ADATA Technology is widely known for DRAM Modules, desktop and notebook PCs, Solid State Drives, External Storage (HDD, SSD, Enclosures), USB Flash Drives, and other media/memory related accessories. In recent years, they have gained recognition as one of the global leaders of top-tier DRAM products. With annual earnings of $680 million USD in 2017, they were ranked number two out of all DRAM sellers globally (Taipei Exchange October 2017). When one of the world leaders in DDR4 memory offered their product for torture testing, we jumped at the opportunity.
Up for review today is a new kit of memory from ADATA. Today we are looking at the Spectrix D41 series of memory. This is a refresh of their Spectrix D40 line, which has been a popular choice for enthusiasts, overclockers, and computer builders alike who are looking to add RGB memory to their system. It’s still the same great company and memory that many of us have grown to appreciate, but this version of their Spectrix line has undergone a modernized face-lift. In particular, the Spectrix D41 shows off that RGB shine with a less obstructed top-down view compared to the Spectrix D40, so you get more RGB shine for your system.
Specifications and Features
As you may or may not know, ADATA was founded in 2001. Shortly after the release of DDR1 in 2000, ADATA has been a sought-after memory provider for overclockers and computer enthusiasts. They are a Taiwanese-based company that is publicly traded.
The ADATA XPG Spectrix D40 DDR4 is manufactured in China and is widely available at retailers across the globe and carries a lifetime warranty. With modern DDR4 there are two main semiconductor players that provide ICs to companies like ADATA, which include SK Hynix and Samsung. Based on the specifications of this memory, we would expect to see Hynix ICs under the heat spreader. While the actual IC itself may be re-branded to ADATA, this particular kit of memory does indeed utilize Hynix ICs.
The specifications and features below come directly from the ADATA website.
|ADATA XPG Spectrix D41 Specifications|
|Capacity||16 GB (2×8 GB)|
|Tested Frequency||DDR4-3000 (1500Mhz)|
|Kit Type||Dual Channel|
|Pricing||$159.99 at Newegg.com|
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 utilizing Hynix’s M-Die. This memory uses a total of 8 ICs, all of which are located on only one side. To an attentive observer, one can speculate that other kits of memory in the Spectrix D41 lineup may contain different memory ICs such as Samsungs infamous B-DIE. If you would like to check out other offerings in the Spectrix D41 lineup you can download a PDF here: Datasheet XPG SPECTRIX D41 DDR4 RGB Memory.pdf
The packaging is often an overlooked element to a finished product, but it shouldn’t be. The packaging is simply a delivery mechanism to ship the product safely, but it also serves as a preview of what’s to come. We have all opened products with cheaply-made packaging, and whether you think about it or not, I believe that it has an effect on the first impression of the product.
The Spectrix D41 packaging does not disappoint. The box is multi-colored, printed cardboard and includes a flip-up top that boldly displays the key features. Opening the top also give the owner a first glimpse of the product. Inside the box, the memory is housed in a plastic clam-shell that protects the RAM and is large enough to house 4 sticks safely.
Of the two color options available, we are reviewing the Tungsten Gray variety today. This might be the obvious choice for more computer builders since it is more of a universal color tone compared to the Crimson Red option. As mentioned above, this version of Spectrix offers more visible RGB. The entire top of the memory module is one contiguous plastic light diffuser. This serves the function of softening the light that comes from each individual LED and makes the light appear more uniform. The heatsink does not protrude into the RGB diffuser.
As seen below, the Spectrix D41 puts on a very unique and very bright light show. Booting up the system we find that the memory comes from the factory with a pre-programmed, rainbow-like fade effect. Each module cycles through all color options as the color shifts from one end of the stick to the other.
The RGB control software for the Spectrix D41 is free and available for download from their support website. The software is easy to use and is not showy or burdened with unneeded content. I commend them for having the courage to produce software which is extremely streamlined and simple, yet highly effective. Within the software, you will find more options than you thought possible. The combinations of lighting configurations are nearly limitless and the memory changes colors as soon as you hit ‘apply’. The software is compatible with all mainstream motherboards.
