AMD EPYC 3201 8-Core Benchmarks Review and Milestone


AMD EPYC 3201 Market Positioning

Thes chips are not released in a vacuum instead, they have competition on both the Intel and AMD sides. When you purchase a server and select a CPU, it is important to see the value of a platform versus its competitors.

AMD EPYC 3201 v. AMD Alternatives

When I first heard about the AMD EPYC 3201 specs, I admittedly did not understand why AMD omitted hyper-threading. After seeing this CPU in comparison to the AMD EPYC 3101 and the EPYC 3251, it makes sense. AMD ended up with four single-die SKUs that have surprisingly good segmentation.

We wish that we could find someone using dual-die AMD EPYC 3000 SKUs. It would be interesting to see how adding the second die plus adding the Infinity Fabric link between the two impacts performance and power consumption. The dual die parts are officially on my wishlist now for STH to review.

AMD has great segmentation between the AMD EPYC 3201 and the lower end of the 7000 series, the AMD EPYC 7251. Packaging requirements for the EPYC 7000 series are such that fitting them into embedded form factors is challenging. A 4x TDP differential also puts it in a different class.

For the other embedded part from the company that we recently tested, the AMD Opteron X3421 performance is much lower. We most commonly see this chip in the HPE MicroServer Gen10 which utilizes the X3421’s GPU capabilities to drive 4K displays. We think the non-GPU SoC that is focused on CPU performance in the AMD EPYC 3201 is sufficiently segmented enough to not impinge on that domain.

AMD EPYC 3201 v. Intel Alternatives

Comparing the AMD EPYC 3201 to its Intel counterparts is challenging. Intel has many lines available between the Intel Atom C3000 “Denverton”, Intel Xeon D-1500, Intel Xeon D-2100, Intel Xeon E-2100/ Core series, and the Xeon Scalable SKUs.

With so many lines, there is a lot of overlap. Overall, the AMD EPYC 3201 is faster core-for-core than the Intel Atom C3000 series. It falls somewhere between the Intel Xeon D-1500 and Xeon D-2100 series performance.

Compared to the Intel Xeon D-1500 series parts like the Intel Xeon D-1537 and Xeon D-1528, power consumption is similar but the performance of the AMD EPYC 3201 tends to be a bit better. Also, the AMD EPYC 3201 is a more modern SKU so it has more high-speed I/O lanes and internally a quad port AMD NIC. Vendors tend to prefer Intel NICs for ease of integration and support, but the four 10GbE ports are available to embedded systems vendors.

Broadwell EP LGA 2647 Broadwell DE Package Size Comparison
Broadwell EP LGA 3647 Broadwell DE Package Size Comparison

Compared to the Intel Xeon D-2100 series parts like the Intel Xeon D-2123IT, the AMD EPYC 3201 performs well. It has large caches which make up for the lower clock speeds. AMD also has a much lower TDP and the difference in power consumption is substantial, to say the least.

The Intel Xeon E-2100 series and Skylake-SP use more power so they are probably not the best competitors except if you have an edge form factor that is not power and size constrained. We think there is enough differentiation that although these are added to our charts, they are not really competitors.

AMD EPYC 3201 v. Arm Competition

This is coming. Intel and the rest of the industry know it is coming. Currently, the main issue that Arm competition faces is the fact that many of the products in this class are not fully SBSA compliant and do not install vanilla operating systems without at least some work whether from drivers or other areas. The AMD EPYC 3201 works well in this regard since the AMD EPYC 7000 series has trailblazed for the embedded parts.

We think this will change and will have more on this soon, however, if you want x86 at the edge, AMD has a strong differentiation point.

Final Words

Refreshingly, the AMD EPYC 3201 does not try to do everything, nor does it attempt to exactly replicate and Intel SKU. AMD succeeded in making a differentiated part that offers a significant delta between it and the AMD EPYC 3251.

AMD has a great product in the embedded space. The impact of this cannot be understated. Part of the Intel Xeon D strategy has been matching the instruction sets of Intel’s mainstream server CPUs. Some part of that is premised on being able to migrate workloads from big iron in the data center out to the edge. Now AMD EPYC has that where one has an embedded option to match the Zen cores running in your desktop.

