Insights
MediaTek announced a collaboration with Meta to develop its next-generation smart glasses chip. Since Meta has previously used Qualcomm chips for its two generations of smart glasses products, it is speculated that Meta’s expansion of chip suppliers is aimed at maintaining supply chain flexibility and reducing costs. MediaTek, in turn, is poised to leverage smart glasses to tap into opportunities within Meta’s VR/AR devices.
TrendForce’s Insights:
In mid-November 2023, MediaTek hosted the overseas summit, Mediatek Executive Summit 2023, where it announced a collaboration with Meta to develop the next-generation smart glasses chip.
Meta’s first smart glasses, a collaborative creation with Ray-Ban in 2021, differ from the Quest series as they are not high-end VR devices but rather feature a simpler design, focusing on additional functionalities like music playback and phone calls.
In the fall of 2023, Meta introduced a successor product with significant improvements in camera resolution, video quality, microphones, and internal storage. This new device is designed to simplify the recording and live streaming process by integrating with Meta’s social platform. Additionally, the new product aligns with the trend of generative AI and incorporates Meta’s AI voice assistant based on Llama2 LLM.
Notably, the market has shown keen interest and discussion regarding MediaTek’s announcement on the collaboration with Meta, given that Meta’s previous two generations of smart glasses used Qualcomm chips, specifically the Qualcomm Snapdragon Wear 4100 for the older version and the AR1 Gen1 for the new version.
Analysis of Meta’s Motivation: Meta’s decision to collaborate with MediaTek may be driven by considerations of risk diversification among suppliers and overall cost reduction.
Firstly, Meta has been investing in the development of in-house chips in recent years to ensure flexibility in product development. Examples include the MTIA chip, disclosed in mid-2023, designed for processing inference-related tasks, and the MSVP, the first in-house ASIC chip for video transcoding, which is expected to be used in VR and AR devices.
Given Meta’s previous attempts, including collaboration with Samsung, to independently develop chips and move towards chip autonomy, the partnership with MediaTek can be seen as a risk mitigation strategy against vendor lock-in.
Secondly, considering that smart glasses, unlike the high-priced Quest series, are currently priced at USD 299 for both models, MediaTek’s competitive pricing may also be a significant factor in Meta’s decision to collaborate with them.
From MediaTek’s perspective, their focus extends beyond smart glasses to the vast business opportunities presented by Meta’s VR and AR devices. In reality, examining Meta’s smart glasses alone reveals estimated shipments of around 300,000 pairs for the older model. Even with the new model and the anticipated successor expected to launch in 2025, there is currently no clear indication of significant market momentum.
In practical terms, this collaboration with Meta might not contribute substantially to MediaTek’s revenue. The crucial aspect of MediaTek’s collaboration with Meta lies in strategically positioning itself in Meta’s smart headwear supply chain, challenging the dominance of the original chip supplier, Qualcomm.
Looking at global VR device shipments, Meta is projected to hold over 70% market share in 2023 and 2024. There are also reports of an updated version of the Quest device expected to be available in China in late 2024. If MediaTek can expand its collaboration with Meta further, coupled with the gradual increase in the penetration rate of VR and AR devices, significant business opportunities still lie ahead.
From an overall perspective of the VR and AR industry, the current design of headwear devices no longer resembles the early models that required external computing cores due to considerations of cost, power, and heat.
The prevalent mainstream designs are now standalone devices. Given that these devices not only execute the primary application functions but also handle and consolidate a substantial amount of data from sensors to support functions like object tracking and image recognition, VR and AR devices require high-performance chips or embedded auxiliary SoCs. This market demand and profit potential are compelling enough to attract chip manufacturers, especially in the face of the gradual decline in momentum in the consumer electronics market, such as smartphones.
The VR and AR market still holds development potential, making it a strategic entry point for manufacturers. This insight is evident in MediaTek’s motivation, continuing its market cultivation efforts after developing the first VR chip for Sony PS VR2 in 2022 and collaborating with Meta.
News
According to TechNews’ report, during a recent financial conference, Samsung revealed its plans to diversify its sales structure by expanding its clientele in the fields of artificial intelligence semiconductors and automotive, moving away from its previous heavy reliance on the mobile sector.
As of 2023, it is understood that Samsung’s foundry sales distribution includes 54% from mobile, 19% from high-performance computing, and 11% from automotive.