The Spectrix D41 is compatible with Asus Aura software. This unique feature allows users with Asus Aura motherboards to sync both the memory and the motherboard. When paired with an Asus motherboard, the ADATA XPG control software relinquishes control to the Asus Aura software to allow syncing of the various compatible hardware.
One noteworthy feature is that this memory has individually addressable RGB LEDs. This means that with the aforementioned software, the user has the ability to control each of the 5 LEDs separately from each other.
Testing and Overclocking
The overall objective is to evaluate the memory under a variety of different conditions in an effort to simulate daily life. To accomplish this task, We will turn to benchmark programs to examine the performance of the memory and overall system under various conditions. The approach is to first test the Intel XMP profiles. The XMP profiles are built-in to the memory and they are expected to work flawlessly across all motherboards. Once I have established that the XMP profiles are working on the test system, then I will take a look at how these memory modules overclock.
For the purposes of this review, I will examine the overclocking potential without expressly voiding the warranty due to over-voltage. 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 less than 1.50 V.
Below are the test system and resulting memory speeds that will be used to evaluate the memory and run the benchmarks.
|CPU||Intel i7 7740X @ 4.0 GHz (4.7 GHz Cache)|
|Cooler||Alphacool Eisblock XPX CPU Block with custom radiator loop|
|Motherboard||ASRock X299 OC Formula|
|Graphics Card||Asus 980Ti Matrix|
|Solid State Drive||Team Group L5 120GB|
|Power Supply||Seasonic 1200W Platinum PRIME|
|Operating System||Windows 10 x64|
|Memory Speeds Compared|
|Intel XMP 2 ~ 2666 CL16-16-16 1.20 V|
|Intel XMP 1 ~ 3000 CL16-18-18 1.35 V|
|Overclock ~ 3200 CL15-16-16 1.35 V|
|Overclock ~ 3333 CL15-16-16 1.40 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 I use two primary tools which include AIDA64 and ASRock Timing Configurator. AIDA64 is a powerful system diagnostic and benchmarking tool, that can be purchased for a reasonable price. Next, I will be using the ASRock Timing Configurator, which is a free piece of software that allows users of all major motherboard brands to see the primary, secondary, and tertiary timings which have been applied in the bios.
Below is the XMP profile 2. Of the two XMP profiles which are stored in this memory kit, profile 2 is the one which is less stressful for your motherboard to operate. The XMP Profile 2 should always be considered the failsafe profile because it operates the memory at lower frequencies than they were intended to run. The operating voltage of this profile is 1.20 V.
Below is the XMP Profile 1. This is the profile that most enthusiasts and computer builders should choose to select. This profile sets the optimal frequency and latencies which the memory is intended to run at. The operating voltage of this profile is 1.35 V.
Due to the fact that this kit of memory utilizes Hynix ICs, I expect that the overclocking potential will be substantially less than kits composed of Samsung ICs. That being said, this memory still has plenty of overclocking headroom and overall system performance is expected to improve with overclocking.
My methodology is to set my maximum working voltage of 1.490 V and see what could be accomplished. Regardless of the fact that the ASRock X299 OC Formula is one of the best motherboards for memory overclocking, it soon became apparent that this kit of memory was not able to achieve stable frequencies other than 3333 MHz. While testing, I was able to achieve speeds of 3433 MHz, but sadly that speed was not stable in all benchmarks. Once I knew the highest frequency that was passable, I spent some time to figure out latencies and other frequency/voltage combinations that resulted in stable memory.
Below is the resulting overclock from using the same voltage as XMP 1 (1.35 V) however this substantially overclocked compared to XMP 1. Not only is it running 200 MHz higher, but the latencies are also heavily reduced to increase overall performance.
Below is the resulting overclock with the optimal latencies and the maximum operating frequency this memory is capable of. A noteworthy point is that to achieve this stable maximum overclock the RAM was only 0.05 V above the XMP. The resulting stable voltage for the overclock is 1.40v
Just to slake my own curiosity, I decided to void the warranty and see what would happen if I applied 1.75 V or higher. This humble kit of memory had no problems running higher voltages such as 1.75 V, but regrettably, it did absolutely nothing for frequency or latency overclocking headroom. I would not recommend any voltage higher than 1.40-1.45 V on this kit of memory as it did not yield better results.
