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

FRONTPAGE G.Skill Trident Z 16GB DDR4-3200 C16 Memory Review

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

Overclockers.com

Member
Joined
Nov 1, 1998
G.Skill continues to release new DDR4 memory kits for Intel Skylake platform and of course we are happy about this fact. After a great Ripjaws V series we have a chance to test Trident Z whose design is simply exceptional. In this review we will present probably one of the most interesting memory kits for computer enthusiasts and gamers, the Trident Z 2x8GB DDR4-3200 CL16. Stay with us to find out how much you can expect from this memory and why it's one of the best options on the market.

Click here to view the article.
 
Nice Woomack, looks like a pretty good kit.
 
Hi Woomack,

After your reply here: https://forum-en.msi.com/index.php?topic=262420 I've been following your memory reviews

With default CPU settings of MSI Z170I's 1.34 BIOS, I tested XMP (3000MHZ kit instead of 3200MHZ), yet it failed at Prime95 with a shutdown, SPD speeds are pretty stable tho, no stress test issues, no usage issues (Also the temps are at acceptable levels on all tests, as far as I remember, they were around/below 70C when the shutdown occurred, XMP heats the CPU more than usual for some reason)

I'm wondering whether you did your test with:
- Exactly Standard BIOS CPU settings
- Balanced power usage of Windows (the funky steppings really help detect/cause/catalyze issues)

Without the above defaults, sometimes the issues went unnoticed for some time

At this point I'm chasing the issue out of curiosity and worry, wondering why my system isn't performing while others do, I'm hoping my system is just slightly weak instead of pretty weak, and funky SA/IO voltages with XMP of BIOS is causing these issues and a BIOS update might fix them at a later time
 
Hi Woomack,

After your reply here: https://forum-en.msi.com/index.php?topic=262420 I've been following your memory reviews

With default CPU settings of MSI Z170I's 1.34 BIOS, I tested XMP (3000MHZ kit instead of 3200MHZ), yet it failed at Prime95 with a shutdown, SPD speeds are pretty stable tho, no stress test issues, no usage issues (Also the temps are at acceptable levels on all tests, as far as I remember, they were around/below 70C when the shutdown occurred, XMP heats the CPU more than usual for some reason)

I'm wondering whether you did your test with:
- Exactly Standard BIOS CPU settings
- Balanced power usage of Windows (the funky steppings really help detect/cause/catalyze issues)

Without the above defaults, sometimes the issues went unnoticed for some time

At this point I'm chasing the issue out of curiosity and worry, wondering why my system isn't performing while others do, I'm hoping my system is just slightly weak instead of pretty weak, and funky SA/IO voltages with XMP of BIOS is causing these issues and a BIOS update might fix them at a later time

Two things:
1) you'll get better help if you start your own thread
2) when you start this new thread please list your detailed system components, including screenshots of the SPD and Memory tabs of CPUz
 
Hi Woomack,

After your reply here: https://forum-en.msi.com/index.php?topic=262420 I've been following your memory reviews

With default CPU settings of MSI Z170I's 1.34 BIOS, I tested XMP (3000MHZ kit instead of 3200MHZ), yet it failed at Prime95 with a shutdown, SPD speeds are pretty stable tho, no stress test issues, no usage issues (Also the temps are at acceptable levels on all tests, as far as I remember, they were around/below 70C when the shutdown occurred, XMP heats the CPU more than usual for some reason)

I'm wondering whether you did your test with:
- Exactly Standard BIOS CPU settings
- Balanced power usage of Windows (the funky steppings really help detect/cause/catalyze issues)

Without the above defaults, sometimes the issues went unnoticed for some time

At this point I'm chasing the issue out of curiosity and worry, wondering why my system isn't performing while others do, I'm hoping my system is just slightly weak instead of pretty weak, and funky SA/IO voltages with XMP of BIOS is causing these issues and a BIOS update might fix them at a later time

I'm preparing all my tests and reviews at 4.2GHz CPU clock. Cache is 4.2GHz for Z170 platform and 3.5GHz for X99. These settings are showing differences in memory performance better than stock and are easy to achieve even on stock coolers.
MSI Z170I Gaming Pro AC has some power delivery issues. First BIOS version was working fine when you made motherboard boot without issues and I mean no vdroop etc.
If you have instant shutdowns then set power limit to the max ( short and long time ). Also maximum current limit to the max and turn off power saving features. It works for me but still have to set higher cpu voltage to keep it stable ( huge vdroop on current BIOS versions ).

