AMD Zen 5 Server CPU Rumors: Turin With Up To 128 Cores, Turin-X With Up To 1.5 GB L3 Cache, Turin Dense With 192 Zen 5C Cores

Hassan Mujtaba
ecddbee0-8495-486a-a3a0-f9806cbe6ced

In addition to the Ryzen 8000 leak, information regarding AMD's Zen 5 & Zen 5C powered EPYC Turin, Turin-X & Turin-Dense CPUs have been revealed too.

AMD Readies At Least Five Zen 5 & Zen 5C CPU Families For EPYC Servers: Turin, Turin-X, Turin-Dense, Turin AI & Sorano

Once again, these new rumors come from Moore's Law is Dead latest roadmap which points out the next-generation EPYC CPU families. The roadmap covers at least five EPYC CPU lineups which can be expected in 2024-2025. AMD has already confirmed Turin as its next-gen EPYC family and it will utilize both Zen 5 and Zen 5C cores,

Related Story AMD Says It Had To Rebuild The Ryzen 5 5800X3D To Bring It Back For AM4’s 10th Anniversary
Image Credits: MLID

AMD EPYC Turin & Turn-X With Zen 5: Up To 128 Cores, 4nm Process

Starting with the first family, we have the AMD EPYC Turin (Classic) which will stick with the chiplet design and house up to 128 cores, 256 threads, and TDPs of up to 500W which can be configurable on certain SKUs up to 600W (as revealed in today's Gigabyte leak). In a previous leak, it was shown that the EPYC Turin chips would feature the same L2 and L3 cache as Zen 4 with a small upgrade to the L1 cache.

The major change will come in how AMD arranges the cache on Zen 5 chips which are expected to utilize a "Ladder" hierarchy. The chips are shown to enter production by Q1 2024 & will utilize the TSMC 4nm process node. The actual launch is expected to take place by the third quarter of 2023.

AMD's Zen 5 CPU cores are rumored to feature a brand-new cache design. (Image Credits; AdoredTV)

Moving on, we have the AMD EPYC Turin-X chips which will be outfitted with a 3D V-Cache. These chips will retain the 64MB of 3D V-Cache per CCD which totals 1024 MB across the 16 CCDs & 512 MB of standard L3 cache. Totaling up to 1536 MB or 1.5 GB of L3 cache. If we combine the L2 cache which is 1 MB per core or 128 MB in total, that increases to 1664 MB of total cache which is still not including the L1 cache. That's a 33% higher cache compared to the upcoming Genoa-X CPU family.

AMD EPYC Turin Dense & Turin AI With Zen 5C: Up To 192 Cores, 3nm Process

Moving over to the Zen 5C side of things, we first have the AMD EPYC Turin Dense chips which will be succeeding Bergamo. Turin Dense isn't an official name for now but it is expected to utilize the 3nm Zen 5C cores in up to 192 core SKUs. These chips will feature up to 500W TDPs but the most interesting thing is that they are expected to hit production before the standard Turin chips. MLID states that this is due to AMD speeding things to compete directly against Intel's Sierra Forest 144 core chips which are also expected around the same time in the first half of 2024.

There's also a secondary Turin Dense chip known as Turin AI which is expected to feature the same Zen 5C cores but with an AI chiplet. There aren't many details regarding this specific SKU but we can expect Xilinx IP to power the AI chiplet for certain applications. AMD is making AI its no.1 strategic priority and it makes sense to incorporate more AI-specific hardware within its EPYC CPUs for the customers that demand it.

Lastly, there's the follow-up to AMD's Siena CPUs known as Sorano. The AMD EPYC 8004 "Siena" lineup makes use of Zen 4/4C cores on the more mainstream and low-cost/low-TCO/low-power platform known as SP6. It has 6-channel memory support and 96 PCIe gen 5 interconnects. These chips will retain up to 64 cores and 225W TDP and production is expected in the 2H of 2024 with availability sometime in 2025.

