The quest to be the first foundry in advanced lithography is slowly being less and less prioritized by Samsung, as a new Exynos chipset roadmap reveals that the Korean giant will stick to its 2nm GAA node for a few years before finally taking a leap of faith towards 1.4nm. Thanks to this delay, TSMC will potentially beat Samsung as the first semiconductor manufacturer to introduce its sub-2nm process, but that’s no disadvantage.
Exynos 2900 rumored to be Samsung’s first 1.4nm SoC, gravitating from 2nm to give Samsung time to refine its yields
With TSMC expected to commence mass production of its 1.4nm process in 2028, Samsung will be an entire year behind its foundry nemesis, with a previous report stating that the commercialization of the Korean behemoth’s 1.4nm node will begin sometime in 2029. The Exynos 2600 represents Samsung’s foray into the 2nm chipset realm, with TSMC’s first family of SoCs on the same lithography arriving later this year.
The rumor from Kaulenda indicates that the Exynos 2700 and Exynos 2800 will remain on the 2nm process, with Samsung bringing gradual improvements to this node in performance, efficiency, and area. Based on the new claim, the Exynos 2900 will be Samsung’s first 1.4nm node. For the Exynos 2800, the third-generation 2nm GAA process, also known as SF2P+, will be utilized, marking what appears to be a strategic shift from Samsung.
Previously, the company would always embroil itself with TSMC in obtaining bragging rights by unveiling a newer lithography, and while that’s an ideal move for accumulating eyeballs briefly, it’s a terrible long-term approach, especially when Samsung has always been unable to overcome its yield problems. Now, it’s taking its sweet time tackling an issue that has long plagued the firm, with the company hoping that this decision paves the way towards profitability.
Stabilizing 2nm yields takes importance over being the first company to unveil the 1.4nm technology
We’ve reported previously that the Exynos 2800 won’t shift to the 1.4nm process, with Samsung preferring to stick with SF2P+ to boost its yields while investing time and resources to make sure the 1.4nm matures to a stage where it’s commercially viable to begin the Exynos 2900’s mass production.
With each TSMC 1.4nm wafer estimated to cost a mammoth $45,000 per wafer, Samsung’s LSI division, which is responsible for the Exynos brand, will be paying more or less the same dollar amount to pay its foundry segment. Keep in mind that Samsung won’t give preferential treatment or discounts to its own business arm, meaning that LSI will be paying more for each Exynos wafer.
The last thing Samsung wants is to rush to mass production on an immature node, costing the LSI division more per Exynos SoC fabricated. In short, it’s a smart business decision from Samsung and one that we hope proves fruitful in the coming years.
News Source: Kaulenda
About the author: Omar Sohail is a reporter and analyst for Wccftech's mobile section, specializing in the technology and business of the mobile industry. His expertise lies in the intricate hardware supply chain, covering developments in semiconductor manufacturing, chip lithography, and camera sensor technology.
Follow Wccftech on Google to get more of our news coverage in your feeds.
