AMD’s New EPYC 7F52 Reviewed: The F is for ᴴᴵᴳᴴ Frequency
by Dr. Ian Cutress on April 14, 2020 9:45 AM EST- Posted in
- CPUs
- AMD
- Enterprise
- Enterprise CPUs
- EPYC
- SP3r2
- CPU Frequency
- Rome
- 7Fx2
Conclusions
There are three main ways to increase modern computing performance: more cores, higher frequency, and a better instruction throughput per cycle (IPC).
The one everyone loves, but is the hardest to do, is to increase IPC – most modern processor designs, if they are evolutions of previous designs, try to ensure that IPC increases faster than power consumption, such that for every 1% increase in power, there might be 2% increase in IPC. This helps efficiency, and it helps everyone.
As we’ve seen with some recent consumer processors, IPC is nothing unless you can match the frequency of the previous generation. Increasing frequency should sound easy: just increase the voltage, which gives the unfortunate side effect of heat and decreases the efficiency. There’s also another element at play here, in physical design. The ability to produce a layout of a processor floorplan such that different parts of the CPU are not affecting the frequency is a key tenet to good physical design, and this can help boost maximum frequencies. If you can’t get IPC, then an increase in frequency also helps everyone.
An increase in core count is harder to quantify. More cores only helps users that have workloads that scale across multiple cores, or gives an opportunity for more users to work at once. There also has to be an interconnect to feed those cores, which scales out the power requirements. Cores doesn’t always help everyone, but it can be one of the easier ways to scale out certain types of performance.
With the new 7F range of Rome processors, AMD is hoping to stag that first second rung of the ladder. These new parts offer more frequency, but also improve the L3 cache to core ratio, which will certainly help a number of edge cases that are L3 limited or interconnect limited. There is a lot of demand for high frequency hardware, and given the success of the Naples 7371 processor from the previous generation, AMD has expanded its remit into three new 7F processors. The F is for Frequency.
The processor we tested today was the 7F52, the most expensive offering ($3100) which has 16 cores with a base frequency of 3.5 GHz and a turbo of 3.9 GHz. This is the highest turbo of any AMD EPYC processor, and this CPU is built such that there is 256 MB of L3 cache, offering the highest core-to-cache ratio of any x86 processor. At a full 16 MB per core, this means that there is less chance for congestion between threads at the L3 level, which is an important consideration for caching workloads that reuse data.
Our tests showed very good single thread performance, and a speedy ramp from idle to high power, suitable for bursty workloads where responsiveness matters. For high throughput performance, we saw some good numbers in our test suite, especially for rendering.
Personally, it’s great when we see companies like AMD expanding their product portfolio into these niche areas. High frequency parts, high cache parts, or custom designs are all par for the course in the enterprise market, depending on the size of the customer (for a custom SKU) or the size of the demand (to make the SKU public). AMD has been doing this for generations, and in the past even created modified Opterons for the Ferrari F1 team to do more computational fluid dynamics within a given maximum FLOPs. I’m hoping AMD lets us in on any of these special projects in the future.
Threadripper, Rome, Naples. AMD introducing RGB to CPUs
Facebook
Twitter
RSS
97 Comments
View All Comments
Santoval - Thursday, April 16, 2020 - link
"This cpu line is low margin and unable to seriously beat Intel big superiority in raw core performance."Zen 2 based servers CPUs already beat Intel in "raw core performance". This "F" series AMD introduced is not meant to beat Intel, since they already have. It is meant for certain customers who want fewer but faster cores/threads. Examples might be high-end workstations rather than servers (or workstations disguised as servers), which scale well up up to 16 - 24 cores and do not need CPUs with 32 - 64 cores which provide less performance per core as a trade-off.
As for the server market share AMD is going to exceed -or rather, *was* going to exceed, before Covid-19 froze everything -a 10% market share in Q2 2020 already (rather than a mere 5%). Source (in the 8th paragraph) :
https://www.forbes.com/sites/tiriasresearch/2020/0...
tyleeds - Thursday, April 16, 2020 - link
We used to custom order for customers what we called "The Oracle Special". Due to the way Oracle lays out their licensing on the database, you're looking for relatively low core counts, but screaming fast with a lot of cache. The price of Oracle licensing means you can safely say "power be damned" and just get the fastest core you can manage.This looks a lot like that...
Lord of the Bored - Monday, April 20, 2020 - link
"Intel on 14nm""AMD are late a lot"
You, sir, are a comedy genius!
schujj07 - Tuesday, April 14, 2020 - link
Yes the Intel counterpart is on 14nm and has a 205W TDP, but as we all know Intel's TDP is only measured off of base clock. During actual usage its TDP is much higher than 205W. This is why we see the Threadripper 3970X using less power than the 18 core Intel 10980XE even though the 10980 has a much lower TDP. https://www.servethehome.com/amd-ryzen-threadrippe... For here the 7F52 has higher performance than the 6246R and when you have workloads that are frequency sensitive that extra power doesn't matter as much.Deicidium369 - Tuesday, April 14, 2020 - link
Their server CPUs are a different thing than the desktop - You give people a little bit of info and all of a sudden they are freaking experts on power usage. So 205 is 205. NO ONE overclocks server CPUseek2121 - Tuesday, April 14, 2020 - link
It actually has little to do with overclocking on the Intel side. A stock Core i9 9900K will blow through it’s limit (both power and heat) with the vast majority of motherboards out there today.Their server CPUs, however, adhere to TDP.
schujj07 - Tuesday, April 14, 2020 - link
Not exactly true when there is a load. Max draw on dual 8280's is 685.1W for a 205W TDP. Due to the boost nature, the CPU will draw a lot more power. https://www.tomshardware.com/reviews/intel-cascade...Note the Epyc also draws more than its TDP as well and the review doesn't say whether this is total system or just CPU.
Oxford Guy - Saturday, April 18, 2020 - link
Reviewers or someone...There needs to be serious pressure to create a triple metric.
1. Maximum power the CPU can draw with a synthetic workload that maxes it as completely as possible.
2. Maximum power the CPU can draw with a real-world program (come up with an industry consensus).
3. For consumer CPUs: Maximum power the CPU can draw using the world's most demanding real-world gaming title. For prosumer and enterprise CPUs: Maximum power the CPU can draw with a second real-world program that is very different from the other one.
Stop enabling useless metrics that don't match reality.
bug77 - Tuesday, April 14, 2020 - link
And this is despite you having power measured on the second page of this review.schujj07 - Tuesday, April 14, 2020 - link
The chip on the 2nd page is the 6226R which will not compete with the 7F52. The competing chip from Intel is the 6246R.https://ark.intel.com/content/www/us/en/ark/produc... - 6226R
https://ark.intel.com/content/www/us/en/ark/produc... -6246R
The added 500MHz base clock brings the TDP from 150W to 205W.