Today we’re taking a look at Adata’s XPG Spectrix D60G DDR4 memory. Sporting dat sweet sweet RGB in a sleek stylish heat sink, how does this memory compare to my tried and true non-RGB G.Skill Trident Z memory with similar specs? Let’s find out.
Specifications for the Adata XPG Spectrix D60G are relatively straightforward.
- Price $149.99 (Newegg)
- Model Name DDR4-3200
- Speed PC4-25600
- Size 8GB x 2
- Latency CL16-18-18-38
- Voltage 1.35V
- Heat Sink Color Tungsten Grey
A Note on Clocks and Timings
But before I move forward, it’s important to understand just what memory timings and latency actually are. At the risk of becoming overly derivative and increasingly granular, look at this as a higher level explanation. In general, higher speed plus lower timings are better. In general.
Data is stored in a row x column “grid” within memory. The timings refers to how many clock cycles it takes to access data in a specific location in this grid in nanoseconds. This is different to memory speed or clocks, which in this case, is 3200 MHz when XMP is enabled (more on XMP in a minute). Memory speed tells you how many transfers per second the memory can perform.
You’ll typically notice four numbers when discussing timings, though recently, it’s only three. For the Adata XPG Spectrix D60G, the timings at XMP are 16-18-18-38. Techquickie actually has a great explainer of this, which I’ll summarize here.
The first number is CAS Latency (Column Address Strobe) or the time it takes to start responding to a request for data. Keep in mind, slower clocked RAM paired with lower CAS Latency can be faster than higher clocked RAM with higher CAS Latency.
The second number is tRCD, or Row Address to Column Address Delay. Remember, data is stored in a row x column “grid” in memory. This refers to the delay between row and column access.
The third number is tRP, or Row Precharge Time. This refers to the latency when opening a new row within memory. Finally, the fouth number is tRAS, or Row Active Time. This refers to the minimum amount of clock cycles this specific row needs to stay “open” in order for the data contained within to be written and read properly.
XMP, or Extreme Memory Profile, is an auto-overclock of sorts which adjusts clocks and timings to improved settings. Note, you can feasibly adjust beyond these, but just be careful of any artrefacting which may occur.
There’s a lot more to RAM timings and speeds than what I’ve covered here, but this should suffice for our understanding. Gamers Nexus has a brilliant write-up which I encourage you all to read.
Design
I love the way the Spectrix D60G modules look when plugged into my system. They’re not overly designed, yet sport some great angled lines along the heat sink
Of course, the literal highlight of the design is the RGB. Some may hate the RGB craze, but I love it. I love how the RGB is along almost the entirety of the heat sink, really showcasing the lines and curves of the RAM modules.
The surface area of the RGB is a lot larger than other RAM modules I’ve seen, yet it isn’t garrish. It’s hardly subtle, rather, feels respectful of the RGB lighting in the rest of my rig. I love it.
Testing
I conducted productivity and gaming tests with both my G.Skill Trident Z modules and the Adata XPG Spectrix D60G modules. I tested the Adata at both stock and XMP, while testing the G.Skill at XMP only. The goal was to understand how the G.Skill modules at XMP compared to the Adata modules at both stock and XMP.
Productivity included Sandra for both bandwidth and cache latency, plus 7-Zip for compression. Games included GTA V, Assassin’s Creed Odyssey, and Shadow of the Tomb Raider. Please see the considerations below, including clocks and timings for both the G.Skill and Adata modules:
The test bench used included the following:
Synthetic and Productivity Results
First up, memory bandwidth results:
My old G.Skill memory sits between Adata stock and Adata XMP here at 29.53 GB/s. The Adata at XMP turned in 30.53 GB/s, however, with such a small margin between the G.Skill at XMP and Adata at XMP, their performances are nigh on similar.
Here again, the Adata at XMP performed similarly to the G.Skill at XMP with only a tenth of a nanosecond between the two. Adata at stock turned in slower results at 20.6 ns.
Finally, I ran a 7-Zip test compressing 2.5 GB of .mp4 video. Here, the Adata at XMP significantly outperformed the G.Skill at XMP by six seconds. In fact, the Adata at stock was only one second slower than the G.Skill at XMP.
Gaming Results
Gaming results were mixed when compared to the G.Skill at XMP.
The largest gain was in GTA V, with the Adata at XMP posting a 2.7% increase in performance when compared to the G.Skill at XMP. However, performance was nearly identical to the G.Skill in both AC Odyssey and Shadow of the Tomb Raider.
This shouldn’t really come as a surprise, given how similar these two sets of memory are to one another. At XMP, they boast nigh on identical clocks and timings, with the sole exception being the slightly improved tRAS timing on the Adata over the G.Skill.
Conclusion
Even though my performance with the Adata XPG Spectrix D60G modules was similar to that of my G.Skill Trident Z, I actually prefer the Adata purely because of the RGB. Design does matter to me, and the lines, curves, and overall implementation of the RGB on the Adata XPG Spectrix D60G really impressed me.
Moreover, I didn’t lose performance when switching from the G.Skill to the Adata. At $149.99, the Adata XPG Spectrix D60G may be a bit pricey for some, especially considering the G.Skill’s far more affordable $84.99 price tag for nearly identical performance.
However, things like design matter quite a lot to me. And while I wouldn’t necessarily recommend someone on a budget to spend $65 more on nearly identical RAM just for “some RGB,” if you are in the market for good quality performant 16 GB modules, you can’t really go wrong with the Adata XPG Spectrix D60G.
Pros
- Extremely stylish
- Beautiful RGB
- Performant
Cons
- Not budget-friendly
The product described in this article was provided by the manufacturer for evaluation purposes.