FEATURED Team Group Dark Pro 8Pack Edition 2x16GB DDR4-3600 CL16 - TDPPD416G3600HC16GBK

[Woomack]

Team Group Dark Pro 8Pack Edition 2x16GB DDR4-3600 CL16 - TDPPD432G3600HC16GBK

Today I have a bit unique memory kit as it's marked as the 8Pack Edition while the famous overclocker 8Pack is out of the scene for some longer. Either way, there is no Dark Pro memory kit listed on the Team Group website above DDR4-3466 while the tested memory kit is rated at DDR4-3600 CL16-16-16 1.35V. What makes it special is that the kit contains two 16GB memory modules and is also available at a reasonable price but the availability is limited to the overclockers.co.uk online store.
With the higher demand on RAM, I decided to purchase this kit for my daily PC. I often run a couple of applications, multiple website tabs, and a game at the same time so sometimes I'm passing 16GB of RAM.

Back to the memory itself. As it was expected, it's using Samsung B-die IC so if we wish to overclock and keep optimal performance then it's still the best IC around, even though not manufactured anymore.
The memory has no RGB lighting what more often is an advantage. I guess that all that RGB stuff is way too popular and many users have enough of that. Manufacturers started to focus on the lighting instead of improving performance or compatibility.

Below are some photos of the memory kit. It's black so perfectly fits my next build with the MSI Z490I Unify.

Spoiler

The Dark Pro has one XMP profile at 3600 16-16-16-38 1.35V. Officially there is a voltage range up to 1.45V but on the label, we will see 1.35V. Of course, there are no problems at XMP settings.

Spoiler

I'm not sure if it's a matter of early BIOS or something else but the memory overclocks great up to DDR4-4133 while above that can't even boot. Also scaling with the voltage is a bit weird and looks like:
- 3600 1.35V
- 3800 1.40V
- 4000 1.45V
- 4133 1.50V
Below these voltages, memory is not even booting, regardless of timings and other settings. For example, the memory can boot at 4000 CL14 1.45V but can't at 4000 CL19 1.40V. Also, I'm not sure if it's a matter of BIOS or the memory itself as I had no time on more detailed tests on other motherboards.

What you can count on is:
3200 CL13-13-13 1.40V+
3200 CL14-14-14 1.35V
3600 CL14-14-14 1.45V
3600 CL16-16-16 1.35V
3800 CL16-16-16 1.40V
4000 CL16-16-16 1.45V
above is still in tests and the highest stable setting so far is 4133 CL18-18-18 1.50V what doesn't look any special.

The tightest at DDR4-3600 is CL12-12-12-28 1N 1.6V. It wasn't stable but I wasn't expecting that. Here is the AIDA64 screenshot.



All tests were performed on the i5 10500 CPU, MSI Z490I Unify motherboard and MSI GTX1660Ti graphics card. Nothing really special but enough for some good results.

 

[Woomack]

Overclocking results

3200 CL14-14-14 1.35V

Spoiler

3600 CL14-14-14 1.45V

Spoiler

4000 CL16-16-16 1.45V

Spoiler

4133 CL18-18-18 1.50V

Spoiler

I will try some more when MSI releases a new BIOS as I feel that the motherboard is limiting the results. I may also check some other motherboards but have some problems with the time recently so we will see.
For sure this memory kit is a keeper and highly recommended if you are looking for a 32GB memory kit that overclocks well and even at XMP settings delivers exceptional performance.

 

[rrul]

Hey thanks, nice review.

Thinking of buying it as well, but was doubting compared to the G.Skill 2x16 3600 C16 B-Die set (F4-3600C16D-32GTZR) which actually does slightly better out of the box, 16-16-16-36, instead of 38.
I do like the look of these. Are you still satisfied despite the not so revolutionary overclock or would you in hindsight also consider something like the G.Skills ?
Price is the same but G.skill also has RGB, which although I personally don't like it, makes the price of the actual ram itself a bit lower (if you would subtract the RGB price component). And the G.skills are more mainstream. Not sure whether that's a pro or con for getting the best chips, but if the 8packs were indeed binned as thorougly as it's marketed I wonder why they don't have at least the same or maybe even better timings than the G.skill.

