Table of Contents
AMD’s new X570 platform introduces enhanced memory speed capabilities. The integrated memory controller in previous generations of Ryzen CPU’s severely limited the frequency and range of timings. To take advantage of the all-new memory capabilities, G.Skill introduced the Trident Z Neo. The Neo’s built-in XMP memory profile is optimized to work with AMD Ryzen 3 series CPU’s and X570 motherboards. We got our hands on the new G.Skill Trident Z Neo, so join us as we evaluate this memory using an X570 motherboard.
Specifications and Features
The all-new Trident Z Neo is optimized for AMD Ryzen 3000 series, but what does that mean? If we assume that the Neo sticks are using common retail ICs, such as Samsung, Micron, or Hynix, the only way to make them specially formulated for X570 is to build a special XMP profile. That is precisely what G.Skill has done.
With AMD’s new Ryzen 3000 series motherboards and chipsets, we have found that many older XMP profiles simply don’t work. The reason for this is that Intel’s Z390 and AMD’s X570, for instance, require different secondary and tertiary timings to work effectively. This might be confusing for some because we find 3600 CL16 memory marketed for both Intel and AMD; however, the difference lies in what we refer to as the sub timings. To ensure maximum compatibility, the Neo lineup has been tested across a wide range of motherboards to strict quality standards.
G.Skill introduced the Trident Z Neo with a whopping 32 different configurations to choose from. With speeds ranging from 2666 MHz to 3600 MHz and primary timings from CL14-14-14 to CL18-22-22, the Neo lineup has something for everyone. Most users and PC enthusiasts will choose the 16 GB memory kits composed of two 8 GB sticks. If we focus our search down to only 16GB kits, we find the 2666MHz CL18 at the low-end and 3600MHz CL14-15-15 at the high-end.
Today we will be evaluating the 3600MHz CL16-16-16 memory kit. In the table below, we examine the particular details of our test memory.
|G.SKILL Trident Z Neo For AMD Ryzen 3000 Series|
|Capacity||16GB (8GB x 2)|
|Speed Spec||PC4 28800|
When purchasing DDR4 memory, the main factors to consider, other than memory size or physical features, are the operating frequency and timings. The XMP is a memory profile stored inside the actual memory, which allows the user to easily apply the rated frequency and timings.
For those interested, here is a closer look at the particular details of this kit of memory. Below is a screenshot of Thaiphoon Burner, which is a wonderful free tool that allows one to read the Serial Presence Detect (SPD) firmware of the DRAM. The SPD information is critical in determining how the stick will perform and how the computer will recognize it.
As the Thaiphoon Burner screenshot shows, this specific kit of memory is composed of Samsung B-DIE IC’s. This memory uses a total of 8 ICs, all of which are located on only one side. We found this to be an interesting choice as Samsung officially classified its B-Die series as EOL (end of life) in the second quarter of 2019.
In the months after Samsung’s EOL status of B-Die, the computer enthusiast community focused on new products from Hynix and Micron to hopefully fill the void of B-Die. The new ICs showed incredible overclocking capabilities and exceptional compatibility with AMD. We had anticipated a new release from G.SKill that implemented ICs made by either SK Hynix or Micron Technology, instead, we see the status quo of Samsung ICs.
It’s not easy to spot, but a sharp eye will see that this memory utilizes the modern A2 style PCB layout. This refers to the orientation of the 8 ICs on the actual memory PCB itself. While it may seem like a useless tidbit of information, this is relevant, because it can play a very substantial role in the relative overclocking ability on some motherboards.
G.Skill opted to package the new Neo in what we would consider a mid-level packing option. We have seen everything from high-end, fabric-lined boxes–which the Trident Z Royal comes in– down to simple plastic blister packs from G.Skill. Given the price point, we feel that this is a nice packaging option. It comes with a full-color box and even included a nice preview window.
Gently sliding out of the box, we are greeted with the classic plastic clamshell to safely ship and store the memory. G.Skill also included the now prototypical red case sticker inside the box.
We did get a sneak peek at the heat sink design from the box, but there’s nothing like seeing the real thing. Once freed from the packaging, we are met with a new adaptation of the same Trident Z design that we have come to know and love.
The new heat sinks sport two different colors, as well as two different material finishes. Half of the stick should be recognizable to most of you. It’s the same matte-black, anodized aluminum, which has a sexy brushed look to it. For the other half of the Neo’s redesigned heat sink, we are introduced to a semi-gloss, silver finish. The silver section makes use of a process known as power-coating. This leaves the surface a little textured with very small bumps. The powder-coating finish is prized for being extremely durable. While it might not fit the color scheme of all the systems out there, the new heat sink design does stand out and make an impression.
The color scheme happens to be a perfect match for our test system today. As listed above, we are running the ASRock X570 Taichi. The Neo looks as if it has been custom made just for this particular motherboard.
