Review AMD Ryzen 9 3950X [AM4]: Specs | Test | Hasrate | CPU performance| Set-up | Config – Moving to a standard far beyond what most of us might need, but in doing so demonstrating what it is capable of doing right now, AMD surprised the market with a 16-core processor in the AM4 socekt. The Ryzen 9 3950X based on the famous Zen 2 architecture offers performance that can get along very well under high-demand multi-threaded scenarios, as well as in video games (a certain transition from socket AM4 to TRX4).
To understand more what makes it a great beast of computing at an enthusiastic level, let’s see how it is shaped internally.
In this way the CPU is modularized, and we saw this with Ryzen Threadripper and Epyc processors for servers, all using the high-speed highway called Infinity Fabric 2.0, which will communicate both CCD units through the cIOD.
With these new processors, and new support for PCIe 4.0, it became necessary to update the chipset. For this, the new X570 appears, which provides more SATA and NVMe ports, all through the PCIe 4.0 interface, as well as 10Gbps USB ports (USB 3.2 Gen2). Obviously communication with the CPU continues via a PCIe 4.0 x4 link.
Everything is in the routing
In order to communicate between the cores, all the routing of tracks passes through the IO die, which is ultimately in charge of communicating the CCDs with each other, and the CPU with the memories and the other components. In order to achieve this, the number of layers of substrate had to be increased to 12.
A close up of one of the layers of the CPU, you can see at a glance, the Infinity Fabric lines, ranging from the CCD to the IO. There is no direct communication between the 2 CCDs.
New design topology
With this new design challenge to incorporate more cores, within the same socket size, the clearest move was to move to a chiplet topology. In this way, they can add more and more cores, just by adding or removing chips from the gasket.
The basic unit of Zen2 is the CCD, which is made up of 2 CCX units (Core Complex) of 4 cores each, and 16MB of L3 cache (twice as much as before). Therefore, each Zen2 CCD unit has a maximum of 8cores / 16Threads, and 32MB of L3 cache.
This topology is what we see in the 3rd generation Ryzen 5 and Ryzen 7 processors. Since the maximum number of cores is 8. Remember that communication with the IO is through the Infinity Fabric, and it is this chipset that is in charge of communicating with the other components.
When you want to go beyond 8 cores, it is necessary to add one more CCD, with this the maximum will be 16 cores / 32 Threads and 64MB of L3 cache. This is why a processor like the Ryzen 9 3950X (16C / 32T) is possible.
Internal improvements (Better cache and new stages)
At a high level, the kernel looks a lot like what we already know about Zen. The highlights of Zen 2 design include a different L2 branch predictor known as a TAGE predictor, a duplicate cache micro-op, a doubling of L3 cache, an increase in integer resources, an increase in load / storage resources, and support for single-operation AVX-256 (or AVX2).
Despite all these changes, AMD has stated that there is no frequency penalty when using AVX2, depending on its frequency platform with energy recognition.
AMD has also made caching adjustments, most notably the L1 instruction cache, which has been halved to 32kB, but the associativity has doubled. The L1 data cache and L2 caches have not changed, however the translation buffers (TLB) have increased their compatibility. AMD also claims that it has added more virtualization support regarding security, which helps enable the features in the future.
Taking a quick look, it’s easy to say that duplicating the micro-operative cache will offer a significant IPC improvement in a number of scenarios, and combining that with an increase in load / storage resources will help drive more instructions. The dual L3 cache will aid in specific workloads, as AVX2 single-operation support would, but the improved branch predictor will also show an increase in raw performance. In summary, for a paper analysis, AMD’s + 15% CPI improvement appears to be a very reasonable number to promote.
Security: Something that stays with Zen 2
One of the important issues of the past year, and one that remains throughout 2019, is security in CPUs, specifically those based on the X86 architecture. Since its design and microcode allowed malicious people to be able to perform tasks remotely, or execute malicious code to extract data, without one noticing.
Like all processors based on the Zen architecture, these new Zen 2 have mitigations at the hardware level and also at the operating system level already implemented by Microsoft. Therefore, they would not be affected by the recent and better known security problems in the CPUs, which on the Intel side, has become a major headache.
Memory Support and Overclock
Another point that this new family of Ryzen processors improves, is the support and overclocking in the memories. Something that in current generations of Ryzen processors caused a lot of headaches. Well, with this new Zen 2, and the improved I / O chip, using high-frequency memories will not be a problem, nor will it be to take them to the highest in their frequencies.
In fact, they promise that frequencies of up to 5100MHz can be achieved in the memories (which allow it, clearly) and all this by air. This could be accomplished, in part by the above, by separating the chips in the CPU and providing a better memory controller in the IO chipset, in conjunction with the X570 chipset. In addition to that with Infinity Fabric 2.0, dividers are implemented automatically, to maintain the stability of the system as a whole.
Whenever we install fast memories in our new system, Infinity Fabric will operate at a 1: 1 ratio with the frequency of the memories, this until a frequency of 3733MHz is reached. When we exceed that frequency, IF will automatically enter a 2: 1 ratio mode, this means it will operate at half the memory frequency.
