Rampage VI Extreme problems, need assistance please.

[Franziska]

Hey everyone

It's been almost 6 months since I have build my new PC with the R6E and so far it's been like a wild rollercoaster ride, ups and downs and new surprises every week.

First off my detailed system specifications:

ASUS Rampage VI Extreme (latest BIOS version 1301, previously used 1004)
Intel i9-7940X (bought delidded from caseking.de)
G.Skill Trident Z RGB 64GB (8x8) F4-3733C17-8GTZR (on QVL)
Nvidia Titan Xp
Intel Optane SSD 900p 280GB (U.2 2.5" version)
Seasonic PRIME 1200W
NEC SpectraView Reference 322UHD-2

Caselabs Merlin SM8
EKWB CoolStream XE480+360 Radiator
Aquacomputer cuplex kryos NEXT .925 silver CPU block
Aquacomputer kryographics GPU block

Windows 10 Enterprise 1709 and prevously used 1607, both up to date.


The story so far:

When I first build this PC in early November 2017 everything worked fine for about 2 weeks, stock settings, no overclock of any installed components, and then the system just randomly froze up in Windows 10. Turned out to be a defect memory kit which I replaced with a different one that is on the QVL for this board.

The ASMedia USB ports (first row on the back) also randomly stopped working, with a device manager code 43, defective hardware detected. This happened every couple days and it took a few power cycles to get them working again. The Intel USB ports worked fine.

So after that first incident with the memory life has been good for 2 months and then the next surprise happened, random bluescreens not pointing to anything, just saying CRITIAL_PROCESS_DIED. No dumpfiles generated. Again this also happened with default settings and no overclock.

I used Windows 10 Enterprise 2016 LTSB at that time which is based on 1607 and decided to try a clean installation of 1709 FCU.

Everything worked again for a few weeks, USB ports on and off as they please, and then the next thing happened: I installed the latest Aura software to play around with it for a while and the next day after a cold boot I was greeted by 48GB RAM. 2 of the 8 memory sticks were no longer detected by the board.

Another defect memory kit? I think not. Just like the USB problems it took a power cycle to get the missing memory back. Everything was fine again for a few days, 64GB RAM and all. Another cold boot after I have been away over the weekend (wall plug pulled) and this time the board would no longer detect the graphics card, VGA LED lighting up white and the QCODE display saying something about NVRAM. Like before another power cycle fixed this.

A few other side effects I noticed since I'm on this X299 system:

Some games stutter bad and that's not related to anything-GPU. GTAV for example is unplayable for me on X299 no matter what settings, what drivers, nothing I tried fixed this. Other games like The Division, Ghost Recon Wildlands or the newly released Final Fantasy XV are running perfect. Overall my gaming experience has been worse than on my old X99 system (R5E10/6950X).

10G works for the most part, I connected a Synology DS which also has 10G and performance was good. Sometimes the Aquantia 10G port stops working and it has to be disabled/enabled in Windows to get it working again. Rare but it keeps happening.

Other components such as the graphics card, power supply or SSD have been tested with my old X99 system and all of them work fine.

Stability tests I used on this system (all passed for 8-12 hours)

Intel XTU
HCI Memtest Pro
RAM Test (Karhu Software)
Google StressAppTest (Tried on Arch Linux)
Handbrake, FFmpeg and several other encoding tools
ASUS RealBench


I read a few posts on overclock.net where other users reported having issues with delidded CPU's from caseking, like they unknowingly damage them during the delid process. Apparently sometimes a CPU still kinda works if some of the small capacitors got damaged. I don't know if that is the case here.

I'm frustrated at this point because even though this system passes all the stability tests I listed above it is still very unreliable because of the random problems that have been happening to me in the last months.

I'm thinking about sending the board in for RMA and the CPU as well to have caseking check on it.

Anyone else experienced similar problems?

