AV2 vs H.266: The 20-Year War Over Open Video Codecs

AV2 vs H.266: The 20-Year War Over Open Video Codecs

The release of AV2 marks the next major phase in one of the internet’s longest-running infrastructure battles: the conflict between proprietary, royalty-bearing video codecs and open, royalty-free alternatives.

For more than two decades, video compression standards have shaped:

• Global bandwidth economics
• Streaming platform scalability
• CDN operating costs
• Hardware roadmaps
• Patent licensing ecosystems

What appears to users as a simple “play” button is actually supported by an enormous technical and economic foundation.

Every percentage point of compression efficiency directly affects:

• Streaming quality
• Storage costs
• Network utilization
• Cloud infrastructure spending

At internet scale, even a modest improvement in codec efficiency can save platforms hundreds of millions of dollars annually.

With AV2 officially arriving as the successor to AV1, the broader industry is now entering another long transition cycle that may redefine the economics of video delivery through the end of the decade.

🎬 Why Video Codecs Matter So Much

#

Video codecs are effectively the compression engines of the modern internet.

Without aggressive compression, large-scale streaming platforms would be economically impossible.

Modern platforms such as:

• YouTube
• Netflix
• Bilibili
• Douyin
• Tencent Video

deliver billions of viewing minutes every day.

At this scale:

Compression efficiency = infrastructure cost

A 20% improvement in compression efficiency can translate directly into:

• Lower CDN traffic
• Reduced storage requirements
• Lower transit costs
• Improved streaming quality at fixed bitrates

This explains why codec wars are never merely technical debates.

They are economic wars over internet infrastructure itself.

⚖️ The Core Conflict: Royalties vs Open Standards

#

Historically, the dominant video standards have been controlled by patent licensing pools.

H.264 (AVC)

#

H.264 became the universal codec of the internet era because it balanced:

• Good compression
• Broad compatibility
• Hardware acceleration
• Mature tooling

However, it is heavily patented and managed through licensing organizations such as:

MPEG LA

H.265 (HEVC)

#

HEVC significantly improved compression efficiency over H.264.

But it introduced a major industry problem:

Fragmented patent licensing

Instead of a single licensing body, HEVC became divided among:

• MPEG LA
• HEVC Advance
• Velos Media

This created enormous legal and financial uncertainty for streaming companies.

H.266 (VVC)

#

VVC pushed compression efficiency even further.

Technically, it is extremely advanced.

But commercially, it inherited the same licensing complexity that made HEVC controversial.

For hyperscale streaming platforms, this complexity represents substantial operational risk.

🌍 The Rise of Open Codecs

#

The modern open-codec movement began when Google acquired:

On2 Technologies

in 2010 and open-sourced:

VP8

Google followed with:

VP9

in 2013.

VP9 achieved widespread deployment inside YouTube and significantly reduced Google’s bandwidth costs.

However, VP8 and VP9 remained heavily associated with Google itself.

The broader industry wanted something more neutral and collaborative.

🤝 The Birth of AOMedia

#

In 2015, major technology companies formed:

Alliance for Open Media (AOMedia)

Founding members included:

• Amazon
• Cisco
• Google
• Intel
• Microsoft
• Mozilla
• Netflix

The alliance had a clear objective:

Create a royalty-free, next-generation video codec

This effort eventually produced:

AV1

released officially in 2018.

🚀 AV1 Became a Major Industry Victory

#

When AV1 first launched, adoption was extremely slow.

Early problems included:

• Very slow encoding speeds
• Minimal hardware support
• High compute costs
• Immature tooling

At the time, many in the CDN industry viewed AV1 as impractical.

But over the following years, the ecosystem matured dramatically.

Today, AV1 support exists across:

• Modern GPUs
• Browsers
• Smart TVs
• Mobile chipsets
• Streaming platforms

📉 AV1 Changed Streaming Economics

#

AV1 delivered major bandwidth savings.

Typical estimates showed:

• ~50% improvement over H.264
• ~30% improvement over H.265

at comparable visual quality.

This dramatically altered streaming economics for hyperscale platforms.

Netflix reported significant gains over VP9, particularly on mobile workloads.

Major Chinese streaming companies reached similar conclusions internally.

Even with higher encoding costs, the overall infrastructure economics still favored AV1 deployment.

🖥️ Hardware Ecosystem Maturity

#

AV1 adoption accelerated once hardware decoding became widespread.

GPU Support

#

Beginning around 2020:

• Intel
• AMD
• Nvidia

all introduced dedicated AV1 decode hardware.

Mobile Support

#

By 2022:

• Qualcomm
• MediaTek

began integrating AV1 support into flagship SoCs.

Apple Support

#

Apple joined later, introducing AV1 hardware decoding starting with:

Apple M3

This was a major ecosystem milestone because Apple historically influences media pipeline adoption across the industry.

⚠️ AV1’s Biggest Weakness: Encoding Speed

#

Despite its success, AV1 still suffers from a major limitation:

Encoding complexity

Encoding AV1 video remains significantly slower than:

• H.264
• H.265

This limits its usefulness in:

• Live streaming
• Real-time conferencing
• Low-latency media pipelines

For on-demand streaming and archival storage, AV1 works extremely well.

For real-time workloads, the computational overhead is still problematic.

🔬 AV2 Is Designed to Address Operational Bottlenecks

#

Development of AV2 began under the:

AOM Video Model (AVM)

research initiative.

The next-generation codec is expected to focus on more than just raw compression efficiency.

Its broader objectives include improving practical deployability.

🎨 Improved HDR and Wide Color Gamut Support

#

Modern displays increasingly depend on:

• HDR10
• Dolby Vision
• Wide Color Gamut (WCG)

AV2 introduces improved handling for:

• Color metadata
• HDR encoding workflows
• High-end display optimization

This becomes increasingly important as:

• 8K displays
• Professional monitors
• Premium TVs

continue to grow.

