Technology – The Ambarella Difference
Low Power, High Quality HD Video
AT The heart of Ambarella's Products is an exceptionally power-efficient multi-core processor architecture. Innovative algorithms that leverage the architecture’s flexibility achieve a level of video quality that has become the industry’s performance benchmark. Ambarella tapes out multiple chips every year at the most advanced process node available allowing us to deliver high-performance chips that are both very low power and cost efficient. In 2005 we delivered the world’s first Full HD H.264 encoder requiring only a single 16-bit DDR memory, and in 2008 delivered the world’s first 1080p60-capable encoder — years ahead of everyone else.
High Performance Image Processing
In any digital camera the raw video data originating from the imaging sensor requires intelligent pre-processing prior to video encoding. Ambarella’s ground breaking imaging algorithms combined with ultra-high-speed pixel processing enhance dynamic range, eliminate noise and deliver outstanding color fidelity. Optimized over many generations of designs, it delivers the highest-quality images under the most challenging of lighting conditions. This technology enables unrivaled image quality for camera applications ranging from consumer HD hybrid cameras to professional high-resolution surveillance cameras.
Broadcast-Class H.264 CODEC
All H.264 encoders are not created equal. Leveraging Ambarella’s experience from professional broadcasting, we picked the right tools from the H.264 specification and combined them with our innovative algorithms to tailor chipsets for individual markets. Ambarella's renowned H.264 CODEC is tuned to deliver the highest possible HD video quality at very low bit rates. Shoot more pristine HD video before filling up a storage card. Deliver the highest quality video over bandwidth-restricted networks.
MCTF for Efficient Noise Reduction
Ambarella's motion-compensated temporal filtering (MCTF) implementation cleans up image noise with minimal motion blur or loss of sharpness. This algorithm analyzes every pixel of data in every frame in both the spatial and temporal domains, and is then able to intelligently remove static and dynamic noise patterns. The resulting video is dramatically cleaner. This also increases compression efficiency—bandwidth isn't wasted on encoding unwanted noise. Used in high-end video processing gear for many years, Ambarella is first to bring this level of technology to consumer price-point devices.
Highly Programmable Multi-Core Architecture
Our customers are constantly dreaming up new ways to deliver amazing feature sets to the end user. They challenge us to provide the technology that makes those features possible. They need a flexible, programmable architecture that can be tailored for custom applications. Ambarella’s multi-core DSP architecture is highly scalable and strikes a perfect balance between software programmability, hardware-accelerated performance with low power. Combined with extensive on-chip peripherals and high speed ARM CPUs it enables fast time to market with product differentiation.