Tech News
Back to home
Technology

Nvidia tipped to be TSMC's first 16A customer, ahead of Apple — Feynman GPUs could make full use of GAA transistors and backside power

Source

Tom's Hardware

Published

TL;DR

AI Generated

Nvidia is set to be TSMC's first customer for the A16 node, utilizing GAA transistors and backside power delivery in its upcoming Feynman GPUs, designed for AI applications. This move breaks Apple's tradition of being the first to adopt new TSMC process nodes. A16's backside power delivery allows for improved performance and architectural freedom, potentially giving Nvidia a competitive edge in AI and HPC markets. By skipping the N2 process and going straight to A16, Nvidia aims to lead in advanced silicon, though at a higher cost. This strategic shift positions Nvidia as a frontrunner in AI chip innovation and could influence the industry's future roadmap decisions.

Read Full Article

Similar Articles

Framework's new RTX 5070 12GB graphics module costs a whopping $1,199 — 72% more expensive than $699 8GB version, says pricing is beyond its control

Framework's new RTX 5070 12GB graphics module costs a whopping $1,199 — 72% more expensive than $699 8GB version, says pricing is beyond its control

Nvidia released a new 12GB version of the RTX 5070 mobile GPU with upgraded memory chips, increasing memory throughput. Framework introduced a new graphics module for its Framework Laptop 16 featuring this GPU, priced at $1,199, a significant increase from the $699 8GB version. The high cost is attributed to the expensive GDDR7 memory and the ongoing global memory shortage. Framework clarified that the pricing is influenced by external factors and not within its control, highlighting the challenges faced by consumers due to the current market conditions.

Tom's Hardware•
More details emerge about how Intel now earns more revenue from each wafer by looking to the edges — analyst reports say reduced yield variability across each wafer leads to more sellable CPUs

More details emerge about how Intel now earns more revenue from each wafer by looking to the edges — analyst reports say reduced yield variability across each wafer leads to more sellable CPUs

Intel has seen improved revenue per wafer by reducing yield variability across each wafer, resulting in more sellable CPUs. The company's focus on tightening yield distribution across the wafer edges has led to increased margins and productivity. By implementing edge-specific process correction methods, Intel can extract more high-quality and sellable dies from a single wafer. These improvements are node-independent and have been attributed to disciplined execution improvements under new manufacturing leadership. Intel's efforts have led to better output and demand for CPUs, with even lower-quality chips now being sold as viable products.

Tom's Hardware•
Nvidia quietly launches 12GB RTX 5070 laptop GPU — midrange mobile gaming gets more VRAM amid the RAMpocalypse

Nvidia quietly launches 12GB RTX 5070 laptop GPU — midrange mobile gaming gets more VRAM amid the RAMpocalypse

Nvidia has quietly released a new 12GB RTX 5070 laptop GPU, offering 50% more memory capacity than the existing 8GB version. This new option aims to address the ongoing memory shortage crisis, providing gamers and creators with more VRAM for improved performance. The 12GB RTX 5070 will sit between the 8GB and Ti 12GB versions in Nvidia's mobile GPU lineup, offering enhanced capabilities for AAA gaming and higher graphical settings. This move marks a rare instance of Nvidia offering higher VRAM options amidst industry trends of reducing memory capacities on consumer hardware.

Tom's Hardware•
Nvidia RTX 3060 comeback in 2026 could alleviate soaring GPU prices and memory shortages — rumored RTX 5050 9GB abruptly shelved amid speculation

Nvidia RTX 3060 comeback in 2026 could alleviate soaring GPU prices and memory shortages — rumored RTX 5050 9GB abruptly shelved amid speculation

Nvidia is rumored to be reintroducing the RTX 3060 graphics card in June 2026 to address GPU price hikes and memory shortages. The company has reportedly paused the launch of the RTX 5050 9GB variant, potentially due to these market challenges. The RTX 3060, with its 12GB of GDDR6 memory, could offer a cost-effective solution for gamers, especially with the current scarcity of GPUs. The decision to bring back the RTX 3060 may be influenced by the lower cost of GDDR6 VRAM and the ease of manufacturing using Samsung's 8nm process.

Tom's Hardware•

We use cookies

We use cookies to ensure you get the best experience on our website. For more information on how we use cookies, please see our cookie policy.