At COMPUTEX 2025, ASUS showcased its comprehensive AI infrastructure solutions, targeting the computing needs of data centers, edge computing, and a wide range of intelligent applications. With a robust product portfolio, ASUS brings to life the NVIDIA AI Factory vision, illustrating how AI will be deeply embedded into enterprise IT infrastructure. This evolution continues to drive forward applications and services that shape the future of AI and high-performance computing

Linxens, a global leader in component-based solutions for authentication and traceability, and Dracula Technologies, a pioneer in ambient light energy harvesting, have announced a strategic collaboration to develop next-generation autonomous IoT solutions for traceability and smart labelling applications. This collaboration paves the way for smarter and more sustainable connected devices

Technology news, product launches and show highlights from the event attendees available for download (PDF).

As AI workloads scale across global cloud infrastructure, chip designers face a persistent challenge: balancing compute throughput, bandwidth, and power efficiency amid escalating design complexity.Emerging architectures built around UCIe-based chiplets, PCI Express 7.0 and CXL 3.1 interconnects, and 2.5D/3D memory-integrated packaging have made integration-not computation-the new bottleneck.Progate Group Corporation (PGC, TPEx: 8227), a member of TSMC DCA, leverages its long-term partnership with Synopsys to solve that bottleneck.Through a unified ASIC turnkey flow, PGC enables innovators to design, verify, and manufacture advanced-node SoCs and chiplet-based accelerators for cloud and AI applications with greater efficiency and lower risk.PGC integrates the latest generation of Synopsys Interface IP, spanning high-speed interconnects, memory, and networking subsystems critical to modern cloud and AI processors.The portfolio includes UCIe 40G Die-to-Die, CXL 3.x, and PCI Express 7.0 controllers that enable seamless chiplet-based compute integration for next-generation server processors.It also incorporates HBM4 and DDR5 Gen2 memory subsystems optimized for multi-stack 2.5D and 3D integration, as well as 224G Ultra Ethernet PHY solutions designed for high-throughput data-center networking.Validated across TSMC 3nm to 2nm process nodes, these silicon-proven IPs address the cloud industry's most demanding bandwidth and power-efficiency challenges while maintaining full compatibility with Synopsys EDA and PDK environments.Through its TSMC Design Center Alliance (DCA) membership, PGC provides a verified turnkey service covering RTL design, APR layout, wafer fabrication, 2.5D/3D packaging, and final test coordination.By integrating Synopsys EDA tools (Design Compiler, Fusion Compiler, ICC2, PrimeTime) and complete IP deliverables (Lib, LEF, GDS, and testbench models), PGC helps customers accelerate RTL-to-signoff cycles through pre-verified IP subsystems, SIPI (signal and power integrity) analysis, and IP hardening support-as described in Synopsys' IP Accelerated methodology.This unified approach reduces integration risk, minimiclozes manual iteration, and ensures design closure consistency across advanced-node ASIC projects."Cloud and AI systems are evolving into distributed compute platforms built on multiple dies and protocol layers," said Fred Lai, CEO of PGC."By aligning Synopsys' IP Accelerated methodology with TSMC's leading-edge processes, PGC helps customers scale performance while maintaining design predictability and manufacturing reliability."Unlike large turnkey providers that concentrate solely on a handful of mega-scale clients, PGC combines advanced-node design capability with engineering agility, making it a trusted partner for both hyperscalers and rapidly growing AI startups.Beyond simplifying multi-vendor coordination, the company maintains unified technical accountability across design, IP integration, and manufacturing, ensuring that every stage of development proceeds seamlessly.PGC's engineers perform SIPI co-optimization using Synopsys EDA and PDK models to guarantee signal and power stability, while TSMC-certified signoff flows deliver predictable yield and reliability.Supported by cross-regional teams in Taiwan, Japan, China, and the United States, PGC provides rapid collaboration and execution across the entire ASIC turnkey cycle.PGC's ecosystem extends from design integration to manufacturing, combining TSMC's advanced-node foundry services, Synopsys' comprehensive IP portfolio, and partnerships with leading OSAT partners for 2.5D/3D packaging enablement.Together, this network provides a complete foundation for high-bandwidth, energy-efficient ASIC innovation across AI, HPC, and cloud computing markets.