Reviewed Trident Z 2x8GB 3200 memory was working fine on mentioned MSI with XMP enabled and VCCIO/SA at auto.
 
Thanks a lot for the info, in the meantime, I got a reply from MSI stating that they accepted my issue and are currently addressing it, which relieved me

And thanks for the suggestion ATMINSIDE, when the BIOS gets an update, I might open a separate thread, as I will probably attempt a 1.25V manual safer OC instead of the 1.35V XMP
 
It's the safe limit tho, also so far I've been burnt too much by XMP, RAM's guarantee these speeds/voltages, yet as fas as the CPU specs go, 2133mhz+ isn't even supported

My information/experience is a bit tainted since I was dealing with faulty/non-optimal stuff, yet XMP as it is bumped temps almost 20C, that's why once a stable BIOS is out, I will aim for a low-voltage/more-stable OC

There is a 15-15-15-35 2800MHZ 1.25V TridentZ for example, would be my choice if it was available and if I had the opportunity :)
 
IIRC 1.5V is the max allowed for an XMP 2.0 rating, and is the recommended max voltage for daily usage.
1.35V is very much in the safe range, just like 1.65V is very safe for DDR3.

The manufacturer put the spec at 1.35V, with a lifetime warranty, for a reason. It is safe.
 
I don't agree with you, it is safe for the RAM, but it's definitely not guaranteed to be safe for the CPU, as the CPU's don't even support 2133+ officially

My CPU's IMC seems weak for example, XMP as-it-is is definitely not safe for me, this is why I'm inclined to achieve a sub-1.35 voltage

Edit: Like I hinted earlier, I draw conclusions from personal experience + online reading/research, it might be the Auto IO/SA Voltage's that are causing the CPU stress, or a bandwidth limitation of the CPU IMC, I'm not even sure whether that 1.35V is routed through CPU or fed directly from the Motherbord, if it's fed directly from the Motherboard, feeding 1.35V to the RAM is obviously the easiest option, tho the RAM's get to 45C's with usage even at 1.2V, it's unclear how the temps will be at 1.35V, so I personally prefer lower voltages
 
If you don't want to believe us that it's safe, go right ahead with your own thoughts.
The manufacturers of the CPU, motherboard, and RAM all say that 1.35V for the DRAM is safe though.
 
Yes 1.35V is safe, but it's the limit

http://ark.intel.com/products/88195/Intel-Core-i7-6700K-Processor-8M-Cache-up-to-4_20-GHz

The Rated Memory XMP limits are not safe tho, these modules are sold like they are supported by every 6th gen cpu out there, while in reality the situation is opposite

I don't get why you have a problem with me deeming 1.25V safer than 1.35V, low voltage, less wattage, cooler system, less issues

If you also look at any online forum, it's filled with people experiencing issues with XMP, I didn't even make a universal statement like "everyone should aim for < 1.35V" - personally I will aim for the lowest voltage I can get as I've been burnt too much
 
How, exactly, is 1.25V safer than 1.35V if neither of them cause any issues?
You're being given a lifetime warranty on the RAM at that voltage, the CPU is rated for it, and the board is rated for it...

Most of the XMP issues I've been seeing aren't voltage related, they're BIOS and/or new technology related.
 
