Team Group Delta RGB DDR4-3000 16GB Review

Why you can trust Tom's Hardware Our expert reviewers spend hours testing and comparing products and services so you can choose the best for you. Find out more about how we test.

Benchmark Results & Final Analysis

Delta RGB supported slightly tighter timings under manual configuration compared to the company’s Night Hawk RGB, but we feel that was probably luck-of-the-draw, since both sets came with the same ratings. Same-rating modules from a single brand are usually identical underneath the heat spreaders.

Swipe to scroll horizontally
 DDR4-3200DDR4-2666DDR4-2133
Team Delta RGB 16GB TF4D416G3000HC16CDC0116-17-17-34 (1T)13-14-14-28 (1T)11-11-11-28 (1T)
Team Night Hawk RGB 16GB TF2D416G3000HC16CDC0116-18-18-36 (1T)13-15-15-30 (1T)11-12-12-28 (1T)
Gskill Trident Z RGB 32GB F4-3600C16Q-32GTZR15-16-16-32 (1T)12-12-12-28 (1T)10-11-11-28 (1T)
Mushkin RL Ridgeback 16GB MRB4U320LLLM8GX218-19-19-38 (1T)15-16-16-32 (1T)12-13-13-28 (1T)

The Delta RGB also overclocked better than the Night Hawk RGB, nearly catching up to the two-module O/C of G.Skill’s DDR4-3600. We were able to reach the CPU’s DDR4-3733 ratio at full stability, but any additional tuning via BCLK eventually caused errors in Prime95.

The bandwidth numbers for G.Skill’s DDR4-3600 at XMP settings might have some readers wondering if that higher data rate can really make this much difference, and the answer is no. As with the Skylake core that came before it, the memory controllers of Kaby Lake processors have a strong preference for four-bank configurations. The Trident Z got there by having four modules, but our previous review of Team Group double-sided modules shows a two-DIMM alternative.

Identical rated (XMP) timings made us expect identical Sandra memory latency results at XMP default, but the Delta RGB started a little slower there. Lower achievable timings helped it pass its Nigh Hawk sibling under DDR4-3200, 2666, and 2133 manual configurations.

Sensitive to both bandwidth and timings, F1 2015 shows good performance for all three 16GB sets. None of these can keep up with G.Skill’s 4-DIMM configuration, particularly when it’s clocked to DDR4-3600.

Metro is only slightly sensitive to memory performance, requiring large differences in specifications to produce noticeable differences in framerate. It’s also a better representation of the “general gaming experience” concerning memory performance, compared to the EGO game engine of the F1 series.

Blender is also slightly sensitive to minor changes in DRAM performance. It strangely shows the Delta RGB lagging at all settings.

Like the F1 racing series, 7-Zip file compression shows noticeable performance gains for faster RAM. Delta RGB beats its two-single-rank-DIMM competitors by completing the task in less time, though once again the four-DIMM set shows that four ranks offers greater improvement.

Since our benchmarks are optimized for 12GB of RAM, all of the 16GB sets offer similar performance-per-dollar, and the 32GB set, with twice as many modules, falls behind.

Comparing the above results to differences in capacity gives us a true inkling of the performance value per DIMM, since it corresponds to the performance value per gigabyte in a comparison of 8GB models. The Trident-Z is hyper-priced to match its hyper-speed DDR4-3600 data rate. Meanwhile the Delta RGB’s strong overclock puts it in the value lead. Those who don’t overclock might find better value in Team Group’s Night Hawk RGB.

The above value chart is a little peculiar because of scarce availability for Delta RGB, where we were forced to rely on MSRP rather than street pricing. We had a similar conundrum concerning the Night Hawk series, which only achieved reasonable availability after being reviewed. And Night Hawk has since dropped in price by around $2, and that $2 difference was just enough to give it the win.

Because Night Hawk RGB provided RGB capability at very little cost, we gave it our stamp of approval. Our Delta RGB samples overclocked a little better and are similarly priced, so they probably deserve that same honor. Meanwhile, we’ll continue looking for a break-away product worthy of an even greater honor.


MORE: Best Memory


MORE: DDR DRAM FAQs And Troubleshooting Guide


MORE: All Memory Content

Thomas Soderstrom
Thomas Soderstrom is a Senior Staff Editor at Tom's Hardware US. He tests and reviews cases, cooling, memory and motherboards.
  • AgentLozen
    I see this got a Tom's Approved award. Is there a place where you guys describe what these awards mean and what variables are factored into them? Something like:

    Editor Approved: This product doesn't suck at the very least.
    Editor Recommended: This product is better than many of it's rivals.
    Editor's Choice: This product is the Bernie Sanders of the computer world.
    Reply
  • shrapnel_indie
    20165646 said:
    I see this got a Tom's Approved award. Is there a place where you guys describe what these awards mean and what variables are factored into them? Something like:

    Editor Approved: This product doesn't suck at the very least.
    Editor Recommended: This product is better than many of it's rivals.
    Editor's Choice: This product is ...

    Let me correct this:

    Editor Approved: It meets expectations: it doesn't suck.
    Editor Recommended: This product is better than many of it's rivals.
    Editor's Choice: This product has no equal in performance for cost. PR/Advertising hasn't made false promises.

    More officially, I think they did specify in better detail at one point in time.
    Reply
  • Crashman
    20165859 said:
    20165646 said:
    I see this got a Tom's Approved award. Is there a place where you guys describe what these awards mean and what variables are factored into them? Something like:

    Editor Approved: This product doesn't suck at the very least.
    Editor Recommended: This product is better than many of it's rivals.
    Editor's Choice: This product is ...

    Let me correct this:

    Editor Approved: It meets expectations: it doesn't suck.
    Editor Recommended: This product is better than many of it's rivals.
    Editor's Choice: This product has no equal in performance for cost. PR/Advertising hasn't made false promises.

    More officially, I think they did specify in better detail at one point in time.
    Thanks!


    Reply
  • Zaporro
    Ughh if only not the white casee, do they come in black?
    Reply
  • Crashman
    20166373 said:
    Ughh if only not the white casee, do they come in black?
    Yes, but we reviewed the white ones.

    Reply
  • Zaporro
    20166410 said:
    20166373 said:
    Ughh if only not the white casee, do they come in black?
    Yes, but we reviewed the white ones.

    yesssssssssssssss
    Reply
  • James Mason
    This article needs a video/60fpsGIF/Webm of the RGB rainbow/color shifts in action. The picture of the static rainbow isn't really impressive, but when you see for example the g.skill trident z RGBs actually changing colors it looks really cool.
    Reply
  • FritzEiv
    20167349 said:
    This article needs a video/60fpsGIF/Webm of the RGB rainbow/color shifts in action. The picture of the static rainbow isn't really impressive, but when you see for example the g.skill trident z RGBs actually changing colors it looks really cool.

    That's a really good idea. Crashman: Let's do this for the RGB things!

    Also:
    Editor Approved: It meets the expectations for the category. (Someone said "it doesn't suck," but I might say instead that it gets a passing grade.)
    Editor Recommended: Not only does the product do well, but it also is a good value for the category.
    Editor's Choice: Pricing or other considerations aside, this is a best-in-class for this category.

    Some editors are tougher graders than others, but I encourage everyone to think these 3 awards through, to rarely give the top 2 unless really warranted.
    Reply
  • chaosmassive
    PRO : Configurable via Asus Aura software
    CON : Configurable via Asus Aura software

    seems legit
    Reply
  • juanrdp
    CON : Configurable ONLY via Asus Aura software
    Reply