According to a report from Wccftech, senior executives at Samsung have indicated that major players such as super-scale data centers, automotive original equipment manufacturers (OEMs), and other clients have been in contact with Samsung, considering the adoption of Samsung’s foundry services to manufacture their designed chips.
This includes the in-development 4-nanometer artificial intelligence accelerator and the 5-nanometer chips for the top-ranked electric vehicle company. Currently, Samsung is gearing up with its advanced packaging solution called SAINT (Samsung Advanced Interconnection Technology), aiming to compete with TSMC’s advanced packaging, CoWoS. Based on information disclosed by Samsung, there might be a collaboration with AMD in the field of artificial intelligence, involving the manufacturing of certain chips.
In fact, recent rumors suggest that Samsung has already reached an agreement with AMD to provide HBM3 and packaging technology for the upcoming Instinct MI300 series. Additionally, AMD might adopt a dual-sourcing strategy for the Zen 5 series architecture, choosing TSMC’s 3-nanometer process and Samsung’s 4-nanometer process technology for manufacturing the next-generation chips.
According to sources, besides the artificial intelligence domain, Samsung is likely to have received orders from the electric vehicle giant Tesla. The speculation points towards the possibility of fulfilling orders for Tesla’s next-generation HW 5.0 chip, designed for fully autonomous driving applications.
Read more
(Photo credit: Samsung)
News
In the global landscape of self-developed chips, the industry has predominantly embraced the Arm architecture for IC design. However, Meta’s decision to employ the RISC-V architecture in its self-developed AI chip has become a topic of widespread discussion. It is said the growing preference for RISC-V is attributed to three key advantages including low power consumption, high openness, and relatively lower development costs, according to reports from UDN News.
Noted that Meta exclusively deploys its in-house AI chip, “MTIA,” within its data centers to expedite AI computation and inference. In this highly tailored setting, this choice ensures not only robust computational capabilities but also the potential for low power consumption, with an anticipated power usage of under 25W per RISC-V core. By strategically combining the RISC-V architecture with GPU accelerators or Arm architecture, Meta aims to achieve an overall reduction in power consumption while boosting computing power simultaneously.
Meta’s confirmation of adopting RISC-V architecture form Andes Technology Corporation, a CPU IP and Platform IP supplier from Taiwan, for AI chip development underscores RISC-V’s capability to support high-speed computational tasks and its suitability for integration into advanced manufacturing processes. This move positions RISC-V architecture to potentially make significant inroads into the AI computing market, and stands as the third computing architecture opportunity, joining the ranks of x86 and Arm architectures.
Regarding the development potential of different chip architectures in the AI chip market, TrendForce points out that in the current overall AI market, GPUs (such as NVIDIA, AMD, etc.) still dominate, followed by Arm architecture. This includes major data centers, with active investments from NVIDIA, CSPs, and others in the Arm architecture field. RISC, on the other hand, represents another niche market, targeting the open-source AI market or enterprise niche applications.
(Image: Meta)
News
On the 15th, Microsoft introducing its first in-house AI chip, “Maia.” This move signifies the entry of the world’s second-largest cloud service provider (CSP) into the domain of self-developed AI chips. Concurrently, Microsoft introduced the cloud computing processor “Cobalt,” set to be deployed alongside Maia in selected Microsoft data centers early next year. Both cutting-edge chips are produced using TSMC’s advanced 5nm process, as reported by UDN News.
Amidst the global AI fervor, the trend of CSPs developing their own AI chips has gained momentum. Key players like Amazon, Google, and Meta have already ventured into this territory. Microsoft, positioned as the second-largest CSP globally, joined the league on the 15th, unveiling its inaugural self-developed AI chip, Maia, at the annual Ignite developer conference.
These AI chips developed by CSPs are not intended for external sale; rather, they are exclusively reserved for in-house use. However, given the commanding presence of the top four CSPs in the global market, a significant business opportunity unfolds. Market analysts anticipate that, with the exception of Google—aligned with Samsung for chip production—other major CSPs will likely turn to TSMC for the production of their AI self-developed chips.
TSMC maintains its consistent policy of not commenting on specific customer products and order details.
TSMC’s recent earnings call disclosed that 5nm process shipments constituted 37% of Q3 shipments this year, making the most substantial contribution. Having first 5nm plant mass production in 2020, TSMC has introduced various technologies such as N4, N4P, N4X, and N5A in recent years, continually reinforcing its 5nm family capabilities.