I'm using my CPU at 1.20V for example, can get the same performance compared to Auto voltages, with much less than 5% performance loss, while Auto voltages push 160W's, with this voltage limitation the CPU is around 80W's at most

Similarly a cooler RAM is in my opinion better, if it can roughly provide the same performance at lower temperatures, I would prefer the lower voltage/temperature

It also doesn't matter if the RAM producers give lifetime warranties or not, they have a better look at the economical situation, they could be easily affording a 5% failure rate for example, I wouldn't want my system to fail with 5% probability, but they can easily afford it

You seem to be very interested in refuting my voltage choices but you are doing a weak job at this imo, you can refute it strongly with some solid information, if the 1.35V never touches the CPU for example it would be a strong refutal, but since there are warnings against running 1.5V Ram's with Skylake CPU's, the DRAM voltage obviously plays a role in stability, lower is better in my opinion
 
RAM voltage has nothing to do with CPU or anything else. It affects RAM and nothing else.
Memory manufacturers calculated that 1.35V or even 1.4V is safe limit for memory at which it won't fail even after long work. No one would give lifetime warranty on a product which can fail even in 3-4 years. The same made almost all memory manufacturers on the market. Kingston even set 1.5V for their early Predator kits.
Temperature between 1.2V and 1.35V is like 3-4*C difference tops ( you can check it on many modules with thermal sensors ). During long work DDR4 is reaching about 40*C max while max safe for most consumer grade modules is 80-85*C. At this voltage range what may cause failure or fast degradation is temperature which is far from safe limit even after overclocking.
Simply there is no difference if memory will work at 1.20, 1.25, 1.35 or even 1.40V.
It's not recommended to set high voltage on DDR4 at all but there is no clear rule what voltage is max safe. It's not only related to Skylake. No one will recommend to run memory at 1.5V or more when it was designed to work at about 1.3V ( JEDEC general specs is 1.275V max ). Additional tests and better PCB may help in stabilizing memory at higher voltages and keep it safe. That's why most good manufacturers are using better components to guarantee stable work at higher than standard settings.
Good example can be G.Skill which is offering Trident Z kits up to 1.4V with lifetime warranty and additional support. So far I haven't seen problems with their memory kits and many overclockers were pushing them above 1.7V.

Like ATM already said, right now the only issue may cause badly designed XMP and motherboard BIOS. So far I see mainly issues with Corsair DDR4-3000+ modules and Gigabyte motherboards. Most other motherboard manufacturers updated their BIOS and there are barely any issues with memory kits designed for Skylake. However there can be issues with some kits designed for X99 because of slightly different timing tables.

Back to reviewed Trident Z 3200 memory, I was able to stabilize it at DDR4-3466 17-18-18 1.35V and DDR4-3600 at the same timings using 1.45V. About the same results on both my motherboards so MSI Z170I Gaming Pro AC and ASUS Maximus VIII Hero. On both it required higher VCCSA and VCCIO voltages ( about 1.25V ).

Here is 4x8GB Trident Z DDR4-3200 16-16-16 + 16-18-18 @ 3600 16-18-18 at 1.55V after 2h 95% memory load in AIDA64. At these settings this memory is working for a week now.

attachment.jpg
 
Auto voltage on a CPU is a terrible idea anyway, unless you're running stock...

The vast majority of RAM is designed to operate up to 85°C, so you have plenty of room there.

Manufacturers are running <1% on RAM RMA rates.
https://www.pugetsystems.com/labs/articles/Most-Reliable-Hardware-of-2014-616/#RAM(Memory)

Why would there be a motherboard VRM section dedicated to the system RAM if it was being run through the CPU?
The answer is, because it isn't routed through the CPU.

I've been unable to find this for Skylake yet, but here's the Haswell pin map. Page 115 starts this information.
I have yet to find any pins that reference DRAM voltage.
http://www.intel.com/content/dam/ww...h-gen-core-family-desktop-vol-1-datasheet.pdf
 
Thanks for all the information, very useful, good to know

Can't wait to retry ram clocking myself after MSI releases the new BIOS for the Z170I
 
I've seen there is official 1.30 version for Z170I Gaming Pro AC and I have 1.34 beta but I simply had no time to check any of them. Right now I have only ASUS Maximus VIII Hero set for tests for reviews which also got new BIOS and so far it works great as you see on the screenshot in my last post. Earlier BIOS couldn't run with the same memory above DDR4-3466 while new is fully stable at DDR4-3600 with all 32GB.
 
Wow, I missed the 1.3, MSI USA has 1.2 as latest while Global has the 1.3, thanks for the heads up
 
Back