This obviously means that the latency will increase a little, and the performance will be impacted by approximately 1-2%, but this had to be done in order to maintain the stability of the IF, and in the process to achieve higher frequencies in the memories.
Specifications – AMD Ryzen 9 3950X
|Specifications||AMD Ryzen 9 3950X||AMD Ryzen 9 3900X||AMD Ryzen 7 3800X||AMD Ryzen 7 3700X||AMD Ryzen 7 2700X||Intel Core i9 9900K|
|Freq. Base||3.5 GHz||3.8 GHz||3.9GHz||3.6 GHz||3.7 GHz||3.6 GHz|
|Freq. Max||4.7 GHz||4.6 GHz||4.5GHz||4.4 GHz||4.3 GHz||5.0 GHz|
|Sist. Refrigeration||–||Wraith Prism with RGB LED||Wraith Prism with RGB LED||Wraith Prism with RGB LED||Wraith Prism with RGB LED||N / A|
|Price USD||$ 499||$ 329||$ 254||$ 485|
|CLP price||PCFactory: $ 869,990||$ 445,000||$ 389,990||$ 315,900||$ 250,000||$ 439,900|
Hashrate – AMD Ryzen 9 3950X
|Processor||– AMD Ryzen 9 3950X|
|Motherboard||– GIGABYTE X570 AORUS MASTER|
|Memories||– G.Skill TridentZ 3200 MHz 2x8GB|
|Refrigeration||– EK-XLC Predator 240|
|Graphics card||– Sapphire Pulse RX 5700 XT 8GB|
|Power supply||– Corsair RM1000X|
|Storage||– Corsair MP600 1TB M.2 PCIe 4.0|
|Monitor||– ASUS MG28UQ|
Testing Platform and Methodology.
- Windows 10 Pro x64 operating system [Update 1903].
- The tests were carried out in an environment with a temperature of approximately 25ºC.
- The platform was used without a cabinet.
- Video Driver used: Adrenalin 2020 Edition 20.2.2
- The resolutions of the synthetic tests are predetermined by each of the benchmarks.
Although we have to sacrifice the top frequencies of some cores to go to 4.2 GHz for the 16 cores, we see a significant improvement in multi-core performance. The improvement that the overclock made gave us reached up to 10% more.
Mono-core performance is sacrificed, however squeezing out a model like this would point to multi-core rather than mono-core use.
Consumption and Temperature.
The consumption of this 16-core model, has a standby of 32W, while under stress it can develop a total of 133W at 100% load, leaving us with 28W above what is delivered as TDP. Already with overclocking the processor develops a consumption that easily exceeds 200W, reaching 216W at its maximum pick. Clearly in load without overclocking it does not show to be a processor of such characteristics, if it is that we compare it to what the competition could develop or a Threadripper of 16 cores, what all this consumption data seems to be more than appropriate for everything that we delivers this AMD model.
The temperature that could be a section where we could have found some not very pleasant experience with this model, reflex to deliver totally different data. We must remember that this model does not bring factory refrigeration, so experience in this area will be low when they are willing to invest in keeping this model cool. Under a high performance AIO liquid cooling system, a maximum of 74 ° C was experienced under load, not bad for this AM4 socket colossus, under overclocking it was not so bad considering that in consumption it generated up to 62% more, while in temperature it rose by just 13 ° C.
Talking about the AMD Ryzen 9 3950X are bigger words, although we have in front of us only the AM4 socket that has been on the market for more than 3 years, this socket has shown us that it has an amazing capacity by easily delivering the support to a processor the characteristics of this Ryzen9 3950X. AMD was true to its word by delivering a socket with no physical changes to it that lasted until 2020.
The motherboard is something very important when choosing a processor such as the Ryzen 9 3950X, we have in front of us no more and no less than 16 cores, with a performance capable of easily matching workstation equipment in many aspects, so the choice of a companion for this will not have to be taken lightly. Beyond knowing the support that the manufacturer provides, we recommend looking for a robust model at an electrical level so that it can support 100% of this processor.
The lack of a cooling solution in this model can have a pros and cons, the advantage of not having it is that it comes at a more affordable price, considering that a cooling system for a processor like this could easily add around $ 50,000 CLP at th final price (at least), but for those who pay for such a processor, it is very likely that they already have high-performance cooling and well the disadvantage is this extra expense that we must consider.
In Chile we do not have many units of this model, it was clearly a processor with a demand that AMD probably did not expect and was exceeded, so it tends to be in short supply, something similar to what happened to the Ryzen 9 3900X at the beginning, which over time availability stabilized.
To end this section of the review, we must say that this AMD model is a key model between the transition from an “enthusiastic” use platform to a real workstation, very versatile in bringing together the main characteristics of both worlds.
Review AMD Ryzen 9 3950X [AM4]
- – Versatility, delivers great performance in both video games and content creation
- – Low price in relation to what it is capable of delivering and what the competition offers
- – Socket AM4
- – Does not have factory refrigeration
- – Low availability
Users Ranking: /5
( Be the first! votes )
We ust thank PC Factory again for giving us the opportunity to try this AMD Ryzen 9 3950X.