• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

FEATURED Team Group Dark Z FPS 16GB DDR4-4000 CL16 - TDZFD416G4000HC16CDC01

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Woomack

Benching Team Leader
Joined
Jan 2, 2005
Team Group released only one memory kit from the Dark Z FPS series. So far it's a 2x8GB DDR4-4000 CL16-18-18 1.45V kit which already looks pretty good. As expected, it's using Samsung B IC. There are no LEDs or any other flashy things. The heatsinks are quite standard but have a specific design. You may like it or not but the most important is how the memory works.

Some manufacturer's info:

Module Type 288 Pin Unbuffered DIMM Non ECC
Frequency 4000
Latency CL16-18-18-38
Capacity 8GBx2
Data Transfer Bandwidth 32,000MB/s (PC4 32000)
Voltage 1.45V
Dimensions 43.5(H) x 141(L) x 8.3(W)mm
Heat Spreader Aluminum heat spreader
Warranty Lifetime warranty

The Dark Z FPS memory has one XMP profile, programmed to work at DDR4-4000 CL16-18-18 1.45V. It actually works fine on AMD and Intel motherboards but because of memory controller behavior, it seems better for Intel chipsets. I will perform tests on both platforms and mix some results so please read the description or simply check what platform is on the screenshot.

The memory uses Samsung B-die IC which is so popular that I guess I don't have to add more. What's important is that the Dark Z FPS uses higher binned IC so also overclocking results are expected to be better.
Below is how the Thaiphoon Burner software sees the RAM.

TG_FPS_FTPICss2.jpg



Product photos:



Stability at rated speed:

TG_FPS_FTPICss1.jpg
 
All results are to show how high this specific memory kit can overclock. Sub-timings are not tweaked optimally. It shows more regular users how they can set it without bigger problems.

Longer tests were performed on the pretty average Ryzen 9 5900X and MSI X570 Unify.


XMP: DDR4-4000 CL16-18-18-38 1.45V

XMP_FPS.jpg


DDR4-3600 CL12-13-13-28 1.65V

3600c12.jpg


DDR4-4133 CL14-15-15-34 1.60V

4133c14.jpg


DDR4-4400 CL16-17-17-34 1.60V

4400c16.jpg


I'm not sure why but I forgot to save DDR4-4800 CL18-18-18 1.55V results. The DDR4-4800 is actually the maximum possible on my X570 Unify motherboard but the same on the Z590I Unify (below).


Here are additional AIDA64 screenshots made on Intel. All these settings passed longer tests, I simply didn't save all screenshots:

DDR4-3733 CL13-13-13 1.65V Results should be better as long as you can run it with Gear 1. My setup can't :(

TG_FPS_FTPICoc1.jpg

DDR4-4200 CL14-16-16 1.60V

TG_FPS_FTPICoc2.jpg

DDR4-4800 CL18-18-18 1.55V

TG_FPS_FTPICoc3.jpg

At least on my motherboards, the Dark Z FPS RAM doesn't want to work at anything above DDR4-4800. Actually, many Samsung B kits have problems with that, lower or higher binned. Optimal for most users are still lower frequencies at tight timings so it's not an issue.
 
This one seems like the best Samsung B kit that I had for some longer. Likes higher voltages and overclocks well at lower. It also likes tighter timings at still reasonable voltages. Works well on AMD and Intel while some of my other Samsung B kits are not overclocking so high on AMD.

Just one thing. On Intel, I could set about 1-2 ratios higher at the same voltages and timings. For example, while on AMD was 4400 CL16 then on Intel was 4600 CL16. 5000 should work at CL19/20-22-22 1.60V but Samsung B generally doesn't like 5000+.
 
On the new 11000 Intel would you go for the tighter timings at 4000 or push for the looser timings as close to 5000 as I can get?

My personal goal is to get to at least 4400 with this set.
 
As you can see on the screenshots, with a higher clock (even at more relaxed timings) latency is also going down so generally the higher clock, the better results. I would try for DDR4-4400/4600 and CL16 + play with sub-timings. This should give about 65-70GB/s and ~48ns latency.
 
Dear @Woomack, thank you for the review on this kit. However, I have a question. What do you think about this kit performance against the Crucial Ballistix MAX RGB 4400 MHz DDR4 DRAM Desktop Gaming Memory Kit 16GB (8GBx2) CL19 and the Patriot Viper Steel DDR4 16GB (2x8GB) 4400MHz PC4-35200 Dual Memory Kit - PVS416G440C9K. I am building a new system with Ryzen 5800x and x570 Aorus Master v1.2 Revision. End goal is to overclock both the CPU and the Ram for optimal performance. ( Will be using DH-15 for cooling )
I see the review for the Crucial Ballistix on only with Intel CPU and no thread regarding this kit. Noting this is a Micron E die. The Viper Steel DDR4 review is almost a year old with a previous gen CPU.
Would you suggest the Teamforce FPS kits 4000Mhz over these two kits.

Your output on this would be highly appreciated.
 
I'll wait for Woomack to answer your other questions but wanted to chime in on the DH-15 cooler. I am using this Team Group ram with a DH-15 and had to switch out the intake fan for a 120mm because the stock fan hits the ram sticks. The other ram sets are probably just as tall so be prepared for that either way you go.
 
I'll wait for Woomack to answer your other questions but wanted to chime in on the DH-15 cooler. I am using this Team Group ram with a DH-15 and had to switch out the intake fan for a 120mm because the stock fan hits the ram sticks. The other ram sets are probably just as tall so be prepared for that either way you go.

