Maximus VIII Hero q-codes 14, 15 and 99

TheF34RChannel

TheF34RChannel

TheF34RChannel

Associate
Joined
12 Jul 2015
Posts
1,694
Here we go again: Was playing FO4 when the system suddenly shut down and went into a reboot loop again, same as before. Only code 15 this time (pre-memory system agent initialization is started). Reseated and tested all RAM and slots, cleared CMOS. Got code 14 then (pre-memory CPU initialization started). Checked CPU and socket; all okay. Removed and reseated mobo and tightened the screws a bit more than before*. Booted just fine.

*my crazy theory; maybe the mobo screws weren't tightened enough causing the whole thing to sag just enough to short it, causing the problem?

Any ideas what might have caused it, and what I can do if it happens next? I'm not feeling really comfortable now that it might happen again.

Edit 1: Well it was fine in Windows, loaded a save game and boom there it goes again. Thoroughly p***ed off now! I've never had PC issues before and now it's something off and on with a single system I cannot reply on.

Edit 2: Could only get it working after taking out the entire mobo again but this time I replaced the screws for ones with a slightly wider head for more grip. Then I got code 99 with no display. Had to revert to my old DVI display, enter BIOS after another CMOS reset and save and exit to get my new 144Hz DisplayPort monitor working again. I've been at it for hours and am worried that as soon as I launch a game I'll be back to a boot loop (and I was right). Even took all the OCs off. I don't know what else to do :(

I have no extra components to test with nor can I get any.

Online searches yielded people with the same thing across all kinds of generarations of boards, always seems to happen during heavy load ie gaming but no one has found a solution. Replacement boards didn't solve it either for them.

Am going to flash the newest bios to it shortly and even considered the wall socket being bad so tried another, albeit not grounded, one to no avail. Also, I have my GPU zip tied upwards to counter the sag, could that be it? Grasping at straws now.

Edit 3: Because I keep thinking of a short somewhere I removed the side panels (to no avail), reseated all cables in the PSU, removed the top panel and it booted just fine - although I doubt that's the culprit. Here's a pic of the top panel where you can see a ground wire attached to the case, a rolled up LED wire I don't use, and 3 empty sockets from which I removed cables I don't use (front eSATA, USB 2.0 etc.).

Could there maybe be a short caused by something faulty with the case that I simply cannot see? I don't want to buy a new case on a whim, hoping that'd solve it.

osbh9c.jpg


5 January 2016: Big edit below the red edit indication; could use some of your thoughts on my ideas.

With case closed it no longer spins up - the PSU fan I mean.

SSD is in good health.

Testing the AIO: in OCCT CPU temp was 59 C on the hottest core during the PSU test.

So tested without OCs and adding one OC at a time (in OCCT):

PSU: 80 minutes - no issues
GPU: no issues
CPU: no issues

System is still up and running since I refitted the cables into the PSU side so I cannot replicate the issue nor am I convinced it's the cables (when I build the system and it would do something similar it gave me Q-code 13, I tested with stock PSU cables to no avail - unless the matters are unrelated).


EDIT 5 JANUARY 2016:

Incident: watching a second South Park ep in WMP. Immediate system shutdown. Boot loop gave Q-code 25 (= nowhere in the table) (gave 25 earlier, after a clear CMOS button it went to 14 or 15). System tries to boot 2x, then gives up.

What I did: Shut off the PSU and drained power. Removed extension cord from electric socket. Removed all CableMod PSU cables on the PSU side and re-fitted them. Removed PSU cables from motherboard (24-pin only; 8-pin is can only be removed by removing AIO cooler) and GPU. Took out and re-fitted the RAM. Took off and re-fitted the front USB 3.0 connector from the motherboard (next to the 24-pin socket). Tightened the SATA and power connectors on both the SSD and HDD - thinking of Never Back Down's previous issues.

Result: System booted fine instantly.

Thoughts: I may have narrowed it down.

1. It could be the RAM. It's not on the QVL list but that shouldn't be a problem. Is there a good recent RAM test that will replicate the issue if any exists?

2. It could be the front USB 3.0 connector (can that cause an immediate and full shutdown to begin with?), and here's why: perhaps the Q-code when I built the system is unrelated to the current issues. You see, I left the connector off until a few days ago.

3. It's one of the CableMod cables (refer to my Q-code 13 thread where I note I did test cable by cable with the stock Corsair cables both outside and inside the case).

4. On the off chance that the SATA cables might be bad, although they are brand new ones that came with the motherboard, I could replace them if the issue occurs again.

EDIT 6 JANUARY 2016:

After a good night of gaming last night the system (now without the front USB 3.0 attached) decided to shut down again, went into a boot loop until b6 (Clean-up of NVRAM - no idea what that means, neither does Google) and then boot looping to 25 a couple of times. Only refitting the PSU cables like yesterday worked.

I did check again and my RAM (Corsair CMK16GX4M2B3000C15) indeed isn't on the QVL. However the same kit in 4x4 config is.

Found another lad with a similar system (mobo, GPU, exact RAM) with similar issues and I asked both Asus and Corsair if they are familiar with the the mobo & RAM combo being problematic - earlier Asus advised me to try QVL parts only but obviously I don't have other RAM, PSU, GPU and SSD lying around for the fun of it.

My question to you guys and girls is:

1. Could it indeed be the RAM?
2. Could incompatible RAM cause this after working fine for weeks?
3. Alternatively, could it just be a bad CPU OC causing this? I ask because after refitting the PSU cables now and again it says 'OC failed' upon booting. Would be weird if it worked fine for a couple of weeks.

