Following the launch of Arm’s first ever Total Compute solutions in May 2021, we are exploring how the different Premium, Performance and Efficiency solutions can be applied to different consumer device segments. We have already explored the smartphone and laptop segments, but in this blog we are outlining the different Total Compute solutions for the home market, particularly those for the range of Digital Television (DTV) and Set-top box (STB) devices.
DTVs and STBs have evolved and continue to evolve into multi functional devices with far richer user experiences. This not only includes high-quality video streaming from various content providers, such as Netflix, Disney+, Amazon Prime and YouTube, but also other new use cases, such as video calling, health and fitness applications and gaming. Often based on operating systems like Android, these devices are even allowing third-party developers to create and add more applications experiences, just like for the smartphone. As this ecosystem continues to evolve and grow, DTVs in particular are becoming a central control hub, not just for entertainment, but for device management in the home and security. To meet these new opportunities, DTVs and STBs require more system performance, greater artificial intelligence (AI) capabilities and enhanced security protections. The new Total Compute solutions for the Home market are designed to address these requirements through improving system performance, security and developer access.
Arm’s Total Compute solutions offer a full suite of scalable hardware IP (including the latest Armv9 CPUs, Mali GPUs and System IP), physical IP, software, tools, and standards to build the best SoC across different consumer device markets. These solutions offer different configurations for DTV and STB devices across different tiers of devices. Premium DTVs and STBs can utilize Performance solutions. For premium DTVs, this consists of a 4+4 CPU configuration of 4x Arm Cortex-A710 and 4x Arm Cortex-A510 alongside the mid-level Arm Mali-G510 GPU or even the premium Arm Mali-G710 GPU. Meanwhile, for premium STBs, this consists of a 2+2 CPU configuration of 2x Cortex-A710 and 2x Cortex-A510, alongside Mali-G510.
Mid-range and entry-level DTVs and STBs can utilize more efficient Total Compute solutions. For mid-range and entry-level DTVs, this consists of a 2+6 CPU configuration of Cortex-A710 and Cortex-A510 alongside the ultra-efficient Arm Mali-G310 GPU. Whereas for mid-range and entry-level STBs, the CPU configuration would be 4x A510 CPUs, alongside the Mali-G310.
As with the Total Compute solutions for smartphones and laptops, underpinning the CPUs and GPUs in our Total Compute solutions for laptops is our System IP – CoreLink Interconnect CI-700 and NI-700. Both interconnect technologies provide improved energy efficiency and system performance to add further improvements across any Total Compute solution for any tier of DTV or STB device. Moreover, all of the CPU configurations in our Total Compute solutions are bound together by Arm’s new DSU-110, which is the backbone of any Cortex CPU cluster, enabling designers to address different trade-offs in performance and efficiency.
The CPU configurations for premium, mid-range and entry-level DTVs and STBs combine Arm significant and LITTLE CPUs into a single, fully integrated solution. These all provide advanced compute capabilities, faster responsiveness, and increased power savings. Arm big.LITTLE technology is a heterogeneous processing architecture that uses two types of processor. 'LITTLE' cores are designed for maximum power efficiency, while 'big' cores are designed to provide maximum compute performance. With these two dedicated processors for different specialized processing needs, the big.LITTLE solution is able to adjust to the dynamic usage pattern for different consumer devices, which include high-processing and low-processing intensity tasks.
Regardless of the tier of DTV and STB devices, there are now increasing demands for machine learning (ML) performance and more security features. More ML performance is needed to meet the requirements of the new AI-based use cases on DTVs and STBs. These use cases include automated content classification and consumption through streaming services and picture quality enhancements. In fact, AI super resolution is becoming a crucial technology in the picture quality process on these devices. While 4K DTVs and STBs dominate the current market, it is only a matter of time before more 8K DTVs and STBs enter the market and are shipped more worldwide.
In order to address diverse AI requirements, the approach for the different Total Compute solutions is to integrate powerful AI capabilities across all of Arm’s IP. This means our Total Compute solutions are able to offer the broadest range of AI engines for diverse AI use cases across different power and cost constraints (see image below). This is reflected in the IP that were part of the most recent Total Compute solution launch. Matrix Multiply instructions for the Int8 and BF16 and SVE2 in the Armv9 CPUs help to double the ML performance of Cortex-A710 compared with the Arm Cortex-A78 CPU. This also helps Cortex-A510 achieve an ML uplift of 3x compared to Cortex-A55. The step-up in ML performance enhances the various AI and ML workloads and use cases required by DTVs and STBs.
Security is another area where the Armv9 CPUs have added significant features. With all the new experiences and features coming to DTVs and STBs, there is a growing risk of new threats and attacks. Security protections for all the content now arriving via streaming services will be especially important, with content becoming a huge differentiator in the wider Home market. Moreover, with video being operated on by more processing engines in the chipset, the big challenge is ensuring that all the content is kept secure across the entire platform. This means protecting the content, operating system (OS), applications and assets from attacks.
A range of new and existing security features have been built into the architecture of Armv9 CPUs. This means that our partners can achieve better value from software investment into security measures. Leading to a more standardized and scalable security solution that can address a diversity of security challenges on DTV and STB devices and across the broader Home ecosystem. Some of the new security features include:
The GPUs that are part of the Total Compute solutions for DTVs and STBs are Mali-G510 and Mali-G310, and for some premium devices, Mali-G710. Mali-G510 can be used as part of Performance Total Compute solutions for premium DTVs and STBs, whereas Mali-G310 is more appropriate for Total Compute solutions targeting mid-range and entry level DTVs and STBs.
Both GPUs inherit Command Stream Frontend (CSF), the redesigned and additional execution engine, and the redesigned texture unit from Mali-G710 to support premium-like experiences. However, Mali-G510 and Mali-G310 have also been built with key DTV and STB features and trends in mind, which are:
Mali-G510 provides formats for better HDR support, Arm Frame Buffer Compression (AFBC) uncompressed buffers and the new Arm Fixed Rate Compression (AFRC) for bandwidth reductions. Meanwhile, Mali-G310 offers formats for better HDR support and AFBC uncompressed buffers and has optional AFRC.
The introduction of our visually lossless fixed rate compression, AFRC, is important for DTV and STB devices, as it offers exceptional visual quality with high fixed compression rates. The new AFRC technology guarantees a bandwidth and memory footprint reduction, depending on the level of compression and type of content, at a minimum area cost. The use of AFRC framebuffers alone gives a peak of 60 percent reduction in bandwidth, while providing an 80 percent increase in peak performance. This, in turn, reduces the cost of the memory subsystem and DRAM, which then brings down the cost of the SoC itself.
As with the Armv9 CPUs, Mali-G510 provides a significant ML uplift of 100 percent compared to the Arm Mali-G57 GPU. This enables the variety of new ML and AI workloads that are taking place on DTV and STB devices, now and in the future.
The Total Compute solutions for DTVs and STBs are designed to address the variety of new, more enriching use-cases, experiences and compute workloads now taking place on these devices. Arm has different Total Compute solutions and configurations to address the compute needs of different tiers of DTV and STB devices across Premium, Mid and Entry-Level tiers. Whatever the tier of device, there will need to be greater processor power, advanced AI capabilities and enhanced security protections. All of these features are enabled by Arm’s range of Total Compute solutions for multi-functional DTVs and STBs.
To learn more about our Total Compute solutions, watch this new video about how they achieve accelerated performance growth across key compute workloads and use cases, and all consumer devices.
[CTAToken URL = "https://www.youtube.com/watch?v=hGXSmj2_g7g" target="_blank" text="Learn more about Total Compute" class ="green"]