The project named “Baltra”according to one of the Galapagos Islands, is an AI server chip that should be ready for mass production by 2026. Polaze hopes that the development of silicon chips to date and confidence in the development team, which has already proven its capabilities in designing state-of-the-art chips for iPhones and Mac computers, can create a chip that will be able to compete with established products. Apple also cooperates with Broadcom Inc. because their expertise in this area gives Apple more chances to quickly develop the chip.
AMD’s chiplet design will be used
The Baltra chip is expected to use TSMC’s advanced N3P manufacturing process, which is an improvement over the process used for Apple’s latest computer processors. This choice of manufacturing process is in line with industry trends, as other AI chip designers such as OpenAI and Nvidia are expected to use this process to improve performance. Apple plans to use Chiplet design for Baltra, a technique first developed by AMD.
Chiplet design is a modular approach to building processors and complex integrated circuits that addresses several limitations of traditional monolithic chip designs. In this approach, the chip is broken down into smaller, specialized components called chiplets, which are then interconnected within a single package to create a more versatile and efficient system.
Can it be scaled?
Apple’s decision to develop its own AI server chip was apparently driven by the need to scale AI services to support billions of devices, reduce costs associated with building and operating data centers, reduce dependence on third-party vendors such as Nvidia, which currently dominates the AI chip market, and improve privacy and security. of AI data processing.
They left the server market a long time ago
Historically, Apple has not been a major player in the server chip market. In 2011, the company stopped production of its Xserve line of servers and exited the hardware market for business servers.
Baltra’s development could have far-reaching consequences for Apple’s AI capabilities. This could enable more advanced AI features on the devices themselves and improve the performance of cloud AI services like Siri. Moreover, the move could position Apple as a more serious competitor in the AI space, potentially challenging the dominance of companies like Google, Microsoft and Amazon in cloud AI services.