Check few posts back, I did a few 3-hour Prime95 v29.3 (with AVX/FMA3) runs with CPU temps lower than 80c without delidding.
My 8700K is a really rare chip with a very low stock VID.

DDR4-3200 vs 4000+, I'd say very little to no difference but anything lower than DDR4-3200 will bottleneck the system.

Source: TPU

This post has been edited by owikh84: Nov 27 2017, 07:55 AM

CL17/18 is not easy to stabilize, so going back to XMP at
DDR4-4266 CL19 @ 1.40v 4x8GB



ROG Maximus X HERO
8700K Coffee Lake
TridentZ RGB 4266C19 4x8GB


Attached thumbnail(s)

Each motherboard's power design is different.
ASUS and other brands have LLC level 1 as the lowest and up to 8 for the highest.
ASRock has it backwards where level 1 is highest and level 5 is the lowest.

On my ROG boards, I chose level 5 because I find it gives the lowest vdroop.
Check few pages back for my 5GHz CPU OC.
LLC5 used, BIOS set 1.30v, load vCore 1.280v Prime95 AVX 0 offset stable.

Typical voltage is around 1.2-1.4v depending on your VID and board of course.

Check out 8700K OC database from Germany: https://www.hardwareluxx.de/community/f139/...ad-1178256.html

Better check your chip's vid first before pumping any voltage.
For record, my 8700K comes with 1.088v vid.

5 hours+ of HCl 500% 😁
DDR4-4266 CL18 @ 1.40v 4x8G

i7-8700K
ROG Maximus X HERO
TridentZ RGB DDR4-4266C19 4x8GB

VID can be observed when your BIOS setting is set to default.
Hence, in your case your VID is 1.040v which is quite low even lower than my 8700K chip.
Hmm not sure why you need so much vcore but it's likely that the Strix mobo has a weak VRM power.

CPU Load-line Calibration [Level 5]
CPU Current Capability [140%]
CPU VRM Switching Frequency [Auto]
VRM Spread Spectrum [Disabled]
Active Frequency Mode [Disabled]
CPU Power Duty Control [Extreme]
CPU Power Phase Control [Extreme]
DRAM Current Capability [100%]

The bolded part sounds like a defective RAM module. It should have a lifetime warranty, so send it for RMA.

1) Most likely not, you're not pumping extremely high VCCSA like 1.4v+ etc.
2) Prime95 will not kill your RAM, unless you pump extremely high RAM voltage like 2.0v+ etc for long term without a proper cooling. The dead RAM module could be an already weak stick that suffered from ESD etc.
3) Prime95 will not kill your motherboard and CPU because the board itself has OC protection etc to prevent it from getting fried. When your CPU reaches 100C high, thermal throttling kicks in and your OC will be reduced at lower frequency to reduce the temperature. But chip degradation over time is undeniable, so best is to keep it as cool as possible.
4) Answered in number 2.

Wow, that was fast! I think I saw your post about this in a FB group.
How much is the shipping cost?
Last time I used poslaju around RM100, DHL around RM200.
Return trip is free of charge.

hmm I guess you will need a better mobo to confirm as I saw your VID is quite low so couldn't be your chip is worse than mine.
Btw, if you're not doing video encoding stuffs you can use Prime95 v26.6 that doesn't have AVX/FMA3 instruction.
It is less stressful but enough for your daily work. It also requires less voltage, so you should get lower CPU temperatures.
The legendary Noctua NH-D15S is a great air cooler that it performs even better than some AIO coolers in the market.
I also have the NH-D15S and the vanilla D15, maybe be will give it a try when I am free.

I see... My pleasure bro iBenQ.
All da best man.

DDR4-4266 CL17-19-19 with 4 x 8GB...
en route to CL17-18-18 <= doable but not sure can achieve full stability or not 🤔



Core i7-8700K
ROG Maximus X HERO
G.Skill TridentZ RGB DDR4-4266 CL19 4x8GB

8600K - ES sample
8700K - retail
ROG Maximus X HERO - retail
ROG Maximus X APEX - review sample
TUF Z370-Pro Gaming - review sample

This post has been edited by owikh84: Dec 5 2017, 05:19 PM

arghh not again!
btw, good deals during last Black Friday.

hmm... couldn't be that ASRock set -4 as the AVX offset by default.
Inside Windows, you can check it through software like XTU or probably ASRock's own utility.
Inside UEFI, you can manually set it to 0 in case you want to disable this function.

No zero offset option is odd. Auto being -4 offset is more weird
It sounds like the AVX offset is not working properly, for me it's a big bug LOL.
Too bad I don't have any ASRock Z370 board with me, maybe you should contact ASRock for clarification haha.

CPU-Z is not a suitable software to test stability of your OC, neither a good software for benchmarking purpose.

For CPU overclocking, the most standard people would go is Prime95, but take note that it is way too much stressful that your cooling can actually handle the heat.

For me, a couple of hours (3 hours minimum) of Realbench or AIDA64 are more representative of real-world applications.
Between these two, Realbench is better since it stresses both CPU and GPU at the same time.

Anything lower than 100c would be safe. Moreover you're using an air cooler, which is less worrying than custom watercooling with PETG tubing etc. Your normal usage at 65c max looks fine though, just need to retest with a proper stress tool as I mentioned earlier to confirm stability.

Previously, I said go for the none AVX-based Prime95 v26.6 if you're not doing video encoding stuffs but it seems like I was wrong.
Nowadays more and more software including games are using AVX instructions. It is suspected that NVIDIA driver also has AVX instructions in it.
Hence, Prime95 v29.4 is definitely a better choice if you don't want to see the CPU speed fluctuating due to AVX offset.

No delid, my chip is doing 5GHz @ 1.28v under custom wc with max temps never go above 90c.
Every chip was born unique, meaning each CPU comes with its own stock voltage (VID).
If you're lucky enough you will find a good chip, while some others might get bad chips with high VID.
It's all about draw of luck, silicon lottery.

How about the CPU frequency showed in CPU-Z?
I would disable BCLK spread spectrum in UEFI to prevent/minimize the BCLK fluctuations

Should be fine as CPU-Z is showing 5.0GHz.
Just ignore that value in XTU, sometimes I will see 4.99GHz as well. The benchmark score is more important, but expect to get lower score with slower RAM speed like DDR-2400.
My cache is slightly overclocked to 4.5GHz from 4.4GHz.