H.265 vs MP4: When Do You Need H.264?

H.265 (HEVC) is a video codec; MP4 is a container that can hold HEVC or H.264 video. An MP4 extension alone does not tell you whether a device can play the file. Remux only when the destination supports the existing codec; convert to H.264 when HEVC decoding is the obstacle.

Already know that your destination needs H.264? Use the H.265 to H.264 converter for a compatibility copy. Keep the original until you have checked the result.

How can I tell whether an MP4 contains HEVC or H.264?

Inspect the video stream, not the filename. A media-information tool may label HEVC as H.265; in ffprobe, codec_name=hevc identifies HEVC and codec_name=h264 identifies AVC/H.264. MP4 sample-entry tags such as hvc1 and hev1 also indicate HEVC.

ffprobe -v error -select_streams v:0 -show_entries stream=codec_name,codec_tag_string,profile,pix_fmt,width,height,avg_frame_rate -of default=noprint_wrappers=1 'input.mp4'

This reads the first video stream without modifying the file. These examples use Bash-style single quotes. Replace the filename inside the quotes and keep the quotes, for example 'holiday video.mp4'. If a filename contains a single quote, use a copy with a simpler name for these examples. See the ffprobe documentation for stream selection and output options.

When is remuxing enough instead of converting H.265 to H.264?

Remuxing is enough when the destination accepts HEVC and the selected audio tracks, but requires a different container. It copies compressed streams without re-encoding. It cannot turn HEVC into H.264 or fix a missing HEVC decoder. Renaming .mkv to .mp4 does neither.

For an HEVC MKV with MP4-compatible audio, this example copies the first video and optional first audio track into MP4:

ffmpeg -i 'input.mkv' -map 0:v:0 -map '0:a:0?' -c copy 'output.mp4'

The result still contains HEVC. This command omits subtitles and additional tracks; check those separately if needed. An incompatible stream may prevent MP4 muxing. See FFmpeg’s streamcopy documentation for the distinction between copying and transcoding.

Why can an MP4 play on one device but not another?

Container support does not guarantee codec support. Check the target device or application’s specifications for HEVC profile, bit depth, resolution, and frame-rate support. Hardware decoding depends on the device, operating system, and playback software; software decoding may work but struggle to play smoothly. Test on the actual destination before replacing your original.

For a broadly compatible H.264 copy of ordinary SDR footage, 8-bit 4:2:0 (yuv420p) is a common choice. A 10-bit HDR source needs additional care: changing pixel format alone is not HDR-to-SDR tone mapping. Preserve the HDR original and use a workflow that explicitly handles its color information. The MDN codec guide explains codec profiles and compatibility constraints.

Does converting HEVC to H.264 reduce quality or file size?

A typical lossy H.264 conversion introduces another generation of loss. It cannot restore detail missing from the HEVC source. The output can be larger or smaller depending on quality settings and scene content; matching the source bitrate does not guarantee matching visual quality. Our measured example below shows both outcomes.

Our reproducible H.264 settings test

We tested FFmpeg 5.1.9 on August 8, 2026 using a five-second, 1280×720, 30 fps testsrc2 pattern with a 997 Hz AAC audio track. The HEVC source used x265 at CRF 28 and measured 1,356,580 bytes. We then re-encoded only the video with x264’s medium preset, preserved the 1280×720 dimensions and 30 fps rate, and copied the AAC track.

The test shows direction, not a promised compression ratio: a camera scene, animation, grain, and motion all change the result. Reproduce it by generating testsrc2=size=1280x720:rate=30:duration=5, encoding an x265 CRF 28 source, and making x264 medium outputs at CRF 18, 23, and 28. Verify codec, dimensions, frame rate, duration, and audio with ffprobe.

Which H.264 settings affect quality and compatibility?

Codec and pixel format

Select H.264 (AVC) for the video output. For the widest playback support, use the common 8-bit 4:2:0 pixel format, usually shown as yuv420p. Higher bit depths and less common chroma formats can be valid H.264, but some hardware decoders reject them.

Quality or bitrate

A quality-based control is usually the simplest choice when exact file size is not required. Lower constant-rate-factor values in common H.264 encoders preserve more detail and produce larger files; higher values reduce size and quality. There is no single value that fits animation, screen recordings, noisy camera footage, and clean studio video equally well. The dedicated converter uses balanced defaults. In a local encoder, choose a quality setting, then inspect detailed scenes, text, gradients, and fast motion.

Use a target bitrate only when a platform or delivery channel imposes a size or bandwidth limit. Matching the H.265 source bitrate is not a reliable rule: H.264 and H.265 do not achieve the same quality at the same bitrate.

Resolution and frame rate

Keep the original resolution and frame rate unless the destination has a documented limit. Downscaling a 4K source to 1080p can make a smaller, easier-to-play file, but it is a separate decision from changing codecs. Avoid converting 24, 25, or 30 fps material to 60 fps; adding duplicate or synthesized frames does not restore motion that the source never captured.

Audio

If the audio already uses a compatible codec, copying it avoids another audio generation. AAC is a common choice in MP4 when broad device support matters. Check multi-language audio, commentary tracks, and subtitles before conversion because a simple workflow may select only the default streams.

How to make and verify an H.264 compatibility copy

Identify the codec first, convert only if needed, and compare the result with the source before deleting anything.

1. Identify the source video codec, resolution, frame rate, audio tracks, and subtitles.
2. Open the H.265 to H.264 tool for an HEVC source that the destination cannot decode.
3. Keep the source dimensions and frame rate for the first attempt.
4. Use the tool’s balanced defaults; use a local encoder if you need to set a specific quality or target bitrate.
5. Convert, then test the result on the device or application that rejected the original.

How do I verify that the result is really H.264?

Run the same ffprobe check on the output: the video should report codec_name=h264. Compare its dimensions and frame rate with the source, then inspect duration and all audio or subtitle streams:

ffprobe -v error -show_entries format=duration:stream=index,codec_type,codec_name,width,height,avg_frame_rate -of json 'output.mp4'

Repeat with the source filename. Small duration differences can reflect container timing; missing scenes, tracks, or growing audio drift need investigation.

Scrub through the beginning, middle, and end. Listen for audio drift, confirm the full duration, and inspect areas with motion or fine detail. Check rotation on phone footage and verify that subtitles or extra audio tracks survived when you need them. A file opening successfully on the computer that created it does not prove it will work on the original target device.

If you are adjusting a local encode and the result is too large, reduce quality gradually or consider a modest resolution reduction. If it looks blocky, raise quality rather than increasing the frame rate. For a broader set of inputs and outputs, use the video converter; for HEVC that only needs a standard MP4 delivery path, see the HEVC to MP4 converter.

Frequently asked questions

Are HEVC and H.265 the same thing?

Yes. HEVC is the name of the video compression standard also known as H.265. Neither name describes the file container.

Can I convert H.265 to H.264 without quality loss?

Normal lossy H.264 settings introduce loss. A remux preserves the compressed video but leaves it as H.265. Keep the source as your master and choose the compatibility copy’s quality by inspecting demanding scenes.

Will converting HEVC to H.264 always make a smaller file?

No. In our five-second test, CRF 18 and 23 outputs were larger than the HEVC source, while CRF 28 was smaller. That is one synthetic sample, not a prediction for your footage.