H.264 vs H.265 - AVC vs HEVC - What's the difference?
H.265 and H.264 are the original video coding standards established by the video coding expert group of the ITU Telecommunication Standardization Sector (ITU-T).
With the rapid development of video surveillance technology, the current H.264 (MPEG-4/AVC) standard has ceased to meet video encoding requirements to offer higher image resolution. In the not too distant future, 4K UHD and 8K UHD video equipment will inevitably become the standard, as is Full HD today. H.265 has entered the market as the next-generation video compression standard and is gradually being introduced into high-definition IP surveillance products such as HD IP cameras, NVRs.
What is H.264 (MPEG-4 AVC)?
H.264, also known as MPEG-4 AVC (Advanced Video Codec), is a video compression standard developed in 2003 and a widely used format for high-fidelity video recording, compression, and distribution. H.264 is known to be the standard codec for Blu-ray discs. All Blu-ray players must be able to decode H.264. The vast majority of modern video recorders (DVRs) use H.264 as their main codec.
At the time, the development of the H.264 codec was a true technological advance, because it put the people involved in television, IP cameras and conference calls at the same table and gave birth to a standard that, in general, and by a long period of time, was good enough for everyone.
H.264 video resolution
H.264 only recognizes resolutions up to 2048 x 2048
Commonly supported resolutions and width-to-height ratios include:
- 854 x 480 (16:9 480p)
- 1280 x 720 (16:9 720p)
- 1920 x 1080 (16:9 1080p)
- 640 x 480 (4:3 480p)
- 1280 x 1024 (5:4)
- 1920 x 1440 (4:3)
Apple’s decision to use H.264 has contributed to some extent to the popularization of this encoding. This standard has entered millions of households, with hundreds of millions of iPads and iPhones, and has become the absolute standard in image coding with more than 80% market share.
Compared to previous encoding standards, H.264 can produce higher image quality at a lower bit rate, which is why it has been adopted by most people. H.264 is also widely used in network multimedia transmission and in various industries of high definition terrestrial television, satellite television, broadcasting and other industries.
Initially, the problems of the massive introduction of the H.264 codec were due to the fact that decoding of HD video in real-time required great power, for those times, from the hardware part of computer technology. The situation normalized with the entry into the market of multi-core AMD and Intel processors when they became more accessible to the common user.
What is H.265 (HEVC)?
The H.265 or HEVC (High-Efficiency Video Coding) standard, developed in 2012, builds on its predecessor, the H.264 video coding standard, retaining some technologies and improve others. H.265 uses advanced techniques to improve the relationship between codestream, encoding quality, delay, and algorithm complexity for optimal settings.
H.265 supports frame formats up to 8K (UHDTV) with a resolution of 8192 × 4320 pixels
H.265 enhancements over H.264 include:
- increased compression efficiency,
- increased reliability
- the ability to recover from errors
- decrease in real time latency
- reduced channel acquisition time and random access latency, and reduced complexity.
The H.265/HEVC encoding architecture is roughly similar to H.264/AVC and mainly includes:
- internal prediction
- external prediction
- transformation
- quantification
- deblocking filter,
- entropy coding and other modules.
However, in HEVC encoding architecture, everything is divided into three main blocks:
- coding unit (CU)
- prediction unit (PU)
- conversion unit (TU)
H.265 vs H.264
The difference between H.264 and H.265 lies mainly in the transmission bandwidth and storage requirements. H.265, thanks to the optimization of the H264 algorithm, can realize standard definition digital image transmission at a speed lower than 1Mbps. H.265 can realize conventional 720P high-definition video and audio transmission (1280 resolution x 720) at a transmission speed of 1-2 Mbps.
H.265 is designed to deliver higher quality network video with limited bandwidth, and only half the bandwidth of H.264 can be used to play video of the same quality. The H.265 standard also supports 4K (4096 × 2160) and 8K (8192 × 4320) ultra-high-definition video.
The H.265 / HEVC encoding architecture is very similar to H.264/AVC. It mainly includes Intra prediction, inter-prediction, transformation, quantization, unblocking filter, entropy coding and other modules. However, in HEVC coding architecture, everything is divided into three basic blocks: a coding unit (CU), a prediction unit (prediction unit, PU), and a transform unit (transform unit, TU).
Compared to H.264/AVC, H.265/HEVC provides various tools to reduce the bit rate. Regarding the encoding unit, the size of each macroblock (MB) in H.264 is a fixed 16x16 pixels, and the H.265 encoding unit can be selected from the smallest 8x8 to the largest 64x64. At the same time, the intra H.265 prediction mode supports 33 directions (H.264 only supports 8) and provides better vector prediction and motion compensation processing methods.
Stay Ahead in the Codec Game!
Subscribe now to receive in-depth analysis, comparisons, and the latest updates and industry insights
The quality comparison test shows that with the same image quality compared to H.264, the size of H.265 encoded video will be reduced by about 39%-44%. When the bit rate drops by 51%-74%, the quality of H.265 encoded video may be the same or better than H.264 encoded video, which is significantly better than the expected signal to noise ratio (PSNR).
Comparison of H.264 and H.265 codecs
|
H.264 AVC |
H.265 HEVC |
Block size |
Macroblock 16 x 16 |
Blocks with a tree-like encoding structure from 64x64 to 8x8 |
Prediction blocks |
Splitting up to 4x4 |
64x64 to 4x4 + asymmetric prediction |
Conversion blocks |
8x8 and 4x4 |
32x32, 16x16, 8x8, 4x4 + non-square transformations |
Reverse transformation |
Deblocking filter |
Unblocking filter, SAO |
intra-frame prediction |
9 modes |
35 modes |
Motion compensation |
Motion vector prediction |
Advanced motion vector prediction (spatial and temporal) |
Colour depth |
8 bit |
10 bit |
Entropy coding |
CABAC or CAVLC |
CABAC with parallel operations |
H.265 support in protocols
HLS has been compatible with H.265 for a long time. The H.265 format is compatible with the MPEG-TS protocol. In the RTSP protocol, H.265 is supported. There is packaging in both SDP and RTP.