Why you can trust Tom's Hardware Our expert reviewers spend hours testing and comparing products and services so you can choose the best for you. Find out more about how we test.
The fourth-gen EPYC 9003-Series Genoa processors enable unmatched compute density with up to 96 cores and 192 threads in a single socket, a bar that Intel simply won’t be able to compete with any time soon. For now, Genoa's once-unconceivable 192 cores and 384 threads in a single server lead the market by a vast margin, but AMD's own Bergamo will soon slot into the same systems for a total of up to 256 density-optimized Zen 4c cores and 512 threads in a single server.
In the meantime, the Genoa chips also bring a host of other improvements beyond core counts that will provide tangible performance gains across the entire stack. Those advances will be important as Intel looks to leverage its incumbency and pricing power to maintain its high-volume stronghold in the mid-range and lower end of the market.
We are in the earliest stages of transitioning to a new memory, a nearly once-in-a-decade event, so DDR5 pricing remains steep. However, data center users will benefit more from faster memory than we see in the consumer market. They’re also not as price sensitive if the new tech provides significant performance gains. As such, we expect that Genoa will enjoy rapid uptake in memory-constrained and compute-intensive workloads. Meanwhile, the EPYC Milan processors will co-exist in the market with Genoa for some time, providing a less-expensive pathway for lower-intensity applications.
It’s easy to point to the explosive performance gains and market-leading threaded horsepower of the 96-core EPYC Genoa 9654 — this chip simply has no peer in the x86 market in highly parallelized workloads. However, we also witnessed an amazing generational performance uplift with the 64-core EPYC 9554, and the frequency-optimized 32-core EPYC 9374F was simply stellar, providing an incredible leap forward for the mid-range.
It's always hard (and somewhat misleading) to compare pricing for server chips because they typically represent a small part of the overall server BOM, even before considering software licensing costs. In fact, AMD's largest and most important customers will never see those price tags. Regardless of the correlation of the suggested pricing to reality, AMD has raised pricing gen-on-gen, an inevitable result of the current market dynamics. Still, these hikes largely feel acceptable given the increased performance.
AMD's support for PCIe 5.0 and CXL will help foster a broader ecosystem of supporting devices, but Intel's Sapphire Rapids will also feature much of that same technology. For now, the jury is out on Sapphire Rapids and, more importantly, how competitive it will be in both price and performance in the broader portions of the product stack. We expect that Intel will continue vigorously defending its shrinking data center market share with attractive pricing. Still, those efforts will be confined to what is essentially the middle of AMD's Genoa stack — Sapphire Rapids tops out at 60 cores, while Genoa carves out a whole new definition of the high end that it alone occupies.
Given the core count and amazing performance advantages of the EPYC Genoa platform, not to mention the increased compute density and TCO advantages of deploying fewer servers, AMD will likely sell every high-end Genoa chip it can produce, just as it has with Milan.
Paul Alcorn is the Deputy Managing Editor for Tom's Hardware US. He writes news and reviews on CPUs, storage and enterprise hardware.
Okay, for those more knowledgeable than me (I'm not really into server tech), how is it that Intel is so far behind in terms of core count with these systems? Looking at some of the benches (and I might as well be blind in both eyes and using a magnifying glass to scroll through the data!) it seems to me that if Intel were able to increase core count, that they would be comparable in performance to the AMD counterparts? What gives?Reply
Intel have taken the performance crown with ADL and Raptor in the consumer market, but on the bigger scales can't get close.
Well, basically AMD has mastered the art and technology of connecting lots of small chips together. While Intel still has to make chips as one big lump. Big chips are harder and more expensive to produce. Plus when combining small chips you can always and easily add more.Reply
Intel might be competing right now in the consumer space - but they don't earn much profit on their expensive to produce CPUs.
Additionally - AMD is cheating a bit and is several chip nodes ahead in production process - while Intel is struggling to get past 10 nm, AMD is on what, 4 nm? Because of TSMCs impressive technology leadership and Intels stubborness.
Strange that Tom's mentions that DDR5 support for these new Data Center CPUs is a "Con" but they don't mention it in yesterday's article of Intel's new CPUs using the same DDR5.Reply
But then maybe not, knowing Tom's. It's just so obvious.
Roland Of Gilead said:Intel have taken the performance crown with ADL and Raptor in the consumer market, but on the bigger scales can't get close.
Rumour has it that AMD prioritize less on the consumer market, and more on the server marked. So if "one core to rule them all" it means Zen4 is a core designed primarily for server chips. Whether Alder Lake took the "performance crown" or not is at best debateable, as is Raptor Lake vs Zen4 if power consumption is taken into consideration.
Intel does not use anything equivalent of chiplets (yet). The die size of raptor lake 13900 is about 257 square mm. Each Zen4 ccd is only 70 square mm (two in a 7950x). That gives AMD a tremendous advantage in manufacturing and cost.
See oMcsW-myRCU:2View: https://www.youtube.com/watch?v=oMcsW-myRCU&t=2sfor some info/estimates on yields and cost of manufacturing.
You have forgotten the face of your father.Reply
Most likely multiple reasons.
1) Intel struggled for a long time trying to reach the 10nm process
This delayed entire product lines for a couple of years and ultimately led to Intel outsourcing some production to TSMC which wasn't struggling to shrink the fabrication process2) AMD moved to a chiplet strategy long before Intel, which I don't believe has a product for sale using chiplets yet, not sure though
AMD had already been using TSMC as AMD had sold off their manufacturing facilities years before
Large monolitic, high core-count CPUs are harder to make than smaller lower core-count CPUs- An example is AMD putting two 8-core chiplets in a package (plus IO die) for a product that has 16 cores
- Intel has recently countered this by going with heterogeneous cores in their CPUs; a mix of bigger/faster and smaller/slower cores
- I don't believe that the heterogeneous core strategy has been implemented in servers products yet
The DIMM slot fragility issue could easily be solved or at least greatly improved by molding DIMM slots in pairs for lateral stability and sturdiness.Reply
Hile Gunslinger :)bitbucket said:You have forgotten the face of your father.
Fair pointSunMaster said:Whether Alder Lake took the "performance crown" or not is at best debateable, as is Raptor Lake vs Zen4 if power consumption is taken into consideration.
It is just me or do you guys also think that AMD can fit 24 CCDs in the same package in the not so distant future?Reply
Sure there's some small SMDs in the way, but it should be doable.
Just imagine one of those with 192 ZEN4 cores ou 256+ ZEN4c cores.
Maybe with ZEN5?
Intel's only bastion seems to be accelerators and burn as much money as they can on adoption, even worse than AVX512.Reply
I just looked at the numbers of OpenSSL and I just laughed... Intel is SO screwed for general purpose machines. Their new stuff was needed in the market last year.