G.Skill’s Trident Z RGB design is like a warm blanket. We know it well and it makes us feel all happy and content. Seriously though, while the Neo’s heat outward appearance may be slightly different, when it comes to the LED lighting it’s business as usual.
The diffuser’s finish is somewhere between shiny and matte. The opaque white plastic helps to mute the inherent harshness of the super-bright LEDs. The overall effect is quite pleasing to look at, and the diffuser does a good job of hiding the hot-spots.
The Neo’s 8 individually-addressable RGB LEDs can be controlled through most modern motherboards software. The Trident Z Neo is officially supported with the following software: Asus’s Aura Sync, Gigabyte’s RGB Fusion, MSI’s Mystic Light Sync, and ASRock’s Polychrome Sync.
For those users who want to get the full RGB experience without vendor-specific software, G.Skill has created its own lighting control software.
Download the Trident Z Neo control software
Testing and Overclocking
AMD introduced serious changes to the memory capabilities and overclocking of the new X570 series chipset. As we talked about earlier, many of the existing XMP profiles on the market today are simply not compatible with Ryzen 3000 series and will not even boot when applied. It’s because of this that our first goal is to check and see if this memory even works with the XMP profile out of the box. We will attempt to apply the XMP profile and run benchmarking software to assess the relative stability.
If and when the XMP profile has been established to be stable, we will evaluate the memory from an overclocking perspective. We want to see what this memory can do, but without hurting it. Therefore, we will stick to what could be classified as 24/7 stable daily memory voltages. According to the XMP 2.0 certifications, the absolute maximum allowable voltage is 1.50 V VDDR. Thus, all overclocking endeavors will be conducted with no more than 1.50 V.
Below is the test system and resulting memory speeds that will be used to evaluate the memory and run the benchmarks.
|CPU||AMD RYZEN 7 3700X @ 4.3 GHz|
|Cooler||NZXT Kraken X62 280mm – All-In-One Cooler|
|Motherboard||ASRock X570 TAICHI AM4|
|Graphics Card||PowerColor RED DEVIL Radeon RX 580|
|Solid State Drive||Team Group L5 LITE 3D SSD|
|Power Supply||Enermax RevoBron 700W 80+ Bronze|
|Operating System||Windows 10 x64|
|Memory Speeds Compared|
|XMP ~ 3600 MHz CL16-16-16-16 + XMP Sub Timings @ 1.35 V|
|Test Case 1 ~ 4400 MHz CL16-16-16-16 + XMP Sub Timings @ 1.45 V|
|Test Case 2 ~ 3600 MHz CL13-12-12-12 + Tight Sub Timings @ 1.45 V|
|Test Case 3 ~ 4000 MHz CL14-14-14-14 + Tight Sub Timings @ 1.50 V|
As is the case with all overclocking adventures, your results may vary, so proceed only if you assume all risk. To view and examine all of the various memory profiles we use two primary tools which include AIDA64 and AMD Ryzen Master utility. AIDA64 is a powerful system diagnostic and benchmarking tool, that can be purchased for a reasonable price.
The modern DDR4 market offers a vast range of XMP memory. Shopping on Newegg we find everything from DDR4 2400 MHz, all the way up to DDR4 4866 MHz. The improvements in XMP profile speeds is greatly attributed to modern manufacturing processes and memory PCB layouts, however, it might not be attainable on all motherboards. The profile is intended to be a one-click overclock, but it likely only applies to enthusiast-grade motherboards.
If you are shopping for an AMD Ryzen 3000 series platform, G.Skill has taken the guess work out. All of the memory listed in that series is designed to work effortlessly with a huge verity of motherboards. If you are building a computer based on a truly budget motherboard, the compatibility rules still apply. Don’t expect that the fastest Trident Z Neo kit will work on the cheapest motherboard.
As you can see below, we had no problem running our memory using the one-click memory profile. In this case, the FCLK = 1800 MHz and UCLK = 1800 MHz.
We cannot examine the overclocking potential without at least briefly covering the new AMD memory structure. Within the new architecture, AMD decoupled the infinity data fabric clock (FCLK) and the unified memory controller clock (UCLK). In the previous generation, the clocks were inherently linked with memory frequency and dividers. The process of decoupling meant that AMD could push the frequency to substantially higher limits, from about 4000 MHz to over 5000 MHz.
When we go to test the memory and overclocking capabilities we need to take into consideration two new variables, FCLK and UCLK. Even though these variables are somewhat out of our control, they play a huge role in the overall performance and benchmark scoring. That being said, it’s not always fair to compare one frequency against another, because the FCLK and UCLK might be running at different speeds and will, thus, skew the test results. While it’s not within the scope of this memory review to dive deeper into the Ryzen 3 memory structure, just keep in mind that those variables are there.
For the first test case only, the operating frequency was increased with no other settings adjusted. With a voltage increase from 1.35 V to 1.45 V, the frequency was able to be increased by an astonishing 800MHz. Below is the resulting first overclock test case with FCLK = 1800 MHz and UCLK = 1100 MHz.
