← CineCinch Home  /  Compress Video Files on Windows

How to Compress Video Files on Windows Without Losing Quality

A practical comparison of every method — HandBrake, ffmpeg, and dedicated batch tools — with the exact H.265 settings that give the best quality-to-size ratio for your library.

Why video files get bloated in the first place

Not all large video files are bloated. A 2-hour 4K movie at 8 GB is probably fine. A 10-minute screen recording at 4 GB almost certainly isn't. The difference comes down to how efficiently the video was encoded at the time it was created.

Video encoders use different codecs (the compression algorithm) and bitrates (how much data per second). When you record a screen with default settings, export from video editing software, or save a clip from a camera in RAW or ProRes format, the result is often encoded with a very high bitrate — far more than you need for long-term storage.

Common sources of oversized video files:

• Screen recordings — Windows Game Bar, OBS, ShareX, and most capture tools default to high bitrate for edit-friendliness, not storage efficiency.
• Camera exports — GoPro, DJI, Sony, and Canon cameras use high-bitrate H.264 or even MJPEG by default.
• Video editor exports — Premiere Pro, DaVinci Resolve, and similar tools often default to H.264 at very high bitrate settings.
• Old codec files — Videos encoded with MPEG-2, Xvid, or early H.264 before modern encoders existed.
• Downloads and rips — Sources that prioritized quality headroom over file size.

The key insight: these files can typically be re-encoded to a more efficient codec (H.265) at a lower bitrate with no noticeable quality difference on a normal display.

The three main compression methods on Windows

There is no single "right" tool for compressing video on Windows. The best choice depends on how many files you're dealing with, how much control you want, and how comfortable you are with technical settings.

HandBrake

HandBrake is a free, open-source video transcoder with a full GUI. It supports H.264, H.265, AV1, and many other codecs with precise control over every encoding parameter. You load a file, choose a preset, and export.

HandBrake is excellent when you're working with one or a few specific files and want to fine-tune the result. It has a modest batch queue, but it doesn't help you identify which files in a folder need compression — you have to pick them yourself.

ffmpeg

ffmpeg is the underlying engine that most video tools (including HandBrake and CineCinch) use internally. You can call it directly from the Windows command prompt or PowerShell. A basic H.265 compress command looks like this:

ffmpeg -i input.mp4 -c:v libx265 -crf 25 -preset slow -c:a copy output.mp4

This re-encodes the video track using H.265 at CRF 25 while copying the audio unchanged. It's extremely flexible but requires you to know what you're doing. Writing a batch script to process an entire folder is possible, but you still won't get any analysis of which files need compression most.

CineCinch

CineCinch is designed specifically for the "I have a folder full of videos and want to clean it up" use case. It scans your folders with FFprobe, calculates MB/sec for every file, and ranks them by inefficiency — so you can see at a glance which files are the biggest compression opportunities. From there you queue them, pick a preset, and compress in batch.

It uses the same FFmpeg engine as HandBrake and the CLI but with a focused workflow that removes the need to know encoding parameters manually.

Online video compressors

Browser-based tools like Clideo, Compress2Go, and similar services let you upload a video, compress it on their servers, and download the result. Convenient for a single small file, but impractical for any meaningful library cleanup — file size limits, slow uploads, and the privacy implications of uploading your content to a third-party server all make them a poor fit for regular use.

H.265 vs H.264: the real-world size difference

Most videos you'll encounter are encoded in H.264 (AVC) — the codec that has dominated video for the past 15 years. It's widely supported and produces good results, but its successor, H.265 (HEVC), achieves roughly the same visual quality at about half the file size.

In practical terms, re-encoding an H.264 file to H.265 at equivalent quality typically reduces file size by 30–50%. For files that were already encoded inefficiently at high bitrate, the savings are even greater — sometimes 60–80%.

The main trade-off: H.265 encoding is slower than H.264, and hardware decoding support is required on older devices. For storage on a PC, NAS, or media server however, playback compatibility is a non-issue on anything made after 2016.

