It is not the first time that AMD and NVIDIA publicly disagree about the death of Moore’s law, as the first manufacturer believes that it is still alive, but changing. Now, in an interview with the Barron’s portal, the CEO of AMD, Lisa Su, points out that the industry needs to take different paths to overcome the challenges of performance, efficiency and cost.
To put “Moore’s Law” in context, it all started in 1965 when Intel founder Gordon Moore made a prophecy in which he predicted that every 18 months the number of transistors on lithography chips in a new generation would increase by 100 % in relation to the previous generation, however, maintaining costs for the consumer.
AMD has been working on advancing 3D packaging and chiplet technology for years with its first HBM designs in 2015, chiplet processors in 2017, and also the first 3D packaging on a chip with its 3D V-Cache design in 2022—currently the focus of the company according to Su.
The manufacturer will launch its MI300, an exa-scale APU that combines multiple CPU, GPU and memory IPs into multiple chiplets and 3D matrices in a single package.
Lisa also mentioned that despite the increasing cost and decreasing performance benefits with each new generation, they have the prospect of entering the 2 nm range after completing work with 3 nm, which they are currently working.
AMD currently has products that utilize both 5nm and 4nm process nodes with some 6nm and 7nm products in between. Starting next year, the company is expected to introduce its first 3nm chips, which will likely be targeted at the server segment before the company introduces them to the broader consumer market.