Benchmark programs allow us to examine the performance gain or loss from making changes to the various computer components such as the CPU, GPU, or memory. In this case, the goal is to remove all variables and examine only the performance change from memory overclocks. To accomplish this, all benchmarks were run with the CPU held at a constant 5 GHz for every benchmark. I will investigate the overclock potential not just because it’s fun, but because it has real-world implications and can help with the productivity of daily tasks.
First up, I used AIDA64 Cache and Memory Benchmark. Specifically speaking, “Memory bandwidth benchmarks (Memory Read, Memory Write, Memory Copy) measure the maximum achievable memory data transfer bandwidth. The code behind these benchmark methods are written in Assembly and they are extremely optimized for every popular AMD, Intel, and VIA processor core variants by utilizing the appropriate x86/x64, x87, MMX, MMX+, 3DNow!, SSE, SSE2, SSE4.1, AVX, AVX2, and AVX-512 instruction set extension.” (source: www.aida64.com)
In the graph below, it is clearly visible that each of the four different memory speeds compared had a noticeable improvement in the benchmark result. As the frequency goes up, or the latency becomes faster, the resulting score has a nearly linear progression of performance.
Next up is my personal favorite benchmark. The benchmark 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. Specifically speaking, “Geekbench 3 features new tests designed to simulate real-world scenarios. This helps make Geekbench an invaluable tool to determine how your current computer (or your next computer) will handle your tasks and applications.” (source: www.geekbench.com)
In the graph below we see that the memory score is similar to the AIDA64 results as it progresses nearly linearly with increased memory speeds. It’s worth noting that although the memory is performing faster, the various memory speeds tested show a negligible 2.5% gain over XMP 2 in overall system performance.
The next benchmark is fairly new to the overclock world, and it was made expressly for the purposes of competitive overclocking hosted at hwbot.org. I chose this benchmark because it has been known to yield noticeable improvements in the overall score from memory overclocking. Specifically speaking, “HWBOT x265 Benchmark is based on the open source x265 encoder. It can take advantage of modern CPUs instruction sets and multithread support is also very good. However, this benchmarks is also capable of running even on old processors such as the AMD Athlon or Intel Pentium III. Of course on the legacy hardware, the encoding time is rather long. There are two presets available – 1080p and 4k. The main goal of both of them is to convert H264 source video to H265/HEVC and measure average fps.” (source: hw-museum.cz)
By looking at the four memory speeds compared in the graph below, it’s clear to see that memory speed has a small overall effect on the total system performance. This benchmark yields the best results from using lower latencies such as CL12 which can be accomplished using Samsung ICs.
The next benchmarks we will examine are ones centered around 3D rendering and games. The Futuremark 3D Mark suite of benchmarks is a real-time graphical rendering benchmark that also contains an element of memory and CPU testing. For each of the benchmarks from Fire Strike to Time Spy, we will only examine the CPU/memory testing portion and disregard the graphical test elements.
In the graph below we can see that in both cases there were slight gains from memory overclocking. With a gain of nearly 6% over XMP 2 in the Time Spy CPU test, the improvement is small but memory does indeed play a role in the score.
Overall, the ADATA XPG Spectrix D41 kit of memory did well across all memory benchmarks. In every case, the overclocking results above XMP were noticeable but indeed quite small. While AIDA64 showed impressive gains of up to 25% over XMP, the real world benchmarks such as Geekbench 3 show us that the overall performance gain is around 2%. To some, it might be disheartening that overclocking results showed such small real-world performance gains, but the selling point is that this performance gain was done with a tiny increase of voltage beyond XMP 1.
The story of this memory is best told not in terms of it’s overclocking prowess but in terms of its style and compatibility. The RGB diffuser is what makes this memory special. There are very few kits of memory I have seen that show a truly unobstructed top-down view of the RGB lighting. The unprecedented RGB lighting coupled with a total of 5 individually addressable LEDs per module makes this an obvious choice for those looking to maximize the RGB lighting of their system.
The ADATA XPG Spectrix D41 would not be the first choice for serious overclockers looking to maximize performance. However, the lifetime warranty, exceptional RGB lighting, and attractive price of $159.99 would make this an excellent choice for gamers, computer enthusiasts, and system builders alike. Overclockers Approved!
David Miller – mllrkllr88