Samsung Is Using AI to Design Its Next Exynos Chipset for Smartphones
It is a time-consuming process to develop a smartphone chipset, not to mention the insane number of complexities accompanied with that journey. To make things a little less stressful, Samsung is reportedly using artificial intelligence to automate the steps it takes to develop the next Exynos chipset for smartphones.
Samsung Apparently Using AI Features From Synopsys to Develop This Exynos SoC
The leading chip design software company, Synopsys, is used by many firms and will be Samsung’s best friend in developing the next Exynos for smartphones. Aart de Geus, the chairman and co-CEO of Synopsys states the following about the next stage of SoC development.
“What you’re seeing here is the first of a real commercial processor design with AI.”
According to industry experts, Synopsys’ DSO.ai tool accelerates semiconductor development, and the company has years of semiconductor design experience that can be used to train its AI algorithm. Samsung has not confirmed if the recently launched flagship, the Galaxy Z Fold 3 uses an AI-designed chipset, but considering there is a Snapdragon 888 running underneath its hood, we have to take that as a no.
While the Korean giant’s spokesperson has said that Samsung is using Synopsys AI software to design its Exynos chipsets, the firm has not confirmed if any of those designs have gone into mass production or which future products, such as the heavily rumored Exynos 2200, will take advantage of this technology. Ars Technica reports that according to analyst Mike Demler, the use of AI is well suited for arranging billions of transistors across a chip.
“It lends itself to these problems that have gotten massively complex. It will just become a standard part of the computational tool kit.”
Demler also states that using artificial intelligence to design a chip is expensive, as a company requires a ton of cloud computing power to train a powerful algorithm. Fortunately, he expects costs to drop as the technology becomes more widespread and used by other firms. Drafting up a new chip design, such as for the Exynos 2200, is an arduous task, requiring weeks of planning as well as decades of experience to execute.
The inclusion of AI may not train the algorithm to possess the instinctive capabilities of the chip engineers, but some skills derived from years of experience can train the program to an extent. In addition, Synopsys claimed that using AI improved the performance of a chip by 15 percent, with Aart de Geus stating that one result using software can be achieved in a matter of weeks as opposed to a few months using qualified individuals.
“A bit over a year and a half ago, for the first time, we were able to get the same results as a team of experts would get in multiple months in just a few weeks.”
These advantages could prepare Samsung to catch up to Apple in both performance and power efficiency, and possibly the same results could be achieved when Exynos variants of chipsets are designed for future notebook computers. However, we will see how Samsung employs this technology in upcoming SoC iterations, and we will have all the updates for you in the future, so stay tuned.
News Source: Wired