AMD Zen 3 based processors which would be arriving next year and utilize TSMC’s next-generation 7nm+ process node could feature higher transistor density than Zen 2 that arrives this year. AMD’s Zen 3 architecture would be featured across a wide variety of Ryzen, Ryzen Threadripper and EPYC series processors, utilizing the advanced TSMC 7nm+ node that takes advanced of the EUV tech.
AMD Zen 3 Based CPUs Could Feature 20% More Transistors Than Zen 2 – Arrives in 2020
In a report published by PCGamesN, it is mentioned that AMD’s Zen 3 architecture is going to get a major transistor density boost thanks to the TSMC 7nm+ process node. Unlike the Zen 2 CPUs that utilize the TSMC 7nm node, the 7nm+ node utilizes the advanced EUV technology which would be ready for volume production in the second quarter of 2019, as reported by DigiTimes.
Taiwan Semiconductor Manufacturing Company (TSMC) is set to move 7nm EUV process technology to volume production in the second quarter of 2019, according to a recent Chinese-language Commercial Times report.
Currently, the TSMC N7+ node is occupied for the production of HiSilicon’s flagship mobile SOC and a custom version of the process, known as N7 Pro, would be utilized in the production of Apple’s A13 processor for their upcoming iPhones. It is likely that AMD would like to wait out and not jump on the 7nm EUV bandwagon this early considering they are in the process of launching their Zen 2 (TSMC 7nm) processor lineup but we can expect production on the new Zen 3 chips to start at the beginning of next year.
Now according to TSMC themselves, the 7nm+ process node allows an increase of 20% in overall transistor density while increasing power efficiency by 10%. AMD could take full advantage of the process node and we could see a density boost of up to 20% in the Zen 3 architecture while having 10% better power efficiency. These changes are in addition to the architecture revamp which one should expect from the Zen 3 CPUs which will be introduced on the Ryzen and EPYC platforms in 2020 and beyond.
AMD CPU Roadmap (2018-2020)
|Ryzen Family||Ryzen 1000 Series||Ryzen 2000 Series||Ryzen 3000 Series||Ryzen 4000 Series|
|Architecture||Zen (1)||Zen (1) / Zen+||Zen (2)||Zen (3)|
|Process Node||14nm||14nm / 12nm||7nm||7nm+|
|High End Server (SP3)||EPYC ‘Naples’||EPYC ‘Naples’||EPYC ‘Rome’||EPYC ‘Milan’|
|Max Server Cores / Threads||32/64||32/64||64/128||TBD|
|High End Desktop (TR4)||Ryzen Threadripper 1000 Series||Ryzen Threadripper 2000 Series||Ryzen Threadripper 3000 Series (Castle Peak)||Ryzen Threadripper 4000 Series|
|Max HEDT Cores / Threads||16/32||32/64||64/128?||TBD|
|Mainstream Desktop (AM4)||Ryzen 1000 Series (Summit Ridge)||Ryzen 2000 Series (Pinnacle Ridge)||Ryzen 3000 Series (Matisse)||Ryzen 4000 Series (Vermeer)|
|Max Mainstream Cores / Threads||8/16||8/16||16/32||TBD|
|Budget APU (AM4)||N/A||Ryzen 2000 Series (Raven Ridge)||Ryzen 3000 Series (Picasso) Zen+?||Ryzen 4000 Series (Renior)|
Details for Zen 2 are available but in an earlier interview, AMD’s CTO, Mark Papermaster unveiled what to expect from the Zen 3 CPU architecture. First and foremost, AMD states that their claimed performance and efficiency numbers are for a real product while TSMC uses their own methods. Hence, the difference between the power draw and performance uplift estimates from both companies. A node and an actual product are two varying things and you can’t compare them both.
“TSMC may have been measuring a basic device like a ring oscillator — our claims are for a real product,”
“Moore’s Law is slowing down, semiconductor nodes are more expensive, and we’re not getting the frequency lift we used to get,” he said in a talk during the launch, calling the 7-nm migration “
Looking ahead, a 7-nm-plus node using extreme ultraviolet lithography (EUV) will “primarily leverage efficiency with some modest device performance opportunities,”
Competition within the semiconductor manufacturing space is also heating up as a result with major players now investing billions in EUV technology. Samsung Foundry announced this week that they have completed the development of their own 5nm EUV process technology (5LPE) that would offer 10% better performance or improve power efficiency by 20% in the same load tier. It will also reduce the chip die area by 20% compared to the 7LPP process. The Samsung 5LP node will be going into volume production in 2020.
Intel, on the other hand, has already made a Billion dollar investment for their own 7nm EUV based technology. The investment would be used to expand their Oregon fab which would begin from June and take 18 months (+) to complete. The blue team’s 10nm process was delayed for a prolonged period which made competitors rise up to the ranks. Intel’s first 10nm processors are likely to arrive at the end of this year with their Xeon based Ice Lake-SP lineup scheduled for launch next year and are likely going to compete with AMD’s 7nm+ Zen 3 based EPYC Milan processors that would offer increased performance, core count and higher efficiency than their upcoming 7nm based EPYC Rome processors.
AMD is largely on work on their next-generation Zen 4 CPU architecture which is in-design and would replace the Zen 3 CPUs. While we wait for the next generation, AMD has already tapped some major wins with their 7nm based Ryzen, EPYC, and Navi products with Ryzen and Navi to feature inside Sony’s next-generation PlayStation while EPYC Rome would be powering some of the biggest and fastest supercomputers on the planet. AMD’s CEO, Lisa Su, is expected to make groundbreaking technology announcements at Computex 2019 next month as they announce their first consumer-focused 7nm products.