ROG Extreme IV black screen crash issue

[Contrail]

Hi guys,

I'm new to the forum and was hoping to get some help and join the community.

I've been having a problem with my system for about 4-5 months where it seems to crash from Windows to a blank screen which it cannot recover from - happens intermittantly a few times a week. It looks like the system still continues to operate (the motherboard looks okay, fans are still going), however the screen clicks off and I have to hard reboot the system to get it back.

The Windows logs don't seem to contain much info, I did notice a few entries from the GPU driver, but I haven't managed to come up with anything there and it doesn't always log that notification anyway. I've noticed that a fan does speed up inside the system just before the crash happens and I recently installed AS Suite which threw up an alert the other day saying my CPU temp was at 120 degrees!

I've reseated all components, and have also moved the system onto my desk so I can keep an eye on everything inside the case. At the moment I think it could be a sensor issue on the motherboard causing the system to crash, but I'm not sure - so perhaps the next step will probably be a BIOS update.

Below is a list of my build, I'd really appreciate any advise you guys could give on where to go next.

HW:

ASUS ROG IV Extreme
Intel i7 4720k
Corsair Vengeance Pro 2100MHZ 48GB
EVGA SSC GTX970
Corsair AX860i PSU
Corsair H100i
Noctura Fans
Samsung Evo 840 500GB SSD
Samsung Evo 850 500GB SSD

SW:

Windows 10 64
Corsair Link 4
BIOS Version 4801 (I think need to check)

Cheers

[Chino]

Make sure your motherboard is running the latest BIOS. Then clear your CMOS and test your system at complete stock defaults.

[Contrail]

Thanks for getting back to me.

I've updated to the latest BIOS and will monitor it for a week and report back to the forum.

If I get any further problems I'll clear the CMOS and then test again.

[Chino]

Once is more than enough. Make sure it's at stock which means no XMP, no CPU overclock, etc...

[Contrail]

So I upgraded BIOS without clearing the CMOS (just to see if that worked), because I had some fan settings that were tweaked.

The computer crashed again, and when it rebooted the POST message said something about memory overclocking failed - the memory was OC'd to 2100Mhz using the XMP setting in the BIOS, so I removed this and left it at 1333Mhz, which appears to be the default in the BIOS.

I was unzipping a file today and it just crashed again and it seems like the fans (possibly the CPU radiator fans) speed up for a second before the crash.

I've noticed tasks that seem to tax the CPU (like running virtual machines, unzipping files, reading big PDFs) seem to trigger the crash. I'm also aware that the RAM is not on the ASUS qualified vendor list, so I'm wondering if this is causing the issue.

One final thing I've noticed is in the manual there's a power connector (possibly optional) for the RAM which I've never plugged in. I've never noticed it before, but it could be something else that's causing the issue.

Having now cleared the CMOS and am running everything at stock (apart from two fan profiles) I'll continue to test and see what happens.

[Contrail]

I've got two Corsair Vengeance Pro kits in the system. One was a 16GB kit and the other a 32GB kit.

I can see the CAS latencies are different in CPU-Z, although I'm sure I had this problem before I put the 16GB kit in.

There's a post on this forum that talks about problems mixing RAM kits, even if the model is the same - something about sub-latencies being an issue.

This could be the next place to look if the system isn't stable after doing CMOS reset.

[davemon50]

You might check your video card for overheating. Make sure your inside case temps are "cool", and as a preventative measure remove the card, blow it completely clean of dust and debris, and reseat it firmly along with all connections.

[Contrail]

You might check your video card for overheating. Make sure your inside case temps are "cool", and as a preventative measure remove the card, blow it completely clean of dust and debris, and reseat it firmly along with all connections.

I've reseated pretty much every connection inside the PC and cleared out dust etc. I'm also monitoring temps on the mobo using Corsair Link and everything looks good.

One thing I have uncoverred though is the RAM CAS latencies are different on the two kits I've bought. The first kit was a 32GB Corsair Vengeance Pro 2133 11-11-11-27 and the second kit was a 16GB version with 11-12-12-27.

I've now stripped out every DIM except one 8GB module from the 16GB kit and have had a stable machine for a few days. If it remains stable I'll remove that DIM and add a few DIMs from the original kit, test for stability for a week and introduce all DIMs from the original 32GB kit.

Assuming this works, I'll probably have to ditch the second 16GB kit, as I can't see that the differing CAS latencies being good for stability.

[Contrail]

The PC crashed again today just from hibernation mode in the Windows login screen. This was with two 8GB DIMM modules installed with the same CAS rating - 11-11-11-27. Interestingly pings to the NIC were still working from another machine on the LAN.

It also crashed the other day when I was trying to write to a SMB network share - I've noticed before this causes crashes when writing to my NAS running SMB.

I'm keeping a log of the crashes now, I think they are happening when I have the original 11-11-11-27 DIMMS installed.

I have now gone back to a single 8GB 11-12-12-27 DIMM and will stick with that until something happens.

This is such a weird problem!

[cekeu]

I think may be a failure of the graphics card