Did you consider the G.skills in your decision ?

Ps. I think the code in the title may have a typo, shouldn't it be TDPPD432 instead of TDPPD416 ?

 

[Woomack]

The tested memory kit was significantly cheaper than G.Skill (count about 40% cheaper when I was ordering it) and it was pretty much the main reason why I purchased this memory kit. When I decide to buy any memory kit then I also think about something more interesting or what I can't get as a review sample. As you can see on my list (sign or the first sticky in the memory section), I have a long list of G.Skill memory kits and I tested many which are not on the list. There are barely any TG kits.

It doesn't make any difference if tRAS is 36 or 38. It's more for stability than performance. For me it doesn't really matter as I wouldn't use the XMP. For others, as I already said, it also doesn't matter as the performance is pretty much the same.

I still want to test this RAM on AMD and other Z490 motherboards. My results were somehow limited by the motherboard as I don't believe it would end at so "low clock" when Micron B 2x32GB kit could run up to DDR4-4500. I also see how various memory kits act on Z490 motherboards. In short, I had problems with some other Samsung B based kits on various Z490 motherboards. Some 2x8GB kits had a wall at ~DDR4-4300. Literally no boot past that, even though it could run at 4266 CL16-16-16 1.45V without problems and on AMD it could pass DDR4-4800 CL18 in stability tests.
I just had no time to back to this kit but I will test it on AMD X570/B550 once I finish with some others. I still think about my daily rig for which I purchased this memory but plans somehow changed. My daily/gaming PC is always 1-2 generations older than my test rigs but sometimes the upgrage has no point and takes time to reinstall everything.

8Pack brand is weird for me, the same as many der8auer products (most products which are not extreme OC accessories) or anything else branded by a name of a well-known overclocker. 8Pack is not active for some years already and I'm sure he had nothing to do with this memory. He wasn't even a memory guy.
8Pack memory kits are not better than other Team Group kits at the same specs. You can see they have a wide range of voltages, the same as some other series like Xtreem. It suggests they can't guarantee low voltages for every kit. It's a matter of luck if your Samsung chips run at very low timings or like higher voltages or anything like that. I had mixed results on some other kits in the past. On the other hand, the same is with G.Skill. Simply all these DDR4-3600 CL16 or lower, or anything DDR4-4000+ from Team Group or G.Skill that is using Samsung B chips is the best you can get. There is still the luck factor.

The PN was taken from a single module. Thanks, I will change it as it suggests other product.

 

[gameinn]

Woomack in your opinion is G. Skill's recently unveiled 32GB kits at 4000 better bin?

F4-4000C16D-32GTZR

16-19-19-39 at 1.40v

Overclock at 16-16-16-36 with 1.45v you think to match this review kit performance?

I was also concerned about tRAS being at 38. There might be an influx of people enquiring about 4000 kits soon. It has been rumored today that Zen 3 CPU sweet spot is now 4000 memory.

 

[Woomack]

I doubt there is any significantly better bin. For example I was recently testing the best bins from G.Skill Neo kits and they're pretty much the same as HyperX Predator or Patriot Viper Blackout from new series. The main difference is that G.Skill guarantees higher clock or tighter timings while others not.
In this review, I had to set a bit higher voltages and as I mentioned, the motherboard had problems to pass some frequency. I have to test this Team Group kit on another rig. If I find some time then I will test it with Ryzen 4650G and Crosshair VIII Impact.