On the other hand, the embedded market is slow. If one thinks that the server market is slow, the embedded market is much more conservative. Many products in this segment are expected to last 7 years or more, much longer than the expected lifecycle of a data center product. Getting AMD EPYC 3000 out to the point that our readers can get a few boards to test is taking just over a year from its launch. That lag is much longer than the Intel side which makes sense, given that Intel has put such a focus on this market.

We are seeing companies likeĀ Packet launching microservers with Netronome SmartNICs and EPYC 3000 CPUs. Still, AMD needs more engagement with the EPYC 3000 series. We think the product has specs to be successful, but the company needs its first wave of channel products, along with the traditional embedded space, to be successful. Competition is good in this space and the AMD EPYC 3201 is different enough that customers will see different applications that the Intel Xeon D offerings do not currently cover well.


  1. That’s a great point on the supplier diversity. We’ve exclusively used Intel for over 6 years but that AVR54 bug hurt and we didn’t have an answer. We’re going to buy whatever comes out first if it fits our use and see if we can qualify a different part. We’ve got a few customers that would prefer their HA pair to be similar performance but with different components so they don’t get caught stuck with that AVR54 worry

  2. Xeon D-2100 pricing is f*ing crazy. Dilly Dilly to AMD getting in there and stopping runaway embedded pricing.

    If AMD’s cheaper, I’m going to buy 1-3 just to show support and keep them going.

  3. What I don’t understand about the current ecosystem is that nobody seems to be building embedded parts based on the 3000 series. While you can purchase several Xeon D boards with embedded CPU I haven’t seen anything indicate that Epyc parts are coming. Even SuperMicro officially announced that they have no parts in the planning.

    So here’s to hope that AMD will produce boards with embedded CPUs themselves to get the market to follow.

  4. Patrick, thanks for all the hard work – I don’t suppose you know the volume pricing on the 3201 do you?

    I’d love to know if it comes out cheaper than the 3758? It feels like a direct competitor to me, considering it’s only 5w more and in /most/ benchmarks it’s anywhere from 5 to near 40% faster.

    I can only hope for good things for this platform moving forward, hopefully lots of linux, windows, freebsd / freenas support.

    Good to have some options.

  5. I am committed to producing a viability study for the practicality of immediate commencing manufacturing of the very similar products. (this said, I cannot say that the actual product designs are very much more exciting than what is seen in the market and its also fact that we are wanting to manufacture for customers with demand for unique product and services for their businesses, and unlike the public market in our area we have a lot of capacity to entertain high quality and finish quality products in retailers that would be very expensive sales for comparison. If you have any more interest then please contact me by all means the potential to be financed in a manner that will permit paid full equity in the additional compensation, very possible to who is relaxed to take the product from the errors which must be eliminated to last inspection, with the least personal concerns of stress and who is as concerned about why everything is slow in this market and no explanation is certain and sufficient. We want to be odm and to be retail for potential cases design as well as selling the limited edition stocks that are different completion of bigger customer service and specifications making.

  6. Why Supermicro has crippled this great platform?

    We could have:
    8x SATA (RAID 0,1,5,10)
    4x 10 GbE
    2x M.2 Slots (like ASUS P11C-M/4L has)
    1x 24-pin ATX power connector for PicoPSU

    I can only hope another manufacturer will solder Epyc 3201 to a mATX board, so I can get rid of my power hungry 2011-3 RDIMM based homelab. ;-(

  7. Simon – I think that page will get updated.

    Stephan – with these, you simply need a DC power cable and do not need the PicoPSU. You can run them directly from a power brick. The appeal of this model is you can skip the extra power supply.

  8. Patrick – Maybe I can somehow connect a 12V DC power brick to the mainboard itself (DC plug to 8-PIN EPS adapter – part number?) but how do I then power SSDs & HDDs? Retail package contains 4x SATA cable but CBL-PWEX-0982 offering only one SATA power connector and there is no information in manual about how much milliampere we can drain from J6 without smoke.
    Recommended case E300 has a DC connector but the cable inside the case to 8-PIN mainboard connector is not in parts list.
    Supermicro already made a mainboard I’m searching for: A2SDV-8C-TLN5F. I want this board with Epyc 3201 and 8x SATA.
    It seems to be modern these days to sqeeze everything to the smallest possible formfactor at any costs.
    Low power server CPU and ATX form factor for expansion cards and features? – exactly one mainboard available: Gigabyte MB51-PS0. This is so sad.


Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.