NVIDIA rumored to switch to Intel for advanced packaging for next-gen AI GPUs, away from TSMC

NVIDIA has reportedly moved to Intel for advanced packaging of its next-gen AI GPUs to stop shortages, the company is bottlenecked by TSMC's CoWoS tech.

1 minute & 41 seconds read time

NVIDIA can't make enough of its AI GPUs to meet current demand, let alone the non-stop demand for AI computing power... with the company reportedly working with Intel using its advanced packaging services.

NVIDIA rumored to switch to Intel for advanced packaging for next-gen AI GPUs, away from TSMC 08

The news is coming from UDN, which reports that TSMC's (Taiwan Semiconductor Manufacturing Company) and its CoWoS advanced packaging production capacity is "insufficient" which has left NVIDIA turning to Intel. Intel will use its advanced packaging production for NVIDIA, providing them with around 5000 pieces per month.

Intel's help here for NVIDIA will allow the AI GPU giant to pump out, even more, AI GPUs, which have been strained in supply because of various factors, mostly in the advanced packaging supply chain. Once Intel begins advanced packaging production for NVIDIA, it's expected that NVIDIA's total advanced packaging production capacity will increase by a significant 10%.

TSMC still remains as the clear main advanced packaging supplier for NVIDIA, with the Taiwan giant "sprinting" to expand its advanced packaging production capacity. Monthly production capacity in Q1 2024 is expected to rise up to 50,000 pieces per month, up from 40,000 per month in December 2023, representing a chunky 25% increase in production capacity.

Intel will be using its Oregon and New Mexico facilities in the US for advanced packaging production, while Team Blue is "actively expanding" its advanced packaging at its new factory in Penang. CDN points out that Intel has previously said it has open arms for customers to only use its advanced packaging solutions, with its end goal of giving customers -- like NVIDIA in this case -- more production flexibility.

AI GPU chip shortages were previously because of three different factors: not only insufficient advanced packaging production capacity but also low volumes of new HBM3 memory production capacity, and some cloud service providers placed multiple orders. These factors have slowly been eliminated, and now production capacity can increase for NVIDIA.

Buy at Amazon

ASUS TUF GeForce RTX® 4090 OC Edition

TodayYesterday7 days ago30 days ago
Buy at Newegg
* Prices last scanned on 2/23/2024 at 5:19 am CST - prices may not be accurate, click links above for the latest price. We may earn an affiliate commission.
NEWS SOURCES:twitter.com, money.udn.com

Anthony joined the TweakTown team in 2010 and has since reviewed 100s of graphics cards. Anthony is a long time PC enthusiast with a passion of hate for games built around consoles. FPS gaming since the pre-Quake days, where you were insulted if you used a mouse to aim, he has been addicted to gaming and hardware ever since. Working in IT retail for 10 years gave him great experience with custom-built PCs. His addiction to GPU tech is unwavering and has recently taken a keen interest in artificial intelligence (AI) hardware.

Newsletter Subscription

Related Tags