Dear @BugFreak, thank you for your insight on NH-D15. I am well aware of the limitations when its come to RAM clearance with this cooling fan unit. I intend to swap the Noctua fans with Deepcool CF 120mm fans for Ram clearance. I am currently using Deepcool Mattrexx 55 V3 ARGB case which had Cooling fan clearance of 168mm. My current setup is with NH-D14 which gives me a clearance of 8mm. With NH-D15 i will have fan clearance of 3mm. I know it will be a tight fit however that is how it is considering 140mm fans on the NH-D14. The Noctua NH-D15 fans will be installed on the top mount for more optimal airflow. Thanks again for your input on this. Much appreciated. Waiting for @Woomack 's thoughts on this. Mind me asking which Team Group are currently using on your setup and the Ram height for my knowledge.
 
I would recommend getting 2x16GB DDR4-3600 14-15-15 or 16-16-16 like one of the other Team Group series, Dark Pro or Xtreem. This is because the 1:1:1 ratio will be faster than anything up to ~DDR4-4600 on Ryzen, and 2x16GB is generally dual-rank what helps in performance in some games. The only single-rank 16GB modules are Microns in 2x16GB Ballistix MAX kits.

The Dark Z FPS has a higher binned Samsung B IC which supposed to overclock better than the Patriot Steel 4400 (which is Samsung B too). Ballistix MAX 4400 is Micron. It won't be faster because of AMD architecture. To take advantage of higher frequency memory kits, you will need new Intel with a good memory controller. AMD runs the best between DDR4-3600 and DDR4-3800. Some CPUs can handle up to DDR4-4000 at a 1:1:1 ratio but not many. Most perform the best closer to DDR4-3800.
 
I would recommend getting 2x16GB DDR4-3600 14-15-15 or 16-16-16 like one of the other Team Group series, Dark Pro or Xtreem. This is because the 1:1:1 ratio will be faster than anything up to ~DDR4-4600 on Ryzen, and 2x16GB is generally dual-rank what helps in performance in some games. The only single-rank 16GB modules are Microns in 2x16GB Ballistix MAX kits.

The Dark Z FPS has a higher binned Samsung B IC which supposed to overclock better than the Patriot Steel 4400 (which is Samsung B too). Ballistix MAX 4400 is Micron. It won't be faster because of AMD architecture. To take advantage of higher frequency memory kits, you will need new Intel with a good memory controller. AMD runs the best between DDR4-3600 and DDR4-3800. Some CPUs can handle up to DDR4-4000 at a 1:1:1 ratio but not many. Most perform the best closer to DDR4-3800.

Dear @Woomack Thank you very much for your recommendations. Just a quick Q you mentioned about the 1:1:1 ratio, the dark z fps as per your comments "This one seems like the best Samsung B kit that I had for some longer. " Wouldnt this be an ideal candidate for the build ?
 
If you are searching for a 2x8GB kit then yes. However, 2x16GB is generally faster on new platforms because is dual-rank. The difference depends on the software and can be between ~0-7%. For example, some games show +5-10FPS more on dual-rank memory and Ryzen 5000 or Intel 10/11 Gen CPUs. On the other hand, in most cases, it's 5-10FPS more when is already 100FPS+ ;)
I see it this way. If you only play games and don't use additional software then 16GB is fine. When you run multiple web browser tabs + other soft while playing games, then 16GB is sometimes not enough. When you don't play games but use audio/video/photo editing software then 16GB is usually not enough.
If you are planning to keep this build for 2-3 years then I recommend getting 32GB of RAM. If you are planning upgrades within' a year then get 16GB and save money on something else.
 
@Woomack thank you for your info on this. I have bought this kit as it was cheaper when looking at other kits @ 4000mhz cl16. However, I need your help. All your overlocks have been 1.6v - 1.65v. Were you able to do anything at 1.5v - 1.55v ? If so could you post your zen timings or attach the settings you opted for testing this kit.
 
Most 1.60V settings could work at 1.55V. 4200 CL14-16-16 could work at 1.55V or at 4266 14-16-16 1.55V (tested later). 4800 CL18-18-18 works at 1.55V. Probably 4600 CL16-17-17 will work at 1.55V too.
In short, you have some examples so can try on your own as your kit can overclock better or worse. It's not a rule. Just check desired frequency at 1.50-1.55V and play with timings like 14-14-14, 14-15-15, 14-16-16, 16-16-16, 16-17-17, 16-18-18, 18-18-18, ... If you won't make something like 4266 CL14-16-16 work then it will probably run at CL15 or 16 at 1.5V.
I won't test this memory kit anytime soon so just giving advice. This kit can work at a very wide range of frequencies and timings and it's not guaranteed that everything will work the same on a different motherboard.
 
Most 1.60V settings could work at 1.55V. 4200 CL14-16-16 could work at 1.55V or at 4266 14-16-16 1.55V (tested later). 4800 CL18-18-18 works at 1.55V. Probably 4600 CL16-17-17 will work at 1.55V too.
In short, you have some examples so can try on your own as your kit can overclock better or worse. It's not a rule. Just check desired frequency at 1.50-1.55V and play with timings like 14-14-14, 14-15-15, 14-16-16, 16-16-16, 16-17-17, 16-18-18, 18-18-18, ... If you won't make something like 4266 CL14-16-16 work then it will probably run at CL15 or 16 at 1.5V.
I won't test this memory kit anytime soon so just giving advice. This kit can work at a very wide range of frequencies and timings and it's not guaranteed that everything will work the same on a different motherboard.

Thank you @Woomack. Much appreciated.
 
Back