Any suggestions are welcome

Thanks in advance

[Beto556]

It may seem crazy but make a test: Slacken a few screws that hold the CPU cooler and disable aquacomputer services

[Franziska]

Beto556 wrote:
It may seem crazy but make a test: Slacken a few screws that hold the CPU cooler and disable aquacomputer services

Thanks for the reply. The CPU has been reseated already when the problems first started and because of the intermitted nature of these problems I don't think that's the case. Bent pins or a bad mounted CPU would result in permanent problems right?

As for Aquacomputer services, there's nothing like that installed. I do own an Aquaero 6 but it's not in this system, it's in my old X99. Just the CPU/GPU blocks in this one are from Aquacomputer.

Right now everything works fine again I will see how long it stays that way.

Best Regards

[Marko911]

This is gonna be totally inappropriate from me,but I had to ask,where did you buy SM8 in EU? 😄
P.S My advice for you mobo problems is that you send it back to caseking.de,they have great customer support.(in my opinion,I don't think that's CPU problem)

[Franziska]

Marko911 wrote:
This is gonna be totally inappropriate from me,but I had to ask,where did you buy SM8 in EU? 😄
P.S My advice for you mobo problems is that you send it back to caseking.de,they have great customer support.(in my opinion,I don't think that's CPU problem)

Hey, no problem 🙂 My SM8 came directly from Caselabs I bought it before best-cases.eu or what they're called became a EU distributor. It got very expensive in the end with customs and tax but well worth it.

As for my R6E yea I will see. Its just bugging me that sometimes everything works for 2-3 weeks and then things happen like the board randomly not detecting some components. Btw this always happened after cold boots, if I leave the PC in soft-off (S5) over night its fine.

[Marko911]

Franziska wrote:
Hey, no problem 🙂 My SM8 came directly from Caselabs I bought it before best-cases.eu or what they're called became a EU distributor. It got very expensive in the end with customs and tax but well worth it.

As for my R6E yea I will see. Its just bugging me that sometimes everything works for 2-3 weeks and then things happen like the board randomly not detecting some components. Btw this always happened after cold boots, if I leave the PC in soft-off (S5) over night its fine.

Huh,yeah,I was too looking at bestcases.eu but decided against it,and I still don't know if i'm ready to pay around 2000$ for STH10 from USA with all the taxes,shipping alone is 450$ 😄
I understand,had zero problems like that... 😞 it's so hard to do RMA process,when you think about all the lost time....Hope you do successfully and quickly RMA,I really think that will resolve all your issues. 🙂 (it can happen to anyone,I had bad luck with two EK D5 pumps and even singularity computers reservoir started to leak).

[Beto556]

I have 7900X dellided by caseking + Kryos Next Silver w/ Vision + GSkill 64C14....2 cores at 4.600 and rest 4300
Watercooler system type "Frankstein" + 1 960PRO
Asus Optimal Overclock setup

No problems !!!

[Silent_Scone]

I'd recommend installing a dummy OS and see if the problem persists before you take any further action.

[Franziska]

Silent Scone wrote:
I'd recommend installing a dummy OS and see if the problem persists before you take any further action.

Yeah I did that with a clean installation of Windows 10 Enterprise 1803. When up and running it's perfect, the problems seem to happen during POST after a cold boot. Like yesterday it was stuck at NVRAM and no VGA detected. A few days before that it only detected 6 out of 8 memory sticks.

I wonder if the memory problem was related to Aura, SPD corruption on Trident Z RGB. Thaiphoon Burner doesn't work with Skylake-X and in HWiNFO all 8 sticks look fine now. As for the graphics card, no idea, it was detected with no problems today leaving the system in S5 over night. The POST time varies though sometimes I feel like the board needs longer to detect all components.

[Silent_Scone]

The problems you describe in the above post are simply down to memory instability, it won't be related to either of those things. You'll need to tune out the instability with VCCSA and VCCIO voltage rails. 64GB kits can be tough work for some CPU, especially beyond 3600. See how far you're able to get with 3600 ratio.