
It is critical to pick an
appropriate Codec or format that can be handled by a video IP to
support
any given application, according to system-on-chip (SoC) designers. It
is also imperative to select the right video IP with standard
interfaces so that the video IP can be as close as possible to
plug-and-play in terms of the SoC integration.
Commenting on the importance of selecting the right video IPs for SoC designs, Ashwin Gayatri, vice president, Products & Technology Group, Patni, said that video compression/decompression is inherently very compute intensive. In fact, video encoding is said to be much more compute intensive than video decoding. This actually translates into high power consumption. “If a designer is looking to target the SoC with video capability for a portable device, then the designer would need to look for the most optimized video IP,” he said.
If there is a need to have a video encoder on the SoC for applications like videoconferencing, etc, the output of the video quality becomes a very important aspect as well. The best video encoder optimizes the power and the bit rate requirement for a given video quality.
According to Ravishankar
Ganesan, vice president, SoC IP Business Unit, Ittiam Systems, SoCs
define and integrate multiple specialized blocks or subsystems. Each
one of these specialized subsystems needs to be the best in terms of
its performance, area and power so that the SoC can be the most
competitive and the best suited for the target market. Video
intellectual property (IP) is one of these specialized subsystems, and
critically important for the SoCs targeted for video-based
applications. However, Ganesan cautioned that there is no
“one fits all” video IP.
Sasken Technologies has been putting a lot of video stuff in software
on the DSP. According to Dr G Venkatesh, CTO, strategy officer, and
executive director, Sasken, the company has had a strong presence in
video and a lot of its video IPs are shipping globally.
Sasken strongly believes that, with the exception of certain acceleration blocks, etc, a lot of video functions should be done in software on the DSP. This is the path that Sasken has adopted over the last decade or so. Dr Venkatesh added that a lot of video functions can be carried out by programming some of the more advanced processes in the embedded systems.
The logic behind this is that, as the video algorithms keep changing, so too do they improve. Adding this in hardware means that the designer does not have the flexibility to keep changing it. However, if the video IP is in software, it becomes quite easy to replace one algorithm with another.
Video Profiles
On the aspect of what SoC designers need to look for in terms of video
profiles and Codecs, Ittiam’s Ganesan feels that it really
depends on the application(s) the SoC is likely to address.
“If you are targeting video IP for some mobile TV application
in a cell phone, the profiles and Codecs would be determined by the
appropriate broadcasting system.”
Similarly, if the SoC is targeting the high-definition (HD) DVD player segment, the video Codecs and their profiles/levels need to be determined based on the video encoder configuration that was used to create the content on the DVD disc.
Gayatri of Patni reiterated that the video profiles are determined by the target applications. For bandwidth and power-constrained operation, as is the case with portable devices, entry-level profiles make sense. For high-end devices – with mains power supply and big form factors, where the viewing experience is important – the profiles at the upper end are suitable.
The different Codecs trade off quality against computational complexity and bandwidth requirements. However, it may not make sense to go with some of the older Codecs as the designer might not be able to use it to make complete use of the capabilities offered by the newer silicon. Besides, the output quality and bandwidth requirements would make it untenable.
by Pradeep Chakraborty, Bangalore