HEVC is usually the safer library codec for a mixed-device household when older TVs, streaming boxes, phones, and browsers must Direct Play reliably today. AV1 becomes the better storage-and-delivery choice when the clients that matter can hardware-decode it and the server can avoid expensive fallback transcodes. AV1 can offer higher compression efficiency, but a smaller file does not help if one important endpoint turns every playback session into a server conversion.
Client Decode Support Is the First Gate, Not Compression Efficiency
For a home library, the codec decision should begin at the least-compatible client you intend to keep. A modern television in one room and an older streaming stick in another can produce opposite answers from the same file. The winning codec is the one that keeps the most important sessions on a direct path without forcing unnecessary replacement hardware.
AOMedia describes AV1 as an open codec designed for greater compression efficiency than previous codecs. That higher-efficiency AV1 design goal creates real storage and bandwidth upside, but the specification itself does not guarantee that every household endpoint has a hardware decoder.
If every primary client already supports AV1, efficiency can move to the front of the decision. If one frequently used client does not, HEVC may remain the more practical common denominator even when its files are larger at an equivalent subjective quality target.
HEVC Has the Advantage When the Device Fleet Is Uneven
HEVC has had a longer hardware deployment window across 4K televisions, streaming devices, phones, and GPUs. That does not make support universal, but it often makes HEVC the easier codec to keep direct across a household containing several hardware generations.
Roku's current media-player documentation, for example, lists H.265/HEVC on 4K-capable devices while AV1 is supported on only some Roku models. The exact wording of that device-dependent AV1 support illustrates why a mixed-device library cannot treat codec age as an abstract spec.
HEVC wins this stage when replacing older endpoints would cost more or create more friction than the storage savings from AV1. It stops winning when the household has already moved to AV1-capable hardware and compatibility ceases to be the binding constraint.
AV1 Starts to Win After the Playback Fleet Crosses a Support Threshold
Once the primary televisions, streaming boxes, computers, and mobile devices can decode AV1 efficiently, the tradeoff changes. Smaller files or lower streaming bitrates can reduce long-term storage growth and remote bandwidth pressure without requiring the server to convert the source for each session.
Apple's Safari 17 release notes added AV1 capability reporting for devices with hardware support, which is a useful reminder that software support can still depend on the underlying device. The hardware-dependent AV1 capability reinforces the need to test the actual endpoint rather than the operating-system label alone.
AV1 therefore becomes a strong library default only after the devices that matter cross the compatibility threshold. A new codec is not future-proof if today's family-room client forces a transcode every evening.
The Server Pays for Every Compatibility Miss
When a client cannot accept the stored codec, the server must usually convert the video into something it can decode. That turns a storage decision into a compute decision: one AV1 file may save space but require live AV1 decode plus H.264 or HEVC encode for older endpoints.
Plex publishes device-specific media-format guidance showing that Direct Play depends on the client and supported video encoding. Its client-format compatibility boundary is the right operational test: store the codec that minimizes unavoidable conversions across the devices you actually use.
| Household condition | HEVC tendency | AV1 tendency |
|---|---|---|
| Several older 4K TVs or sticks | Safer default | Higher transcode risk |
| Mostly recent AV1-capable clients | Still compatible in many cases | Efficiency advantage becomes usable |
| Server lacks AV1 hardware decode | Lower fallback risk | Software decode can be expensive |
| Remote bandwidth is tight | Good with suitable bitrate | Stronger if client support is verified |
| Archive is rarely re-encoded | Mature playback path | Worth considering for new media, not automatic migration |
The choice flips only when the reduced bitrate or storage demand is larger than the cost of the compatibility misses it creates. Count actual transcodes over a normal week before deciding that the newer codec is operationally cheaper.
Server Hardware Must Support the Source and the Fallback Path
Codec selection also changes server requirements. AV1 software decoding and encoding can be demanding, so a household that stores AV1 while retaining older clients should verify that the media server can decode AV1 in hardware and encode the fallback format fast enough for the expected simultaneous sessions.
Jellyfin's NVIDIA acceleration guide notes that AV1 hardware decode and encode support arrived in different GPU generations. Its separate AV1 decode and encode generation limits shows why “the GPU supports AV1” is not a complete statement.
The same principle applies to integrated graphics and other vendors: verify both directions of the conversion. If the server can decode AV1 but cannot encode the format required by the client in hardware, the bottleneck simply moves to another stage.
Do Not Re-Encode a Healthy Library Just to Standardize the Codec
A mixed library is not automatically a problem. Existing HEVC files that Direct Play everywhere can remain HEVC while new content gradually adopts AV1 after the client fleet supports it. For most homes, coexistence is cheaper than a large one-time transcode that consumes compute, time, and potentially visual quality.
The ZimaSpace mixed 4K and mobile playback guide already frames media hardware around the least-compatible client and the conversions that cannot be avoided. Codec migration should follow the same rule: change the library only when it reduces total storage or delivery cost without creating a larger live-transcode burden.
Keep originals or a reliable backup before any bulk conversion. A codec decision is reversible only if the source remains available; repeatedly transcoding already compressed media compounds loss and makes future migration harder.
FAQs
Should You Convert an Entire HEVC Library to AV1?
Usually not. Convert new or especially storage-sensitive content first, then watch how often important clients fall back to transcoding. A bulk migration is justified only when AV1 compatibility is already broad enough in your household and the storage savings outweigh the conversion time and quality risk.
Does the Container Matter as Much as the Codec?
It can. A client may support the AV1 or HEVC video bitstream but reject the file container, audio track, subtitle format, profile, or level. Test the complete file on each important client rather than treating video codec support as the only Direct Play requirement.
Is AV1 Always Smaller Than HEVC at the Same Quality?
No fixed percentage applies to every source, encoder, preset, and quality target. AV1 is designed for high compression efficiency, but real file size depends on the encoder implementation and settings. Compare representative encodes instead of applying one universal saving figure.
Choose the Codec That Minimizes Total Household Friction
Choose HEVC when the client fleet spans several generations and reliable Direct Play matters more than maximizing compression efficiency. It remains especially practical when the server would otherwise need to transcode AV1 for frequently used older endpoints.
Choose AV1 when the primary devices can hardware-decode it, the server has a clean AV1 path for the exceptions, and storage growth or constrained remote bandwidth makes the efficiency gain meaningful. That is where AV1's technical advantage becomes an operational one.
If both codecs Direct Play equally well across the household, compare representative file size, encode time, HDR/profile requirements, and future device plans. Stop at the codec that reduces the total cost of storage, bandwidth, and live conversion rather than chasing the newest label.
Product Comparisons
More to Read

Docker vs Virtual Machine for Plex: Which Deployment Route Fits?
A conditional Plex deployment verdict for Docker, virtual machines, or Docker inside a VM, based on shared operational requirements.

8GB vs 16GB vs 32GB RAM for Plex: Which Tier Fits Your Workload?
Choose 8GB for lean Plex, 16GB for moderate shared apps, or 32GB for VMs and bounded RAM workspaces—only when measurements justify it.

Does Dedicated Hardware Acceleration Give Plex a Meaningful Advantage?
Hardware acceleration wins for supported repeated transcodes; CPU-only remains valid for direct play, rare conversions, and unsupported stages.

