H.264 vs H.265 (HEVC) vs AV1: video codecs compared

H.265 (HEVC) and AV1 are modern video codecs that compress video far more efficiently than the older H.264 (AVC) standard, delivering identical visual quality at roughly 40 to 50 percent lower bitrates. Between the two newer codecs, AV1 provides an additional 20 to 30 percent bandwidth savings over HEVC and is completely royalty-free, making it the preferred standard for YouTube and Netflix.

For Mac owners, playback efficiency depends on your processor generation: every Apple Silicon Mac decodes H.264 and HEVC directly in hardware, while dedicated AV1 hardware decoding is built into M3, M4, and newer chips. M1 and M2 Macs still play high-resolution AV1 videos smoothly using multi-threaded CPU software decoders like dav1d, but hardware decoding preserves significantly more battery life during long playback sessions.

What does a video codec actually do?

A video is simply a fast succession of still photographs played at 24, 30, or 60 frames per second. Uncompressed, 4K video at 60 frames per second with 10-bit color takes around 10 gigabits per second: a one-terabyte drive would fill in about a quarter of an hour. A video codec compresses this large stream down to a manageable size by eliminating redundant visual information.

Codecs achieve compression through two primary methods: spatial compression and temporal compression. Spatial compression simplifies repeated patterns within a single still frame, much like a JPEG image. Temporal compression evaluates changes across successive frames over time.

In a typical video scene, the background often remains static while an actor speaks or walks across the frame. Instead of storing the background repeatedly in every frame, the codec stores a single complete reference image and then records only the pixels that move or change in subsequent frames.

Animation

A keyframe, then only what changed

This sequence of frames is organized into a Group of Pictures (GOP):

1. Intra-coded frames (I-frames): Full keyframes that contain an entire image independent of any other frame.
2. Predicted frames (P-frames): Frames that record changes relative to the preceding I-frame or P-frame.
3. Bi-directional predictive frames (B-frames): Highly compressed frames that interpolate changes by looking both forward and backward to surrounding frames.

The differences between H.264, HEVC, and AV1 stem from how intelligently their mathematical algorithms predict these changes, how small their pixel evaluation blocks can be divided, and how efficiently their entropy encoders pack the resulting data.

How much more efficient are HEVC and AV1 compared to H.264?

Video engineers measure compression efficiency using Bjøntegaard Delta bitrate (BD-rate), an objective mathematical curve that quantifies the percentage bitrate savings between two codecs at identical objective or perceptual visual quality. Perceptual quality is typically evaluated using VMAF (Video Multi-Method Assessment Fusion), a machine-learning metric developed by Netflix and academic laboratories.

Extensive industry benchmarks reveal clear performance tiers among the three codecs:

• HEVC vs H.264: published comparisons, including tests by Netflix engineers, put HEVC’s saving at roughly 30% to 50% for the same quality.
• AV1 vs H.264: testing published by Meta (then Facebook) engineers demonstrated that AV1 achieves a 50.3% bitrate reduction compared to x264 Main Profile, and a 46.2% reduction compared to x264 High Profile.
• AV1 vs VP9 and HEVC: the same tests found 34.0% savings against VP9 (libvpx); comparisons with HEVC generally put AV1 about 20% to 30% lower.
• In practice: services such as Netflix and YouTube now stream AV1 to devices that can decode it, because the same quality costs them fewer bits.

Diagram

Bitrate needed for similar quality (H.264 = 100)

• H.264 (AVC)100
• H.265 (HEVC)about 50–70
• AV1about 40–50

Fewer bits for the same picture means higher resolutions over slower connections, and lower bandwidth bills for the services. To learn how bitrate impacts overall fidelity, consult our guide on What is bitrate?.

What is the difference between codecs and encoders like x264 and x265?

People often use terms like “H.264” and “x264” interchangeably, but they represent entirely different concepts in video engineering. Confusing a codec standard with an encoder software implementation is common when reading digital video release labels.

A codec is a formal specification written on paper. For example, ITU-T Recommendation H.264 and ISO/IEC 14496-10 define the exact mathematical bitstream syntax that decoders must recognize to unpack a compliant stream. The specification does not supply software code; it simply outlines the rules.

An encoder is the actual software program or hardware silicon block that ingests raw video frames, runs compression calculations, and produces a compliant bitstream:

• H.264 (AVC) is the standard; x264 is the open-source software encoder developed by the VideoLAN community.
• H.265 (HEVC) is the standard; x265 is the open-source software encoder maintained by MulticoreWare.
• AV1 is the standard; SVT-AV1 (developed by Intel and the Alliance for Open Media) and libaom are software encoders that write AV1 bitstreams.

When you see a file tag labeled x265 in a movie release, it indicates that the file was compressed using the open-source x265 software engine rather than a hardware encoder like Apple VideoToolbox or Intel Quick Sync. You can decode and understand these tags in our detailed walkthrough on Video file names explained.