Also, how about the CPU 8-pin being too tight here:

ehdocw.jpg

Silly question: how about where I connect the CPU and GPU power cables in the PSU?

Right now I have the CPU in slot 1, the GPU in slot 2.

Further, I did update the Nvidia driver to the latest right before the issues began, although doubtful I will revert back, and try the stock Corsair GPU cable.

Been reading threads online left and right but no one has a real cause or solution. The ROG forum is full of them with different gen boards. I am at wits end to be honest. I also don't know if it's related to the code 13 when I assembled the system. I built many PCs in the past but never encountered this issue; I'm puzzled.

It always shuts down under load (only in game) except for once when it did in the bios and in Windows Media Player.

Only this changed before the shut downs:

1. Different monitor
2. Updated GPU driver
3. Some Windows updates
4. Put a very mild OC on the GPU (shuts down with it off also). 1388/3637, stock volts

I only play FO4 currently, but doubt it's the game.

I'll put it all on a box come weekend, but how to continue if it fails? And how if it doesn't?

Images over 1280 wide, I have placed these in spoiler tags
 
I think there's a new BIOS out for that board. Are you up to date?

How stable is the system at stock speed?

Have you tried tuning the system agent an vccio voltages?

What about that GPU overclock? Do you get stability at stock?
 
Last edited:
With or without any overclocks the shutdowns are the same. BIOS is 1302, but also on 1202 the same thing.

Am about to tinker with SA and VCCIO voltages, and looking into Corsair RAM compatibility. Tested all hardware, no issues. New PSU cables, no result. I'm putting it in a new case in case there's a short or case faulty, come weekend. If all else fails, I'll RMA the motherboard I guess.
 
Sounds like you're on the right track, checking each component seperately if you can.There is a new BIOS 1402, I'd definitely try that first. Clear your CMOS too.

My suspicion would be RAM compatibility. Its always an issue with new platforms and new memory; it can take a few months for motherboard vendors to get up to speed, and sure enough the z170 forums have been full of folks having issues with 2800 or faster DIMMs. They seem to be slowly resolving those with new BIOS updates.

You could also try disabling XMP and manually setting your RAM timings?

Other things to consider, obviously, is faulty hardware.
 
Yes will set te memory manually from now on. Over on OC3D TTL said my RAM & mobo combo shouldn't be an issue.

Will flash 1402 yeah!

Weird thing is, system worked fine for about 5 weeks. Every part is tested and in working order, neither me or the repair shop can figure it out. It can run fine for a few days until it doesn't and shuts down. Replugging PSU cables as described above is all that gets it running again.
 
Honestly you've provided probably too much information that's likely irrelevant to the problem so it's difficult to read.

Run the system at stock and retest. (Memory at 2133) 14, 15 can occur when the memory is unstable.

The shut downs can be a sign of instability also, or PSU compliance with low voltage stages. Enter the UEFI and under advanced, CPU Configuration, disable states from C3 onward.
 
Last edited:
TheF34RChannel said:
I have done just that and didn't make a difference I'm afraid.

Ran HCI Memtest filling up 16GB for over 900% coverage as well, no errors with XMP on or off.
Click to expand...

Disable the power states and retest, or soak test for any further shut down events. Have you mentioned what PSU you are using? When running HCI be sure to leave enough memory for the OS to breathe or else you end up simply testing your page file
 
It's in my signature.

How much ram should I leave open? Or rather, how many instances of how much ram per instance should I simultaneously run?

I should mention it always boot loops to 25, after tinkering to 14, 15.

Soak test?
 
Depends on your background activity. You want to allocate at least 85 to 90% of memory and keep background activity to a minimum. Allocating all available memory will simply test the HDD.

You want to run one instance per thread. So for a 6700k that is 8 instances. You'd be looking roughly to allocate at least 1750 per instance.

That PSU should be fine assuming there is no issue. If you are using extension cables or custom ones, revert to the originals (including GPU). By soak test I meant use the machine and see if the shut down is replicable.

If really coming unstuck with all of it, remove the motherboard from the chassis and retest. Reseat the CPU and check for even mounting pressure / bent pins, whilst also reseating the memory.


I would focus on doing the latter part of my post though at first whilst leaving everything at stock.
 
Last edited:
Cheers mate!! How do I verify even mounting pressure precisely?

Some of these I can answer:

Issue cannot be replicated: it happens while gaming, browsing Windows, once in the BIOS.

No bent pins on the socket.

I run HCI Memtest always during the night with nothing else on. Run it with a single or both sticks? I'm tempted to do two and switch to one if errors occur.

I had modded cables but it also happened with the stock cables.

I'm also gonna put it in a new case in case there's a short somewhere in the old one.

What do you suspect the issue is?
 
Last edited:
Who knows, it's either power or instability related. If I were in your shoes I would remove the system from the chassis and retest at optimised defaults. Make sure temps are acceptable obviously too if not done so
 
Yeah temps are great!

What wil having out of the case tell me? What should I connect / attach parts wise?

At what point should I disable power states and test in regard to your previous post?
 
If testing at optimised defaults leave all power states as they are and ignore what I said.


Removing it from the chassis may tell you if you have overtightened anything. Specifically around the memory. You said initially you tightened the screws. Loosening them is what you should have done if anything. This could potentially be why you were getting initialisation codes. That's not a straight fire answer though, you will need to test.
 
Yea, I'd agree with taking this out of the chassis. If the memory is OK, we're probably narrowed down to the PSU or motherboard.
 
You must log in or register to reply here.
Back
Top Bottom