March 29, 2024

Back in 2017 when Zen was first announced it was a big big thing. AMD was finally bringing a processor out that could actually compete with Intel and it’s high-performance parts head on rather than really on the value aspect alone. Last night at AMD’s New Horizon event marked what is possibly the company’s most significant announcement post the initial Ryzen reveal.

AMD’s Mark Papermaster spoke in detail about its new 7nm Zen 2  core and partnership with TMSC, which will be manufacturing it. From the outset, he claims a doubling of core density and halving of power consumption for the same performance, both potentially indicating that we’ll see increased core counts in future from its CPUs. He also claimed that Zen 2 will offer a 1.2x performance boost in IPC over current Zen+-based CPUs helped along by using a second generation Infinity Fabric.

AMD talked extensively about the forthcoming Zen 2 architecture. The goal of the original Zen architecture was to get AMD, at the very least, competitive with what Intel had to offer. AMD knew that Zen would not take the performance lead from Intel, but the pricing and features of its chips made them nonetheless attractive, especially in workloads that highlighted certain shortcomings of Intel’s parts (fewer memory channels, less I/O bandwidth). Zen 2 promises to be not merely competitive with Intel, but superior to it.

Intel’s long-delayed 10nm manufacturing process still has no specific launch date, with the company still using a 14nm process that’s been refined numerous times, stretching all the way back to its ‘Broadwell’ CPUs launched in 2014.

The original Zen used a multichip module design. Chips used one, two, or four dies (for Ryzen, first-generation Threadripper, and Epyc/second-generation Threadripper, respectively) all put together into a single package. Each die had two Core Complexes (blocks of four cores), two memory controllers, some Infinity Fabric links (for connections between dies), and some PCIe channels. This made it straightforward for AMD to scale from the single-die, 8-core/16-thread Ryzen up to the 32-core/64-thread Epyc. Zen 2 is taking a very different approach, albeit one that still uses a multichip design. Instead of having each die contain CPUs, memory controllers, and I/O, the new design splits up the different roles. There will be a single 14nm I/O die, with eight memory controllers, eight Infinity Fabric ports, and PCIe lanes, and then a number of 7nm “chiplets” containing only CPUs and Infinity Fabric. This new approach should remedy some of the more awkward aspects of the original Zen; for example, there is a significant latency overhead when a core on one Zen die has to use memory from another die. With the Zen 2 design, memory latency should become much more uniform.

Zen 2 will be the building block of its Ryzen and Ryzen Threadripper desktop CPUs as well as its EPYC server CPUs and is the most significant step in Zen’s timeline for a number of reasons. AMD needs to keep driving performance upwards to stay competitive and already made significant gains with Zen+, offering higher frequencies and improved boosting algorithms that offered tangible benefits, especially in lightly-threaded workloads. The move to 7nm should allow for higher frequencies as well as increased yields and this will impact all three tiers of its current processor line-up.