Why Is My Video File So Big? Bitrate, ProRes and File Size Explained

The same 10-second video can be a 5 MB file or a 500 MB file. Same resolution, same frame rate, same duration - a hundredfold difference. This is not a glitch; it is the single most misunderstood property of digital video, and it explains almost every "why is my file so big?" question.

This guide walks through the one formula behind video file size, why professional cameras produce enormous files, and how to shrink videos without wrecking them.

The one formula: bitrate × duration

Strip away everything else, and a video file is essentially:

file size (bits) = bitrate × duration
file size (MB) ≈ bitrate (Mbps) × seconds ÷ 8

Some quick examples at 10 seconds of video:

Notice what is not in the formula: resolution, frame rate, codec. Those matter only because they influence the bitrate a camera or encoder chooses. A 4K phone clip and a 1080p ProRes clip can be the same file size; the ProRes one is smaller in pixels but 30x the bitrate.

So the real question becomes: what decides the bitrate?

What actually decides bitrate

1. The codec generation (the biggest factor)

A codec is a compression algorithm. Each generation learns to describe the same picture with fewer bits:

• H.264 (AVC) - the universal standard, ~20 years old
• H.265 (HEVC) - same quality at roughly half the H.264 bitrate
• AV1 - roughly half again, open and royalty-free, but slow to encode
• VP9 - Google's open codec, between H.264 and H.265 in efficiency

If your camera shoots 4K H.265 and your friend's shoots 4K H.264, your file can be half the size at identical quality.

2. Codec purpose: shooting vs delivery (the 100x factor)

This is the part that surprises everyone. Codecs are built for different jobs:

Delivery codecs (H.264, H.265, AV1) are designed to be watched. They use inter-frame compression: most frames are stored as differences from neighboring frames, throwing away everything the human eye won't miss. A talking-head shot compresses incredibly well because frame-to-frame almost nothing changes.

What exactly is ProRes? ProRes is a family of professional mezzanine codecs developed by Apple. One-sentence definition: it is the "working format" of the editing and color-grading stage - professional cameras record straight into it, editing software edits it natively, and its purpose is not to be watched but to give post-production maximum headroom.

That headroom shows up in three concrete ways:

1. Intra-frame compression: every frame is stored complete and independent, referencing no other frames. Scrub to any point on the timeline and decoding is instant - while an inter-frame codec (H.264) must first decode a whole chain of reference frames. This is why ProRes feels buttery smooth in an editing timeline.
2. High bit depth: 10-12 bits per channel (billions of color steps, vs ~16.7 million for the 8-bit H.264 commonly uses). When a colorist lifts the shadows by 3 stops, 10-bit gradients stay smooth while 8-bit ones show visible rings of banding.
3. Full chroma (4:4:4): color is stored as precisely as brightness, so green-screen keys get clean edges.

The cost is size: keeping all that headroom means the bitrate must be 50-100x that of a delivery codec. Avid's DNxHD/HR does the same job in the same family (ProRes is to Final Cut what DNxHD is to Avid).

A useful analogy: H.264 is a printed photo; ProRes is the negative. You don't frame the negative and hang it on the wall (huge, awkward to look at), but retouching should always start from the negative. Converting ProRes to H.264 is developing the negative into a print - you lose the retouching headroom, but the file is 50-100x smaller and looks practically identical for viewing.

That headroom is why an ARRI Alexa shooting ProRes 4444 XQ at 1080p produces ~400 Mbps - about 50-100x the bitrate of the phone in your pocket filming the same scene at the same resolution. Neither file is "wrong"; they are optimized for completely different stages of the pipeline.

RAW and RAW-like formats go further ( uncompressed sensor data) and scale to terabytes per hour.

3. Resolution, frame rate, bit depth, chroma

Within one codec, size scales with:

• Resolution: 4K has 4x the pixels of 1080p - roughly 2-3x the bitrate after compression
• Frame rate: 60 fps carries twice the frames of 30 fps - roughly 1.5-2x the bitrate
• Bit depth: 10-bit vs 8-bit adds ~25% (mainly matters in grading)
• Chroma subsampling: 4:4:4 (full color) vs 4:2:0 (half color both dimensions) is up to 1.5-2x

This is why ProRes 4444 XQ (12-bit, 4:4:4) is nearly 4x the bitrate of Pro 422 Proxy at the same resolution.

4. Content complexity

Grain, confetti, water, camera movement and fire are bitrate hogs. A static screen recording compresses to a fraction of a handheld street shot, same codec, same settings.

Why your iPhone video is 400 MB (and your friend's isn't)

Modern phones expose the shooting codec in settings:

• "High Efficiency" = H.265/HEVC → about half the size
• "Most Compatible" = H.264 → about double the size

A minute of 4K30 video lands anywhere from ~200 MB (H.264) to ~100 MB (H.265) to ~90 MB (10-bit HDR off). If someone hands you a 2 GB "short clip," check whether it came out of Final Cut or a cinema camera - it is almost certainly ProRes, and the size is expected.

How to shrink a video without wrecking it

The practical recipe, in order:

1. Convert editing codecs to delivery codecs. A ProRes/DNxHD MOV converted to H.264 MP4 at CRF 20-23 routinely drops 500 MB to 10-30 MB, and the difference is invisible at normal viewing sizes. Try our MOV to MP4 converter - it runs entirely in your browser.
2. Right-size the resolution. Nobody needs 4K for a clip that will only ever be watched in a chat window. 1080p cuts the bitrate roughly in half.
3. Use a modern codec when the target supports it. H.265 halves size vs H.264; AV1 halves it again. Check compatibility of the devices you care about first.
4. Don't re-encode twice. Every lossy generation loses a bit more. Go back to the original (or a mezzanine file) when you need a different output, instead of converting the converted file.

One warning sign to ignore: after converting ProRes to H.264, the output being 20x smaller is normal, not a quality collapse - see the FAQ above.

Containers don't decide size (mostly)

MP4, MOV, MKV and WebM are containers - boxes that hold encoded streams. A MOV with H.264 inside is nearly identical in size to the same stream in an MP4. The codec inside the box is what matters. The full container-vs-codec breakdown is in our Audio & Video Format Guide.

This also explains why MOV → MP4 is sometimes instant (container rewrite, no re-encoding) and sometimes slow (ProRes inside → full re-encode to H.264).

Quick reference: expected sizes per minute

If your file is roughly in these ranges, the size is normal for its type. If you need it smaller, the answer is almost always: convert to a delivery codec - start with MOV to MP4 or any of our media converters, all running locally in your browser, no uploads.