For the second overclock test case, the same very small voltage increase of 1.35 V to 1.45 V was used. Contrary to the first test case, this time only the memory timings were decreased in order to provide an alternative example of what types of overclocks are possible. The operating frequency was held at the XMP 3600 MHz, but the primary timings were decreased as far as possible while still maintaining relative stability. Below is the resulting second overclock test case FCLK = 1800 MHz and UCLK = 1800 MHz.
For the third, and final overclocking test case, the goal was to achieve the best overclock possible while still staying within the maximum allowable voltage of 1.50 v according to XMP 2.0 specifications. Finding the optimal overclock for a given criterion, such as voltage, is not an exact science and requires a high degree of knowledge, patience, and the right combination of equipment.
We tried to create a memory profile that was implemented at a higher frequency and also tighter timings. The resulting overclock was a 400MHz increase in frequency from XMP, with tighter timings (decreased) as well. In this example, the FCLK = 1800 MHz and UCLK = 1000 MHz.
First up, we used AIDA64 Cache and Memory Benchmark. In the graph below, it is clearly visible that each of the four different memory speeds compared had a noticeable change in the benchmark result, with the exception of memory write performance. Looking at the read performance we found it interesting that overclocking the frequency resulted in a lower score than stock XMP. This can be attributed to the substantially lower UCLK frequency in that case.
If we look at the memory copy results, all of the overclocking test cases improved performance. Considering the big picture for AIDA64, the overclocking test case 3–utilizing 4000 MHz CL14–produced the best results overall. Our goal of a balanced overclock profile was actually realized, at least in this benchmark.
Next up is Geekbench 3, and it has proven itself to be an excellent tool for determining the real-world performance of the system being tested. This type of benchmark is purely 2D calculation based, and there is no graphical processing element, so it’s a great analytical tool to evaluate memory performance.
This is where things get interesting. Firstly, we should point out that all of our overclocking endeavors increased the relative performance across the board. That is not a small feat, given the complexity of Ryzen 3 memory overclocking! Focusing on the overall score, we see that this ‘real world’ test benchmark favors a frequency of 3600 MHz with tight primary and sub timings. We believe this proved to be the best overall performance because the UCLK and FLCK frequency are both using the 1:1 ratio and maxed out at 1800 MHz.
Cinebench R15 scores don’t typically increase much with memory overclocking, but we included it because it’s quite possibly the most popular benchmark for computers today. As expected the results show a maximum of about 1% gain in the score, from overclocking test case 2.
WinRAR is an old benchmark, but still quite useful in modern computing. It evaluates performance by simulating file management tasks such as compression and extraction. Here we see a noticeable trend developing in the results.
Next, we will examine the performance using a few of the memory benchmark tests offered within the SiSoftware Sandra suite of benchmarks. The flagship product, known as Sandra, is a powerful suite of many different benchmarks to evaluate computer performance of all major components, including the processor, graphics, memory, and disk.
The memory bandwidth showed an increase in performance for every sequential memory overclock, with our balanced profiling winning the top spot. Looking at the transactional throughput, however, the latency penalties from decreased UCLK are clearly evident here. In both cases where UCLK was not running a 1:1 ratio, the score was lower. We found it in retesting that for that particular test, all of the overclocking results proved to be worse than XMP.
G.SKill’s Trident Z Neo did not disappoint. It delivered on its promise of AMD Ryzen 3 compatibility and allowed for serious overclocking headroom. G.Skill Introduced the Neo with a new two-tone color scheme and the same award-winning heat sink design. While it may not be the most exciting new heat sink design to be introduced by G.Skill, it’s enough of a departure from the standard Trident Z to keep things interesting.
We were pleasantly surprised by the overclocking potential of the Trident Z Neo. With a scant voltage increase of 10 mV–well within XMP 2.0 specifications–we were able to overclock the memory frequency by a shocking 800 MHz. That’s a testament to the legendary overclocking prowess of Samsung B-Die, but also of the improvements AMD made to the Ryzen 3 series integrated memory controller. However, we proved that memory frequency is not always the answer. In fact, we believe that the native frequency of 3600 MHz was, in fact, the best choice for all-around performance. Keeping the frequency at 3600 MHz meant that the FCLK and UCLK frequencies were maxed out at 1800 MHz, respectively.
G.Skill offers a wide range of memory configurations within the Neo product line, but they don’t offer anything above 3600 MHz. Given the overclocking results we examined today, we feel that G.Skill made a very wise decision for AMD optimized memory lineup. The memory we tested today comes in at just $169.99, which is a reasonable price considering the performance and Rysen 3 compatibility. For the serious users out there, G.Skill offers the Neo in a kit running 3600 MHz with CL14-15-15, priced at $279.99.
David Miller – mllrkllr88