I wouldn't care about the tRAS. It's mostly for stability and not performance. Most brands set it a bit higher because on some platforms there can be some stability issues. For example when I was testing last kits with Ryzen 3900X then I could stabilize them at about 4600-4800 CL18-18-18-38 ... but when I used the same kits with Ryzen 4650G then I had to set tRAS at 40-42 or stability tests were crashing.
On Ryzen I would focus more on sub-timings as they affect the performance more at 4000+. Main timings can't go much lower while sub-timings are usually really relaxed at auto. For example you can drop tRFC from 700+ to about 400 what gives more in latency tests than going down from CL18 to CL16 or even CL14. Matter of clock and some other settings.

I'm not sure if people who write news like that about the sweet spot actually tested any Ryzen 4000/5000. I see that the news is based on some conclusions from Turkish website what says a lot.
My Ryzen 4650G easily runs at 4400 and up to 4533 with IF 1:1. Some results are in G.Skill Neo 16GB 3600 CL14 review. Even when I leave the fclk at auto and set memory at 4533+, then my ASUS motherboards automatically set it at 2200MHz so the mentioned sweet spot will be 4400. This is the same behavior as on Ryzen 3000 motherboards stop at 1800MHz so DDR4-3600 which is generally the "sweet spot".
I assume that the best chips will go up to 2300MHz so DDR4-4600 and 1:1 IF ratio. My CPU boots at this clock but crashes in all tests, regardless of voltages.

 

[gameinn]

Once again fantastic information woomack. I will wait for reviews on the 5000 to see actual fabric values but most likely this TeamGroup kit will be the one.

Just to clarify you think you should also set tRAS at 38 for 4000? Something like 16-16-16-38 1.46v?

 

[Woomack]

It should work at 16-16-16-36 but in case of stability issues can try higher. I have to retest this kit, as I said, but I have 2 others in queue that I have to test before.

 

[BugFreak]

Great review. I have the 3200 B-Die version of this and the same settings worked perfectly. Currently running mine at PC4000 16,16,16 with my 10600K. Thanks as always for a nice write-up.

 

[rrul]

Thanks, appreciate the detailed answer !

Sorry for late reply, didn't get activity notifications for this forum

 

[Woomack]

I finished tests of Team Group VulcanZ 3600 CL18-22-22. Software was suggesting Samsung B but it's C-die so tests were quick. Sadly, couldn't even run at 3733+ without errors. On the other hand, can back to Dark Pro 3600 today. I will run some tests on Ryzen 4650G+Crosshair VIII Impact.

Edit:

Tests were quick. The main difference between my previous results is required voltage.
On AMD I could reach DDR4-4133 CL18-18-18 at 1.35V instead of 1.5V. It passed ~30min stability test.
DDR4-4200 18-18-18 requires 1.50V. What is weird is that DDR4-4200 CL16-16-16 requires the same, 1.50V. 1.45V is not even booting at DDR4-4200.
Any higher clock isn't booting at all up to 1.65V.

I will test some more DDR4-4200 at tighter timings and will post something when I'm done.

4200 14-14-14-34 boots and passes simple tests at ~1.7V but instantly bsod under higher load
4200 16-16-16-36 passes ~30mins of stability test at 1.55V but not much more and up to 1.60V it acts the same
4200 16-17-17-38 1.50V seems stable, 2h 40min stability test below

 

[rrul]

Wow, super, great extra test.

Ordered them today. Bit of a bummer with Paypal exchange rates from EUR to GBP and delivery to NL, in total EUR 23,- more expensive than the B-Die G.Skill RGB 2x16 3600 C16. Afterwards not so sure why I took this one actually, guess I do not like RGB all over the place Plus it fits well with black Unify mobo (although I can hardly see through the tempered glass)

Will aim for 4200 16-17-17-38 @ 1.5v.

That would be better than 4000 16-16-16-38 A 1.45v right ? (frequency bump outperforming delay from trcd and trp 1 notch higher)

You think these sticks / pcb will take 1.5v daily without degradation ? (being rated at 1.45v max)

Hope the z490 Unify will handle it smoothly.