Software encoders like x264 and x265 offer deep algorithmic customization, producing higher visual quality per megabyte than fast hardware encoders, but at the expense of heavy processor utilization and extended encoding times.

Why does patent licensing matter for H.264, HEVC, and AV1?

The commercial success of a video codec is determined as much by legal patent licensing terms as by mathematical compression efficiency. The divergence between HEVC and AV1 illustrates this reality.

H.264 was licensed primarily through a single patent pool administered by MPEG LA. The terms were straightforward, royalties were capped at predictable annual thresholds, and free internet streaming was exempted from per-stream fees. This licensing clarity made H.264 the universal video format across the entire web for over fifteen years.

HEVC, by contrast, suffered from a fractured licensing dispute. Multiple competing patent administrators—including MPEG LA, HEVC Advance, and Velos Media—formed independent pools alongside several non-pooled patent holders. Licensing terms demanded royalties on consumer playback hardware, content distribution, and streaming subscriptions, with no unified royalty cap.

Because hardware manufacturers and open-source browser vendors faced substantial legal and financial liabilities, major web platforms balked at adopting HEVC:

1. Web browsers like Google Chrome and Mozilla Firefox refused to support HEVC playback for years.
2. Streaming giants formed a coalition to bypass patent pool demands entirely.
3. In 2015, Google, Mozilla, Cisco, Microsoft, Amazon, Netflix, and Intel founded the Alliance for Open Media (AOMedia) to develop an unencumbered, royalty-free codec: AV1.
4. Apple formally joined the governing board of AOMedia in January 2018.

Because AV1 carries zero patent licensing royalties, it has rapidly become the standard for YouTube, Netflix, and modern web browsers, avoiding the complex commercial barriers that hampered HEVC.

How does Apple Silicon handle H.264, HEVC, and AV1 hardware decoding?

Every computer processor can decode video using either software decoding or dedicated hardware decoding. The difference between these two approaches determines whether your MacBook battery lasts through an entire transcontinental flight or depletes in two hours.

Software decoding uses general-purpose CPU cores to calculate frame mathematics. This consumes significant wattage, increases core temperatures, and triggers cooling fans. Hardware decoding routes the bitstream through fixed-function silicon circuits on the Apple Silicon chip, known as the Media Engine. The Media Engine processes 4K video frames using negligible power, leaving the main CPU and GPU idle.

Apple’s hardware decode support across chip generations is clearly delineated:

• H.264 (AVC): Supported in hardware across all Apple Silicon processors (M1, M2, M3, M4, and newer) as well as legacy Intel Macs via Apple’s VideoToolbox framework.
• H.265 (HEVC): Supported in hardware across all Apple Silicon processors (M1 through M4) for both 8-bit and 10-bit color, including HDR10 and Dolby Vision.
• AV1: Supported in dedicated silicon hardware decoding on Apple Silicon M3, M4, and newer chips, as well as the iPhone A17 Pro processor. Apple exposes this in CoreMedia under the identifier kCMVideoCodecType_AV1 (av01).

M1 and M2 chips have no AV1 circuits, so players decode it in software. VideoLAN’s open-source dav1d decoder is heavily optimized for Apple Silicon’s CPU cores and usually plays even 4K AV1 smoothly, at the cost of more power. For more on chip compatibility, read our full guide on AV1 on Mac.

H.264 vs H.265 vs AV1: full technical comparison

Evaluating all three codecs across their technical attributes demonstrates why the digital video industry is actively transitioning from legacy standards toward modern alternatives.

The larger block sizes in HEVC (64×64 pixels) and AV1 (128×128 pixels) allow both modern codecs to compress large flat areas—such as skies, walls, and out-of-focus backgrounds—using far fewer bits than H.264’s rigid 16×16 macroblocks.

Which codec should you choose for watching and encoding?

Because each codec presents different trade-offs between compatibility, encoding time, and compression efficiency, the ideal format depends on whether you are distributing media to others or archiving content for yourself.

For watching video:

• On M3 and newer Macs, AV1 streams are decoded in hardware: let services send it.
• Choose HEVC for high-end local movie archives and 4K HDR playback, as it pairs with hardware decoding across all Apple Silicon Macs and Apple TV 4K hardware.
• Rely on H.264 when playing video on older legacy hardware, in-car entertainment systems, or vintage smart televisions.

For encoding and exporting video:

• Choose H.264 if your primary requirement is universal compatibility. If you are emailing a video clip to colleagues or embedding a file on a general website, H.264 plays reliably on every screen manufactured in the last fifteen years.
• Choose HEVC if you are archiving home movies or personal 4K footage. Using Apple’s hardware VideoToolbox encoder exports HEVC files in minutes while halving your storage requirements compared to H.264.
• Choose AV1 if you are publishing content to online streaming platforms or want the smallest file sizes for web distribution, provided you have the processor power to accommodate slower encoding passes.

Pick the codec for where the video will be played: the newest one only helps if the devices that play it can decode it.