Image Credits: MLID

AMD has so far announced its upcoming “AMD Data Center and AI Technology Premiere,” which will be the center stage for various server and data center-specific announcements. Definitely expect to hear more about future and upcoming products at the event which is taking place in June.

AMD EPYC CPU Families:

Family NameAMD EPYC VeranoAMD EPYC VeniceAMD EPYC Turin-XAMD EPYC Turin-DenseAMD EPYC TurinAMD EPYC SienaAMD EPYC BergamoAMD EPYC Genoa-XAMD EPYC GenoaAMD EPYC Milan-XAMD EPYC MilanAMD EPYC RomeAMD EPYC Naples
Family BrandingEPYC 9007EPYC 9006EPYC 9005EPYC 9005EPYC 9005EPYC 8004EPYC 9004EPYC 9004EPYC 9004EPYC 7004EPYC 7003EPYC 7002EPYC 7001
Family Launch2027202620252025202420232023202320222022202120192017
CPU ArchitectureZen 7Zen 6Zen 5Zen 5CZen 5Zen 4Zen 4CZen 4 V-CacheZen 4Zen 3Zen 3Zen 2Zen 1
Process NodeTBD2nm TSMC4nm TSMC3nm TSMC4nm TSMC5nm TSMC4nm TSMC5nm TSMC5nm TSMC7nm TSMC7nm TSMC7nm TSMC14nm GloFo
Platform NameSP7SP7SP5SP5SP5SP6SP5SP5SP5SP3SP3SP3SP3
SocketTBDTBDLGA 6096 (SP5)LGA 6096 (SP5)LGA 6096LGA 4844LGA 6096LGA 6096LGA 6096LGA 4094LGA 4094LGA 4094LGA 4094
Max Core CountTBD9612819212864128969664646432
Max Thread CountTBD19225638425612825619219212812812864
Max L3 CacheTBDTBD1536 MB384 MB384 MB256 MB256 MB1152 MB384 MB768 MB256 MB256 MB64 MB
Chiplet DesignTBD8 CCD's (1 CCX per CCD) + 2 IOD?16 CCD's (1CCX per CCD) + 1 IOD12 CCD's (1CCX per CCD) + 1 IOD16 CCD's (1CCX per CCD) + 1 IOD8 CCD's (1CCX per CCD) + 1 IOD12 CCD's (1 CCX per CCD) + 1 IOD12 CCD's (1 CCX per CCD) + 1 IOD12 CCD's (1 CCX per CCD) + 1 IOD8 CCD's (1 CCX per CCD) + 1 IOD8 CCD's (1 CCX per CCD) + 1 IOD8 CCD's (2 CCX's per CCD) + 1 IOD4 CCD's (2 CCX's per CCD)
Memory SupportTBDDDR5-12800DDR5-6000?DDR5-6400DDR5-6400DDR5-5200DDR5-5600DDR5-4800DDR5-4800DDR4-3200DDR4-3200DDR4-3200DDR4-2666
Memory ChannelsTBD16-Channel (SP7)12 Channel (SP5)12 Channel12 Channel6-Channel12 Channel12 Channel12 Channel8 Channel8 Channel8 Channel8 Channel
PCIe Gen SupportTBD128-192 PCIe Gen 6TBD128 PCIe Gen 5128 PCIe Gen 596 Gen 5128 Gen 5128 Gen 5128 Gen 5128 Gen 4128 Gen 4128 Gen 464 Gen 3
TDP (Max)TBD~600W500W (cTDP 600W)500W (cTDP 450-500W)400W (cDP 320-400W)70-225W320W (cTDP 400W)400W400W280W280W280W200W

Hassan Mujtaba Photo

About the author: A Software Engineer by training and a PC enthusiast by passion, Hassan Mujtaba serves as Wccftech's Senior Editor for hardware section. With years of experience in the industry, he specializes in deep-dive technical analysis of next-generation CPU and GPU architectures, motherboards, and cooling solutions. His work involves not only breaking news on upcoming technologies but also extensive hands-on reviews and benchmarking.

Follow Wccftech on Google to get more of our news coverage in your feeds.

Button