 

[Woomack]

My first tests were made on the MSI Z490I Unify and the memory couldn't pass 4133. I wasn't testing ATX version of Z490 Unify but I heard it overclocks a bit better. For example where I couldn't pass 4933 on the ITX version then ATX could boot a bit above 5000. I have no idea what about the latest BIOS as I wasn't checking it. I'm also surprised that these motherboards got only some BIOS updates after release and now silence for over two months. Every Z490 motherboard that I was testing had some little issues or limited memory OC potential. Like one could run with 2x32GB sticks up to 4500, all others couldn't even boot above 4200. On three different motherboards Samsung B couldn't boot past 4300 while Micron/Hynix could go up to 4800+. One of these motherboards with 4300+ clock issue, could instantly boot at 4800+ after BIOS update but all others had no update since August.

Samsung B is available in mass sales with voltages up to 1.55V and clock up to 4800+ so I guess it's safe. Usually it looks like 1.35V is up to 4000, then 1.40V up to 4266 and 1.45V up to 4500 where, depends on the brand, is 1.50-1.55V up to ~5000.

Btw. you could use Amazon account to transfer money. Not sure if it would be cheaper or not but that store has an option for that.

 

[rrul]

My first tests were made on the MSI Z490I Unify and the memory couldn't pass 4133. I wasn't testing ATX version of Z490 Unify but I heard it overclocks a bit better. For example where I couldn't pass 4933 on the ITX version then ATX could boot a bit above 5000. I have no idea what .

Super, thanks for handy info, will sure use it as reference.


Sticks are out of stock, they're underway apparently but I may be cancelling them and just go with the G.Skill version. Same specs 20,- euro cheaper.

Initially I wasn't in a rush because 3080's aren't available anyways, but I just bought a 9600 GPU for 15,- which allows me to boot into windows, so now I can go and order the rest of the parts for new build in the meantime.


Looking forward and after all that waiting for 3080's not sure I now also want to wait for RAM.


Regardless of brand I will post the mem oc results here back once done.

 

[gameinn]

G.Skill version. Same specs 20,- euro cheaper.

Where from?

 

[Woomack]

DDR4 is not running really hot. It will be hot in touch under longer high load but still safe at 1.50-1.55V. Top kits on the same IC are rated at 1.50-1.55V so if manufacturers think it's safe then it has to be as I doubt they would risk high RMA rate.

Lucky mode was crashing my OS even in quick memory tests. I wasn't digging into this topic so I won't tell you what it exactly does. In the description is something about better OC ... if you are lucky

On AMD, my kit was stable at 4200 16-17-17 1.50V. On Intel there was a problem past 4133 but I think it's related to my motherboard as it's acting a bit weird with other memory kits too. Also on my 4650G, I have to set 16-17-17 instead of 16-16-16 to stabilize memory at higher clock. The same was with G.Skill Neo or Patriot Viper Blackout at higher clocks.

 

[rrul]

Thanks

Still trial and erroring, just passed linpack at 4400 17-17-17-40 at 1.6V, trying to dial it down, but 1.56V just errored.

Let's say I get it stable at 1.57V with SA 1.36 (under full AVX load just under 1.4V) and IO 1.25 (under load equal or lower), if it were your personal gaming rig, would you daily that ?

Or is 1.57 Vram too much ?

Still hard to get a feel for this; people bench without problem up to 1.7, several people say 1.6 is the consensus max for b-die while others swear by 1.5. Just like SA, many state to not go over 1.25 while Buildzoid and other OC-ers state that 1.4 is pretty ok.

I'm sure my compu won't blow immediately and I am looking to stretch it, but within reason. If 1.57 or maybe 1.58 really is too much I will dial it down.


