Apple M5 Chip's Dual-Use Design Will Power Future Macs and AI Servers
Apple plans to use advanced SoIC packaging for M5 chips, enhancing performance for Macs, data centers, and AI tools. Collaboration with TSMC includes carbon fiber composite molding for improved thermal management. Mass production expected in 2025-2026, aligning with Apple's vertical integration strategy.
Read original article
Apple is set to utilize a more advanced SoIC packaging technology for its upcoming M5 chips, aiming to meet the demands of powering consumer Macs and enhancing data centers and AI tools. The System on Integrated Chip technology allows for stacking chips in a three-dimensional structure, offering improved electrical performance and thermal management compared to traditional designs. Apple's collaboration with TSMC on a hybrid SoIC package incorporating carbon fiber composite molding technology is in a trial production phase, with plans for mass production in 2025-2026. References to the M5 chip have been found in Apple's code, indicating its development for AI servers. The M5's dual-use design is seen as a strategic move by Apple to integrate its supply chain for AI functionality across various platforms. This development aligns with Apple's goal of vertically integrating its hardware and software for future computing needs.
Related
TSMC experimenting with rectangular wafers vs. round for more chips per wafer
TSMC is developing an advanced chip packaging method to address AI-driven demand for computing power. Intel and Samsung are also exploring similar approaches to boost semiconductor capabilities amid the AI boom.
US Mac market share grew 22% in Q1, likely helped by new MacBook Air models
The US Mac market share grew by 22.4% in Q1 2024, surpassing the overall PC market growth of 5%. New MacBook Air models drove the surge, with an 8% growth forecasted due to enterprise sales and AI demand. Apple holds a 14.2% share in the US PC market, expected to increase with new Mac features and AI-capable PCs.
Intel CPU with Optical OCI Chiplet Demoed with 4Tbps of Bandwidth and 100M Reach
Intel unveiled a new CPU with an Optical Compute Interconnect chiplet, offering 4Tbps bandwidth and 100m reach. The chiplet merges Photonic and Electronic Integrated Circuits, featuring built-in lasers for improved server communication, especially in AI servers. Intel plans to scale the technology up to 32Tbps, focusing on power savings and enhanced bandwidth. This advancement aligns with the industry's move towards co-packaged optics for better efficiency and performance, marking a significant shift in server architecture towards power efficiency and high-speed data transfer.
Extreme Measures Needed to Scale Chips
The July 2024 IEEE Spectrum issue discusses scaling compute power for AI, exploring solutions like EUV lithography, linear accelerators, and chip stacking. Industry innovates to overcome challenges and inspire talent.
Apple just launched a public demo of its '4M' AI model
Apple publicly launches its '4M' AI model with EPFL on Hugging Face Spaces, showcasing versatile capabilities across modalities. The move signals a shift towards transparency, aligning with market growth and emphasizing user privacy amid ethical concerns.
14 commentsThe stacked logic setup is far more interesting in terms of what it means for die density and efficiency.
It would be of interest because normally you'd be optimizing for very different characteristics.
They have been severely mismanaged for a decade by forcing mandatory yearly upgrades across all of their stacks.
In the past several years the cracks have started to show in their software (bugs lingering for years, features being delayed for months after yearly OS upgrades etc.). Hardware teams seemed to be mostly immune from it, and delivered insane results.
Now it's "omg must have AI don't matter if you're ready for it or no. Upper management saw LLM demos and are now forcing it everywhere. What do you mean you had other plans for M5? It's AI now damn it"
I mean, sure, it’s similar to what you’d surely get running Terminal.app, but I mean I doubt Apple engineers are doing some internal equivalent of apt when they need something.
There’s brew, but it would be… awkward if that’s what they used in their private cloud fleet.
Related
TSMC experimenting with rectangular wafers vs. round for more chips per wafer
TSMC is developing an advanced chip packaging method to address AI-driven demand for computing power. Intel and Samsung are also exploring similar approaches to boost semiconductor capabilities amid the AI boom.
US Mac market share grew 22% in Q1, likely helped by new MacBook Air models
The US Mac market share grew by 22.4% in Q1 2024, surpassing the overall PC market growth of 5%. New MacBook Air models drove the surge, with an 8% growth forecasted due to enterprise sales and AI demand. Apple holds a 14.2% share in the US PC market, expected to increase with new Mac features and AI-capable PCs.
Intel CPU with Optical OCI Chiplet Demoed with 4Tbps of Bandwidth and 100M Reach
Intel unveiled a new CPU with an Optical Compute Interconnect chiplet, offering 4Tbps bandwidth and 100m reach. The chiplet merges Photonic and Electronic Integrated Circuits, featuring built-in lasers for improved server communication, especially in AI servers. Intel plans to scale the technology up to 32Tbps, focusing on power savings and enhanced bandwidth. This advancement aligns with the industry's move towards co-packaged optics for better efficiency and performance, marking a significant shift in server architecture towards power efficiency and high-speed data transfer.
Extreme Measures Needed to Scale Chips
The July 2024 IEEE Spectrum issue discusses scaling compute power for AI, exploring solutions like EUV lithography, linear accelerators, and chip stacking. Industry innovates to overcome challenges and inspire talent.
Apple just launched a public demo of its '4M' AI model
Apple publicly launches its '4M' AI model with EPFL on Hugging Face Spaces, showcasing versatile capabilities across modalities. The move signals a shift towards transparency, aligning with market growth and emphasizing user privacy amid ethical concerns.