Global Unichip Corp. (GUC), the Advanced ASIC leader, and Ayar Labs, a leader in co-packaged optics (CPO) for large-scale AI workloads, have announced a strategic partnership to integrate CPO into GUC's advanced ASIC design services.This collaboration paves the way for high-bandwidth, low-latency, power-efficient optical interconnects in next-generation AI, HPC and networking applications where electrical signaling is reaching its limits. By integrating Ayar Labs'TeraPH optical engines into GUC's advanced packaging and ASIC workflow, the companies are exploring critical areas to enable future CPO deployment."The CPO revolution is at our doorstep. Integrating Ayar Labs'optical engines into our advanced packaging flows is a critical step," said Igor Elkanovich, CTO of GUC. "Our new joint design allows us to address the challenges of CPO integration - architectural, power and signal integrity, mechanical and thermal - ensuring our future customers have access to a robust, high-bandwidth and power-efficient solution."The new XPU multi-chip package (MCP) design replaces traditional electrical interconnects with Ayar Labs'optical engines attached directly to the MCP organic substrate. This architecture enables more than 100 Tbps full-duplex optical interface from the XPU package, more than an order of magnitude improvement over current XPUs.UCIe-S (64 Gbps) provides bandwidth between the optical engines and I/O chiplets over the MCP substrate, while UCIe-A (64 Gbps) is used for communication between the I/O chiplet to the main AI die over local silicon interconnect (LSI) bridges. The design also addresses signal and power integrity challenges associated with the large-scale package. Thermal optimization is done at the XPU MCP level, and a new stiffener design incorporates optical engines' requirements and enables detachable fiber connections while meeting mechanical stress and warpage requirements."The future of AI and data center scale-up will not be possible without optics to overcome the electrical I/O bottleneck," said Vladimir Stojanovic, CTO and co-founder of Ayar Labs. "Working with GUC on advanced packaging and silicon technologies is an important step in demonstrating how our optical engines can accelerate the implementation of co-packaged optics for hyperscalers and AI scale-up."GUC will share more about these technology advancements during a presentation titled "Advanced Packaging Technologies for Modular and Powerful Compute" at the 2025 TSMC Open Innovation Platform (OIP) Ecosystem Forum in Hsinchu, Taiwan, on Tuesday, Nov. 18, 2025.For more information about GUC.For more information about Ayar Labs.

As Taiwan faces demographic challenges from a declining birthrate, the food and beverage (F&B) industry is struggling with severe labor shortages. While digital technologies and automation offer solutions for improving operational efficiency and customer satisfaction, many small and medium-sized eateries - such as breakfast shops and family-run restaurants - still rely on manual processes due to limited budgets. This manual process is slow, error-prone, and struggles to meet rising consumer expectations, making it difficult for small businesses to compete with major chains.DotDot Global addresses these challenges with its one-stop F&B technology solution, integrating online ordering, self-checkout, AI queue management, multiple payment options, membership management, and data analytics. This solution enables restaurants to embrace mobile payment trends, streamline workflows, reduce staffing needs, and enhance customer experiences. To date, DotDot Global has successfully implemented its system in more than 15,000 restaurants across Taiwan.According to Ting Hsieh, CEO of DotDot Global, the company's AI Store Manager replaces traditional manual ordering and checkout procedures, enabling restaurants to achieve front-of-house automation. The AI Store Manager includes intelligent ordering, AI-powered queue calling, and Taiwan's first unified cross-domain payment platform that integrates "points, cash, and coupons." It allows flexible use of loyalty points and discounts while seamlessly connecting credit cards, cash, and digital payments. As a result, restaurants can operate efficiently with minimal staffing - sometimes even with just one person running the entire store.Leveraging AWS cloud services to accelerate innovationDotDot Global's solutions are powered by Amazon Web Services (AWS) and built on AWS IoT Core, which provides two-way communication for real-time monitoring of connected devices worldwide. Through Over-the-Air (OTA) remote updates, the company can upgrade system functions or apply security patches to all devices across Taiwan in under 30 minutes - without dispatching engineers. This significantly reduces maintenance costs while ensuring that every client location has immediate access to the latest features, helping restaurants maintain a competitive edge in the fast-evolving F&B technology landscape.Hsieh explained that since entering the F&B technology space, DotDot Global has continuously gathered feedback from restaurant operators to guide product development. The newly launched AI Store Manager integrates multiple AI applications such as voice queue announcements, on-screen notifications, and adaptive prompt customization based on call frequency or style. Additionally, the system can analyze customer behavior and restaurant sales data using AI, providing insights for menu optimization and combo meal recommendations to improve profitability.Building a cross-domain ecosystem with the Unified Payment Interface (UPI)As mobile payment tools continue to diversify, DotDot Global introduced its Unified Payment Interface (UPI) to foster a new, cross-domain payment ecosystem. With a single integration, business partners can support all payment scenarios - from online to offline, cash to digital. DotDot Global is collaborating with POS system providers, vending machine operators, self-checkout solutions, currency exchange kiosks, and claw machine vendors to expand the ecosystem, generating new traffic and value for its partners.Hsieh added that through participation in the Startup Terrace Kaohsiung AWS Joint Innovation Center (JIC) project, DotDot Global has strengthened collaboration with local industries in Kaohsiung, including parking operators, vending businesses, and F&B brands. The company has also integrated with Agrigo, a cloud-based agricultural e-commerce platform, to enable innovative supply chain payment scenarios. Furthermore, DotDot Global is actively promoting its AI Store Manager and smart restaurant solutions in southern Taiwan, with notable adoption by the well-known Kaohsiung eatery Lao Er Pork Hock on Rice.