What CRF means and which value to use

CRF stands for Constant Rate Factor. It controls the quality-to-file-size trade-off for H.265 encoding. Unlike fixed bitrate, CRF lets the encoder use more data for complex scenes (action, detail) and less for simple ones (static shots, dark scenes), producing a consistently good result at a target quality level.

For H.265, the CRF scale runs from 0 (lossless, enormous files) to 51 (heavily compressed, terrible quality). In practice, the useful range for most content is 20–34:

CRF 25 is a reliable starting point for most libraries. It produces visually excellent results on 1080p content and compresses significantly. If you're working with 4K content, lean toward CRF 22–24. For casual recordings you don't care about long-term, CRF 28–30 is perfectly fine.

How to pick which files to compress first

The naive approach is to sort your video folder by file size and compress the largest files first. This is better than nothing, but it misses an important nuance: a large file is only worth compressing if it's actually inefficiently encoded.

A 2-hour 4K Blu-ray rip at 8 GB might already be well-encoded — compressing it further risks visible quality loss for minimal gain. Meanwhile, a 12-minute screen recording at 3 GB is clearly inefficient and will compress by 80%+ with no perceptible quality change.

The better metric is MB per second: file size in megabytes divided by duration in seconds. This normalizes for length and gives you a direct measure of bitrate density — how many megabytes each second of video is consuming.

General MB/sec benchmarks for well-encoded H.265 content:

• 720p: 0.05–0.16 MB/sec — anything above 0.75 MB/sec is a strong compress candidate
• 1080p: 0.10–0.32 MB/sec — anything above 1.5 MB/sec is worth compressing
• 4K: 0.25–0.75 MB/sec — anything above 3.5 MB/sec is a compress candidate

CineCinch calculates this automatically for every file and highlights Bloated (red), Average (orange), and Efficient (green) files — so you can filter to only Bloated files and work through those first.

Step-by-step workflow with CineCinch

1. 01
   Download and install from the Microsoft Store. CineCinch is available with a free 7-day trial. No configuration needed — FFmpeg and FFprobe are bundled with the app.
2. 02
   Select your folder. Click the folder icon and navigate to the root of your video library. CineCinch will scan all subfolders recursively — you don't need to select individual files.
3. 03
   Choose the right threshold for your library's resolution. Select Strict for 720p libraries, Medium for 1080p, or Generous for 4K. This sets the MB/sec breakpoints used to classify files as Efficient, Average, or Bloated.
4. 04
   Run the scan. Click Scan Videos. Results populate the table in real time, sorted highest MB/sec first. The scan uses FFprobe — no files are modified.
5. 05
   Filter to Bloated files only. Uncheck the Average and Efficient Show filters. You're now looking at only your worst offenders. Uncheck any files you want to skip.
6. 06
   Set your output preferences. Choose "Same folder as source" or a custom output folder. Select an output mode — Standard (CRF 25) is the recommended starting point for most libraries.
7. 07
   Start compression. Click Start Compression. Per-file progress shows in the table. The status bar shows live time elapsed, files processed, and total MB saved so far.
8. 08
   Verify output and clean up. Check the Saved (MB) column to confirm space savings. Spot-check a few outputs visually before enabling Delete Originals for future runs.

What file size savings to realistically expect

Results vary significantly based on what the source files are and how they were originally encoded. Here are realistic benchmarks:

Common mistakes to avoid

• Compressing already-efficient files. Sort by MB/sec first. Files rated Efficient by CineCinch's analysis are probably not worth re-encoding.
• Using too aggressive a CRF value. CRF 28+ on complex content (sports, nature documentaries, action films) can produce visible artifacts. Test on a sample first.
• Deleting originals before verifying output. Always spot-check compressed files before enabling Delete Originals. Pay particular attention to audio sync and subtitle tracks.
• Using the wrong threshold for your resolution. Using Strict on a 4K library will flag everything as Bloated. Use Generous for 4K, Medium for 1080p, Strict for 720p.
• Compressing files you'll need to re-edit. If you might return to a file in a video editor, keep the original. Compressed H.265 is not ideal for editing workflows.
• Expecting 4K files to compress as much as screen recordings. A well-encoded 4K movie may only shrink 15–20%. That's still worth doing at scale, but don't expect 80% savings on every file.

Frequently asked questions