H.265 delivers similar picture quality to H.264 at roughly half the bitrate, so it needs about half the storage and upload bandwidth, but it asks more of your recorders, software and licensing. In the H.265 vs H.264 choice for a security system, that usually makes H.265 the better choice for new 4K cameras and long retention, and H.264 the safer choice where every device in the chain must work without a hitch. This guide compares the two codecs for surveillance and shows the storage arithmetic. If bitrate is new to you, read what bitrate is first.
Key takeaways
- H.264 was standardized in 2003 and H.265, also called HEVC, in 2013.
- H.265 can match H.264 quality at roughly half the bitrate.
- At an example 4 Mbps for H.264, one camera stores 1,296 GB in 30 days. At 2 Mbps for H.265 it stores 648 GB.
- For 16 cameras over 30 days that is about 20.7 TB versus 10.4 TB.
- H.264 is supported by nearly every recorder. Check H.265 support on each device before you switch.
H.265 vs H.264: What Is the Difference?
Both are video compression standards. A camera captures far more data than it can store, so it compresses each frame by describing what changed instead of repeating everything. H.264, also called AVC and published by the ITU in 2003, is the long-time default. H.265, published in 2013, is its successor and was designed to deliver the same quality in a smaller stream, especially at high resolutions. You can read the official specifications at the ITU for H.264 and H.265.
Why the savings grow with resolution
A 1080p frame has 2,073,600 pixels. A 4K frame has 8,294,400, about 4 times more. Larger pictures contain more smooth, predictable areas that a newer codec can describe compactly, so the gap between the two standards matters most on 4K cameras and shrinks on small, low-resolution streams.
H.265 vs H.264 Compared for Security Cameras
| Factor | H.264 (AVC) | H.265 (HEVC) |
|---|---|---|
| Standardized | 2003 | 2013 |
| Bitrate for similar quality | Baseline | Roughly half |
| Example storage, 1 camera, 30 days | 1,296 GB at 4 Mbps | 648 GB at 2 Mbps |
| Example storage, 16 cameras, 30 days | 20.7 TB | 10.4 TB |
| Example upload, 16 cameras | 64 Mbps | 32 Mbps |
| Recorder and software support | Nearly universal | Common, but verify per device |
| Processing needed to encode and decode | Lower | Higher |
| Licensing | Simpler | More complex, several patent pools |
The bitrates in this table are examples used to show the arithmetic, not measurements of any product. Real savings depend on the scene, the encoder and the settings, so test on your own cameras.
Which Codec Should You Choose?
Choose H.265 when storage or bandwidth is the constraint
Long retention windows, many 4K cameras, sites with limited upload, and cloud analytics all reward a smaller stream. A 30 to 90 day retention window is common, and at 90 days an example 4 Mbps camera needs 3,888 GB while a 2 Mbps stream needs 1,944 GB, which is the difference between needing a larger recorder and not needing one.
Choose H.264 when compatibility comes first
If you have older recorders, mixed camera brands, or operators who review footage on many different devices, H.264 plays almost everywhere. Switching to H.265 on a camera whose recorder cannot decode it gives you a black tile, not a saving.
Beware vendor extensions
Many makers add branded modes such as smart codecs or plus modes that cut the bitrate further, often by changing how often full reference frames are sent. They can work well, but they are vendor-specific, so confirm that your recorder and analytics software read the stream correctly. Our guide to DVR vs NVR explains how recorders handle IP streams and why open standards reduce lock-in.
How Does the Codec Affect Video Analytics?
Analytics read the same compressed stream you record. Both codecs work, provided the software can decode them and the quality is good enough to see the object of interest. Argu connects to existing RTSP or ONVIF streams, IP cameras and NVRs and reads feeds at 1080p to 4K and 15 to 30 frames per second over H.264 or H.265, so a mixed estate does not need to be unified first. The point to watch is not the codec label but the bitrate: an over-compressed stream loses small, distant details whichever standard produced it. See how the AI camera options trade off bandwidth, hardware and flexibility, or look at the Vision Agents platform to understand how analytics attach to the cameras you own.
How Do You Test H.265 Before You Switch?
A codec change touches every device that stores, plays or analyzes the video, so test it on a small group first. A simple pilot takes about 1 week and answers most questions.
Run a 4-step pilot
- Pick 2 cameras: one indoor, one outdoor facing the hardest light, such as a gate with headlights at night.
- Record each for 7 days in H.264 and in H.265 at the bitrate you plan to use, and note the daily file sizes.
- Play both on every viewing device, including the operator workstation, a phone and any export tool used for investigations.
- Check that exported clips open for the people who receive them, such as insurers or investigators, who may not have an H.265 player.
Common mistakes
The first mistake is judging quality from a daytime still. Compression artifacts show up at night and on moving subjects, so compare those. The second is changing the codec without lowering the bitrate, which keeps the storage cost and gains nothing. The third is forgetting the export path: footage that plays on the recorder but not on a court officer’s laptop is a failure at exactly the wrong moment. A fourth, quieter mistake is changing frame rate at the same time. A drop from 30 to 15 frames per second reduces data on its own, so change one setting at a time or you will not know which one produced the saving.
What about retention rules?
Retention periods are set by policy and law, not by the codec. Public systems often keep routine footage for 30 to 90 days, and privacy rules such as GDPR in the EU limit how long identifiable footage should be kept. A more efficient codec lets you meet a given window on less disk, but it never changes how long you are allowed or required to keep the footage.
Frequently Asked Questions
Is H.265 better than H.264?
For storage and bandwidth, yes: it delivers similar quality at roughly half the bitrate. For compatibility and simplicity, H.264 remains easier because nearly every device supports it.
Is HEVC the same as H.265?
Yes. HEVC stands for High Efficiency Video Coding and is another name for H.265. AVC, or Advanced Video Coding, is the name for H.264.
How much storage does H.265 save?
Often around half compared with H.264 at similar quality. In the example above, one camera over 30 days drops from 1,296 GB to 648 GB.
Can I mix H.264 and H.265 cameras?
Usually yes, if the recorder supports both. Confirm decoding and playback on every device that will view the footage.
Does H.265 improve image quality?
Not by itself. It keeps the same quality in less data, or gives better quality at the same data rate. The actual picture also depends on the sensor, lens and lighting.
Which codec is better for AI video analytics?
Either can work. Choose the one your cameras and recorders support well, and make sure the bitrate leaves enough detail for the objects you want to detect.
The right codec is the one that fits your recorders, retention and network. If you want to know how analytics would run on the streams you already have, talk to Argu.
Last updated: October 2026
