necessary, use CRF 18-20 (higher quality) to offset generational loss

Rule: Each re-encode cycle reduces quality ~10-15%. Stream copy preserves 100% original quality.

Frame-Accurate Trimming Using Timecode Display

For professional edits requiring exact frame cuts (music video beat matching, dialogue cuts):

Calculate exact frame number:

Frame number = (timestamp in seconds) × (frame rate)
Example: Cut at 2:15.500 in 30fps video = (135.5 seconds) × 30 = Frame 4065

Use keyboard shortcuts: Arrow keys for frame-by-frame navigation, Shift+Arrow for 10-frame jumps. Mark in/out points at exact frames visible in preview.

Optimize File Size Without Visible Quality Loss

When re-encoding is required, balance quality vs. file size using CRF values:

CRF Value	Quality Level	Use Case
18-20	Visually lossless	Archival, professional work
23	High (default)	General purpose, recommended
28	Medium	Social media, fast sharing
32+	Low (visible artifacts)	Extreme compression only

Handling Variable Frame Rate (VFR) Video

Screen recordings and game captures often use VFR (frame rate varies dynamically). This causes issues:

•Symptom: Audio drift after trimming, seeking jumps incorrectly
•Cause: Timestamp assumptions break when frame rate isn't constant
•Fix: Convert to CFR (constant frame rate) before trimming using re-encode mode with -r flag

Detection: Use MediaInfo or ffprobe to check if "Frame rate mode" is VFR. If yes, always use Quality mode with CFR conversion.

Batch Trim Multiple Sections from One Video

Create multiple clips from a single source video efficiently:

Workflow:

1. Load source video once
2. Define all trim ranges: (0:30-1:15), (2:45-3:30), (5:00-6:20)
3. Queue all operations for single-pass processing
4. Generate 3 output files without reloading source

Time saved: 60-70% vs. trimming each segment individually

Preview Quality vs. Output Quality

Browser video preview uses lower resolution for smooth playback. The displayed quality is NOT the output quality. Source resolution is preserved in trimmed files regardless of preview appearance. To verify: check output file properties after download — resolution, bitrate, and codec match source specifications.

Browser Performance and File Size Limits

Memory Constraints

Browser-based processing loads entire video into RAM. Practical limits by device type:

Desktop (8GB+ RAM): Up to 2GB video files (safe limit), 4GB maximum (may crash)
Mobile/Tablet (4GB RAM): Up to 500MB video files recommended
Low-end devices (2GB RAM): 100-200MB maximum to prevent browser crashes

For files exceeding browser limits, use desktop software (DaVinci Resolve, Adobe Premiere, FFmpeg CLI).

Video Resolution	Fast Mode Speed	Re-encode Speed	Recommended Max Duration
720p (HD)	2-5 seconds	30-60 seconds	60 minutes
1080p (Full HD)	3-8 seconds	1-3 minutes	30 minutes
1440p (2K)	5-12 seconds	3-6 minutes	15 minutes
4K (UHD)	8-20 seconds	8-15 minutes	5 minutes

Browser Compatibility

WebAssembly FFmpeg requires modern browser features. Minimum requirements:

Chrome/Edge: Version 57+ (March 2017 or newer)
Firefox: Version 52+ (March 2017 or newer)
Safari: Version 11+ (September 2017 or newer) — iOS 11+ for mobile
Opera: Version 44+ (March 2017 or newer)

Unsupported: Internet Explorer (all versions), Android WebView in apps (varies by implementation).

When to Trim Videos

Optimal Use Cases

✓Social media clips: Extract 15-60 second highlights from longer recordings for Instagram/TikTok
✓Meeting recordings: Remove pre-meeting chatter and post-meeting silence from Zoom/Teams recordings
✓Tutorial creation: Cut mistakes, pauses, and irrelevant sections from screen recordings
✓File size reduction: Remove intro/outro to meet platform upload limits (100MB email, 500MB Discord)
✓Privacy protection: Remove sections showing sensitive information before sharing
✓Presentation excerpts: Extract specific slides or segments from recorded presentations

Poor Use Cases (Use Full Editor)

✗Multi-section edits: Removing 5+ separate segments requires full NLE (non-linear editor)
✗Adding effects/transitions: Trimming is cut-only; use DaVinci Resolve for effects
✗Audio editing: Volume adjustment, noise reduction, music overlay not supported
✗Color grading: No color correction or grading capabilities
✗Text/graphics overlay: Cannot add titles, captions, or lower thirds
✗Format conversion: Trimming preserves original format; use dedicated converter for format changes

Understanding Quality Loss

Generational Loss in Re-encoding

Each re-encode cycle degrades video quality through lossy compression. Cumulative effect:

Example: YouTube video → Trim → Re-upload cycle

Generation 0 (original): 100% quality
Generation 1 (YouTube upload): ~85% quality (YouTube re-encodes all uploads)
Generation 2 (download + browser trim): ~72% quality (85% × 85%)
Generation 3 (re-upload): ~61% quality (72% × 85%)

After 3 generations: Visible artifacts (blockiness, color banding, blurring)

Prevent Quality Loss

✓Always use Fast mode when keyframe precision is acceptable
✓Keep source files (never delete originals after trimming)
✓Use highest CRF setting your file size allows (18-23 range)
✓Minimize number of re-encode cycles (plan edits carefully)

Causes of Quality Loss

✗Using Quality mode on already-compressed video
✗High CRF values (28+) for space saving
✗Re-encoding downloaded social media videos
✗Multiple edit-save-edit cycles on same file

Privacy and Security

✓ Client-Side Processing Advantages

This tool processes videos entirely in your browser using WebAssembly. No data transmission occurs:

Zero uploads: Video files never leave your device
No server storage: No copies retained on external servers
Network independence: Works offline after initial page load
Privacy compliance: GDPR/CCPA compliant (no data collection possible)

⚠️ Browser Security Considerations

→Memory residue: Video data remains in browser RAM until tab closed or page refreshed
→Cache storage: Some browsers cache video data temporarily (clear cache for sensitive content)
→Download folder: Trimmed videos save to default downloads location (move to secure location if needed)

For maximum security with sensitive videos: Use private/incognito browsing mode, clear cache after processing, delete downloads immediately after transfer.

Processing Limitations Disclaimer

This tool uses WebAssembly-compiled FFmpeg running client-side in your browser. Processing speed is 3-5x slower than native desktop applications due to JavaScript execution overhead and single-threaded WebAssembly constraints. Large files (2GB) may cause browser crashes due to memory exhaustion.

Fast mode (stream copy) accuracy is limited by keyframe intervals in source video — typical precision is ±2-10 seconds. Quality mode provides frame-accurate cuts but requires full re-encoding, causing generational quality loss on already-compressed videos.

For professional workflows requiring multi-track editing, color grading, audio mixing, or batch processing, use desktop software (DaVinci Resolve, Adobe Premiere Pro, Final Cut Pro). For automated server-side trimming at scale, use FFmpeg CLI or cloud video APIs.

Authoritative Sources

1. FFmpeg Project. "FFmpeg Documentation - Stream Copy." FFmpeg.org, 2024. https://ffmpeg.org/ffmpeg.html#Stream-copy
2. International Telecommunication Union. "ITU-T H.264: Advanced video coding for generic audiovisual services." ITU-T Recommendation, 2021.

Video Trimmer

About Video Trimmer