Ps. I'm a bit of an idiot, deleted my former post by accident. Was too long anyways, but now the continuity in your thread has a slight bump

 

[Woomack]

I switched my daily/gaming PC from 9700K+MXIG to 10900K+MSI Z490I last weekend and I'm using 2x16GB as in the topic in this PC. It runs at XMP ... Really, out of benchmarks it gives so little difference that I don't care but DDR4-3600 is already a good spot for a daily rig. I also have 10900K set at 4.6GHz ~1.07V so it uses ~50W while gaming. I actually tested it for 92mm AIO but I changed my mind and I left it like this because the PC runs 24/7 so somehow I care about the power draw and noise.

I fell like safe is whatever motherboard sets at max when you push it at higher clocks and auto voltages. In general it's something like 1.45V SA and 1.35V IO. On the other hand, on Z390 motherboards, BIOS had 1.65V SA marked as 'can instantly kill the CPU'. On Z490 it's not marked as a warning. I was pushing SA up to 1.85V on Z390 and all was fine. I'm not saying it's a good idea but I had no problems with CPUs after that. On Z490 I haven't seen any difference above 1.65V so it was like a quick test and that's all.
Keep in mind that most high voltage results around the web were made on LN2 cooling so many overclockers will share high clocks or voltages but usually won't add it's barely benchable and on LN2. Some months ago I had the highest memory clock on hwbot, made on air cooling. I have no idea what about now as I'm not checking these rankings for some longer. Let's say there is too much marketing in that nowadays and various sponsoring practices are killing healthy competitions.

Going back to the TG DDR4-3600 memory, I still find it a bit weird it goes to some frequency and can't even boot above that regardless of voltages. Maybe it's a matter of luck or used motherboards but the same was on Intel and AMD. On AMD it was hitting a wall at a bit higher clock but I would expect it to run higher. I had the same situation with Team Group Xtreem 2x8GB DDR4-4133 CL18 kit. It was reaching some point at quite tight timings and at next memory ratio it couldn't even boot. I know that some other guys on the OCF have the same memory kit and can set it higher on worse motherboards for memory OC.

Anyway, I'm done with the Dark Pro DDR4-3600 memory and I will be busy with some other things soon but I guess I will post 1-2 more memory tests in this year. However, you can expect it to be on AMD as it's more fun nowadays

 

[rrul]

Hi Woomack, thanks.

After 24+ hours trial and error I can confirm your findings for TG; your setting all above, 4200 16-17-17-38 at 1.35 (and 1.36 SA / 1.25 IO) are actually the exact ones I found to be stable, at least 600% HCI stable, didn't do memtest86 yet.

I think that's the sweet spot for freq/timings. Not the fastest, but still decent for a 3600 kit if you ask me.

Several faster combinations boot up fine and some even complete a 5x Linpack 10GB (like 4200 16-16-16, 4266 16-16-16, 4300 16-16-16, 4400 17-17-17) but it all fails eventually, no matter the VRAM/SA voltages. Looser timings like 4600-18 (or 19 don't remember if I tried that) didn't work to get higher speed.

Only variable is that I had secondaries/tertiaries on Auto and Slow Trainng enabled, so it could be that between runs those were configured a bit differently. It all felt somewhat inconsistent; the same settings passed and upon re-boot would fail. So maybe there is some golden config in there after all, but for now I'm focusing on getting the sec/tert down at 4200 16-17-17-38.

As you mentioned, probably the difference in real life gaming is marginal. It's just a bit OCD (no pun intended) and OC-ing is just fun, albeit time consuming and a little frustrating. Nice to know to pull everything out of the system.

Thanks for advice, will be following your tests.

 

[Bar Weinig]

Hi Woomack and rrul,

This thread is just awesome! I have been looking into RAM for a while now. Finding good information about the Team Force Dark Pro is difficult and this thread tags a lot of things I was looking for. I didn't want to mess up this thread with noob questions, so I'd like to ask you guys to check out my new posts:
This post is about RAM and especially the ones talked about here.
This post describes the build I have in mind for the RAM.

I hope you can give me some peace of mind to start ordering...
Thanks in advance.