Maia is tailored for processing extensive language models. According to Microsoft, it initially serves the company’s services such as $30 per month AI assistant, “Copilot,” which offers Azure cloud customers a customizable alternative to Nvidia chips.
Borkar, Corporate VP, Azure Hardware Systems & Infrastructure at Microsoft, revealed that Microsoft has been testing the Maia chip in Bing search engine and Office AI products. Notably, Microsoft has been relying on Nvidia chips for training GPT models in collaboration with OpenAI, and Maia is currently undergoing testing.
Gulia, Executive VP of Microsoft Cloud and AI Group, emphasized that starting next year, Microsoft customers using Bing, Microsoft 365, and Azure OpenAI services will witness the performance capabilities of Maia.
While actively advancing its in-house AI chip development, Microsoft underscores its commitment to offering cloud services to Azure customers utilizing the latest flagship chips from Nvidia and AMD, sustaining existing collaborations.
Regarding the cloud computing processor Cobalt, adopting the Arm architecture with 128 core chip, it boasts capabilities comparable to Intel and AMD. Developed with chip designs from devices like smartphones for enhanced energy efficiency, Cobalt aims to challenge major cloud competitors, including Amazon.
(Image: Microsoft)
Insights
On October 17th, the U.S. Department of Commerce announced an expansion of export control, tightening further restrictions. In addition to the previously restricted products like NVIDIA A100, H100, and AMD MI200 series, the updated measures now include a broader range, encompassing NVIDA A800, H800, L40S, L40, L42, AMD MI300 series, Intel Gaudi 2/3, and more, hindering their import into China. This move is expected to hasten the adoption of domestically developed chips by Chinese communications service providers (CSPs).
TrendForce’s Insights:
In terms of the in-house chip development strategy of Chinese CSPs, Baidu announced the completion of tape out for the first generation Kunlun Chip in 2019, utilizing the XPU. It entered mass production in early 2020, with the second generation in production by 2021, boasting a 2-3 times performance improvement. The third generation is expected to be released in 2024. Aside from independent R&D, Baidu has invested in related companies like Nebula-Matrix, Phytium, Smartnvy, and. In March 2021, Baidu also established Kunlunxin through the split of its AI chip business.
Alibaba, in April 2018, fully acquired Chinese CPU IP supplier C-Sky and established T-head semiconductor in September of the same year. Their first self-developed chip, Hanguang 800, was launched in September 2020. Alibaba also invested in Chinese memory giant CXMT, AI IC design companies Vastaitech, Cambricon and others.
Tencent initially adopted an investment strategy, investing in Chinese AI chip company Enflame Tech in 2018. In 2020, it established Tencent Cloud and Smart Industries Group(CSIG), focusing on IC design and R&D. In November 2021, Tencent introduced AI inference chip Zixiao, utilizing 2.5D packaging for image and video processing, natural language processing, and search recommendation.
Huawei’s Hisilicon unveiled Ascend 910 in August 2019, accompanied by the AI open-source computing framework MindSpore. However, due to being included in the U.S. entity list, Ascend 910 faced production restrictions. In August 2023, iFLYTEK, a Chinese tech company, jointly introduced the “StarDesk AI Workstation” with Huawei, featuring the new AI chip Ascend 910B. This is likely manufactured using SMIC’s N+2 process, signifying Huawei’s return to self-developed AI chips.
Huawei’s AI chips are not solely for internal use but are also sold to other Chinese companies. Baidu reportedly ordered 1,600 Ascend 910B chips from Huawei in August, valued at approximately 450 million RMB, to be used in 200 Baidu data center servers. The delivery is expected to be completed by the end of 2023, with over 60% of orders delivered as of October. This indicates Huawei’s capability to sell AI chips to other Chinese companies.
Huawei’s Ascend 910B, expected to be released in the second half of 2024, boasts hardware figures comparable to NVIDIA A800. According to tests conducted by Chinese companies, its performance is around 80% of A800. However, in terms of software ecosystem, Huawei still faces a significant gap compared to NVIDIA.
Overall, using Ascend 910B for AI training may be less efficient than A800. Yet with the tightening U.S. policies, Chinese companies are compelled to turn to Ascend 910B. As user adoption increases, Huawei’s ecosystem is expected to improve gradually, leading more Chinese companies to adopt its AI chips. Nevertheless, this will be a protracted process.