⚡ Real-Time Encoding Improvements

#

One of AV2’s most important engineering goals is improving:

Encoding throughput

If successful, this could expand AV2 into:

• Live streaming
• Interactive media
• Cloud gaming
• Video conferencing

These are markets where AV1 still struggles operationally.

🧠 Machine Learning-Assisted Compression

#

AV2 also introduces ML-assisted coding tools.

This represents a major structural shift in codec design philosophy.

Machine learning can assist with:

• Prediction decisions
• Motion estimation
• Rate-distortion optimization
• Encoding efficiency

This may allow software encoders to evolve more rapidly over time.

🏗️ Why Major Platforms Strongly Support AV2

#

Large streaming platforms are heavily incentivized to support open codecs.

Bandwidth remains one of the largest operational costs for video platforms.

For services operating at hyperscale:

Small efficiency gains produce enormous financial savings

If AV2 delivers even a modest improvement over AV1, the cost reductions could be substantial.

This explains why companies such as:

• Tencent
• Netflix
• Google
• Amazon

continue investing heavily in AOMedia.

🇨🇳 China’s Role in the Open Codec Ecosystem

#

The open codec ecosystem is increasingly global.

Chinese companies and silicon vendors now participate deeply in AOMedia.

Members include:

• Tencent
• Amlogic
• Realtek
• VeriSilicon

This matters because modern codec adoption requires close alignment between:

• Streaming platforms
• Browser vendors
• Silicon manufacturers
• Smart TV vendors
• Mobile SoC designers

Without hardware support, codec adoption remains limited.

🧩 The Hardware Adoption Problem

#

Historically, codec transitions move slowly.

AV1 launched in 2018 but only became mainstream in consumer hardware around:

2022–2023

AV2 will likely follow a similar trajectory.

The specification release does not mean immediate hardware deployment.

Consumer adoption depends on:

• GPU roadmap integration
• Mobile SoC updates
• Browser optimization
• Smart TV silicon refresh cycles

Realistically, mass-market AV2 hardware support may not fully mature until:

2027–2028

📺 Early AV2 Deployments Will Be Limited

#

Initially, AV2 will likely appear in:

• Server-side archival storage
• Premium VOD pipelines
• Experimental streaming deployments
• High-end enthusiast hardware

Broader deployment across fragmented device ecosystems will take years.

This mirrors the exact path previously followed by AV1.

❓Why Major Platforms Still Avoid H.266 (VVC)

#

Technically, VVC is extremely impressive.

Its compression efficiency is excellent.

So why have platforms like:

• YouTube
• Netflix

not embraced it aggressively?

The answer is almost entirely economic.

Licensing Complexity Remains the Problem

#

VVC inherits the same fragmented patent-pool structure that complicated HEVC adoption.

For global streaming companies, this creates:

• Licensing uncertainty
• Legal exposure
• Negotiation complexity
• Cost unpredictability

By contrast, AOMedia provides:

• Royalty-free licensing
• Mutual patent protection
• Broad industry alignment

For hyperscale platforms, this matters more than achieving marginally superior compression ratios.

🔍 Key Metrics the Industry Will Watch

#

Following AV2’s release, several indicators will determine whether adoption accelerates successfully.

Reference Encoder Performance

#

Industry observers will closely examine:

libaom 2.x

particularly:

• Encoding speed
• CPU efficiency
• Real-time viability

Hardware Vendor Commitments

#

Critical signals will come from:

• Intel
• AMD
• Nvidia
• Qualcomm
• MediaTek

once they announce native AV2 decode support.

Streaming Platform Trials

#

Experimental deployments from:

• Netflix
• YouTube
• Tencent

will strongly influence broader ecosystem momentum.

📈 Codec Wars Are Long-Term Infrastructure Battles

#

Codec transitions are never immediate.

They unfold gradually over many years.

H.264 still dominates much of the internet despite being more than two decades old.

AV1 required nearly five years to become mainstream.

AV2 will likely experience a similarly extended coexistence period alongside:

• H.264
• H.265
• AV1
• VVC

The ultimate winner may not necessarily be the codec with the absolute best compression performance.

Instead, the decisive factors will likely be:

• Licensing simplicity
• Hardware support
• Ecosystem alignment
• Operational efficiency
• Encoding practicality

🔚 Final Thoughts

#

The release of AV2 is not simply another codec launch.

It represents the continuation of a much larger transformation in internet infrastructure.

Over the past 20 years, the industry has steadily shifted toward:

• Open standards
• Royalty-free ecosystems
• Cross-platform interoperability
• Infrastructure cost optimization

AV1 proved that open codecs could compete seriously with proprietary standards.

AV2 now attempts to extend that momentum into:

• Real-time streaming
• AI-assisted encoding
• Next-generation HDR workflows
• Future edge media infrastructure

Whether AV2 ultimately becomes the dominant codec of the late 2020s will depend less on theoretical compression ratios and more on how quickly the surrounding hardware and software ecosystems mature.

As with every previous codec transition, the real battle will not be won in laboratories or standards committees.

It will be decided across:

• Data centers
• CDNs
• Browsers
• GPUs
• Smartphones
• Smart TVs

and the economics of streaming at planetary scale.

📚 References

#

• Alliance for Open Media (AOMedia)
• AV1 and AV2 Technical Specifications
• Netflix AV1 Encoding Research
• YouTube AV1 Deployment Documentation
• MPEG LA Licensing Documentation
• HEVC Advance Licensing Information
• Intel AV1 Hardware Decode Announcements
• Qualcomm Mobile AV1 Support Documentation
• Apple Silicon Media Engine Documentation
• VVC (H.266) Standardization Publications