If you’ve been curious about AV1, but indecisive about committing to adding it to your workflow, you’re in the same spot as a lot of good teams.

You’ve probably seen AV1 benchmarks for years. You know the theory: better compression than H.264 and H.265, cleaner artifacts at low bitrate, better long‑term fit for 4K/8K. But when it comes time to prioritize work, AV1 keeps losing to things that seem like a higher priority, and your stuck with the following thoughts:

• “Re‑encoding the whole catalog sounds expensive.”
• “We can’t afford to break playback on long‑tail devices.”
• “Our ladder is already stable; changing codecs feels risky.”

We designed Qencode’s latest AV1 update specifically to unblock that conversation. The goal is not to convince you that AV1 is the next big trend; it’s to make AV1 a reasonable engineering decision you can ship, measure, and justify as you look at the roadmap for the next year.

What actually changed (and why we think it tips the scales)

From a codec‑standards point of view, nothing dramatic happened, AV1 is the same AV1.

What changed over the years is everything around it:

• AV1 encoding on Qencode is now roughly 20× cheaper per minute than our previous rate. That turns “maybe a few hero titles” into “we can realistically apply this to the majority of our catalog.”
• We’ve aligned our API, presets, and packaging guidance around a simple premise: add AV1 next to your existing H.264/H.265 rungs, without needing a large, flag‑day migration.

From your side, this means you can test AV1 on real traffic with the existing Qencode API and your current manifests, and you can easily roll it back without interruptions if you don’t like what you see.

The compression story: what we actually measured

Let’s move to some specific measurements we have done to assess the true benefit of AV1.

We ran H.264 (x264), H.265 (x265), and AV1 (SVT‑AV1) at typical production settings you’d actually ship, not “slowest preset, no time budget” lab settings. For each codec we:

• Used constant‑quality style encoding (CRF/QP‑driven), the way most per‑title pipelines run today.
• Targeted similar perceptual quality using VMAF in the mid‑90s as a proxy.
• Measured resulting bitrate in kbps at three resolutions: 1080p, 4K, and 8K.

At those “equal‑quality” points we saw:

Compression efficiency of AV1 vs H.264

• ≈ 41% lower bitrate at 1080p
• ≈ 52% lower bitrate at 4K
• ≈ 57% lower bitrate at 8K

Compression efficiency of AV1 vs H.265

• ≈ 30% average bitrate reduction across the same resolutions

For these tests, we used a library with the same source material and same quality targets, measured against different codecs. These are single points on the rate–distortion curve, not full BD‑rate studies. Fast sports and noisy UGC show compression-rates that are different than animation or clean studio content. Stay tuned for a more detailed study with breakdowns per title and per genre in the future.

TLDR: If you hold quality constant, AV1 consistently uses a lot fewer bits to get there, and not by a marginal amount.

From fewer bits to real economics

If you think in terms of cost per viewing hour, three inputs make up the total cost:

1. Average bitrate of the rendition actually watched.
2. CDN price per GB for your stack.
3. Overhead from container, protocol, and packaging.

You probably have a 1080p H.264 rung somewhere that works out to a few cents per hour in delivery cost. If AV1 lets you keep the same perceived quality while dropping the bitrate by ~40%, then almost by definition:

• You’re moving to a bit over half as many GB per hour,
• Your delivery cost per hour can decrease drastically.

At 4K, a 50% reduction is a dramatic change in what a 4K hour costs you. At 8K, it’s the difference can be especially significant.

Device reality

The second big blocker for most teams is the fear that AV1 will break playback on some important device class. That fear is rational and device fragmentation is real. The good news is that the ecosystem is finally at a place where a multi‑codec strategy makes AV1 super safe to roll out.

Here’s the rundown:

• Android: recent versions ship with AV1 hardware decoding, and newer platform requirements push OEMs to include AV1 support. A large and growing share of Android phones, tablets, and TVs can happily play AV1.
• Desktop browsers: modern Chrome, Edge, and Firefox support AV1, subject to underlying OS/hardware capabilities. If your audience is desktop‑heavy, your AV1 coverage is already decent.
• Apple ecosystem: newer devices with dedicated media engines (e.g., recent iPhones and M‑series Macs) expose AV1 decode; Safari can play AV1 there. Older Apple hardware will still prefer H.264/H.265.

So what do you do with that? You actually don’t need to pick a side. You can just start with the following:

1. Generate AV1 renditions for the rungs that matter (start with 1080p and 4K).
2. Keep the H.264/H.265 renditions that are already working in your ladder.
3. Package everything to CMAF and emit HLS/DASH manifests that list multiple codec variants per resolution.
4. Let the player’s capability detection select the best supported codec at runtime.

On capable devices, AV1 gets picked automatically and you enjoy the savings and quality. On legacy devices, the player continues to request H.264/H.265 segments. You haven’t lost reach, and you haven’t split your catalog. AV1 basically becomes an opt‑in optimization for compatible clients and practically insignificant from a risk perspective.

Why it’s not just about cost optimization

It’s tempting to categorize AV1 as just cost savings, but that undersells one of the most overlooked aspects of building a successful video platform. Lower bitrates at the same perceived quality aren’t only about saving money; they’re about maximize user-experience during the most important part of the process, the playback.

Here are some massive UX improvements your AV1 viewers will get that are likely to lead to more engagement, faster growth, and more long-term brand loyalty:

• Startup & time‑to‑first‑frame: a player has to fetch and decode a certain amount of data before it can show anything. If your typical first segment shrinks in size because bitrate is lower, users on congested or high‑latency links see something sooner. That affects session start rates and bounce.
• Rebuffering behavior: if your encoded bitrates better match the actual throughput distribution of your users, ABR algorithms spend less time bouncing between rungs, and segments are less likely to miss deadlines. That reduces the “spinner in the middle of the episode” problem.
• Session length & engagement: small QoE improvements often compound. A stream that starts quickly and rarely stalls encourages users to watch “one more episode” or “one more match,” which directly increases either ad impressions or perceived subscription value.

So while AV1 can shrink your egress bill, we often stand by the fact that the most interesting compression-rate story is about improving the robustness of your experience to make each viewer easier to monetize, upsell, and retain.

Start free → qencode.com/signup