Sorry, long post to describe what is happening....

I was surfing the internet and the computer suddenly and without warning completely shut down.  When I pressed the power button on my tower, the computer would act like it was starting, but only for 2 or 3 seconds, then it would shut down for 2 or 3 seconds, then it would start for 2 or 3 seconds, then it would restart for 3 seconds, etc., until I turned off my power supply.

I thought it was the power supply, but after unseating and reseating the RAM, the computer would boot up seemingly normally and post to BIOS, but the BIOS was only recognizing 4GB of my 16GB of DDR4 RAM, properly inserted in the correct slots.  CPU and MB temp showed around 35C off the top of my head.

I shut the computer down and changed the button battery on the MB, that did not help.

I then used only 2 sticks of RAM at a time, properly inserted in the slots needed when only 2 sticks are used, and I could get it to post, but still BIOS only recognized 4GB of my 8GB of RAM.  Tried to boot to windows and got "stopcode bugcode USB driver error."  Couldn't get Windows 10 to boot.

OK, was thinking bad RAM so I numbered the RAM to keep track of which was being used, and swapped RAM modules.  However, no matter which combination of two I used, I could not get the system to post, even with the old combination of two that worked before.

I'm thinking my MB went bad.  If it was the power supply, I wouldn't have been able to post or boot to the Windows 10 boot screen at all.  It it was the RAM, the switching of the various modules would have eliminated the bad stick and I would have been able to boot to BIOS.  Plus, should I be able to Post to BIOS with no RAM sticks in?  The system won't even do that.

Any advice appreciated.  I'm baffled and have never had a rig just suddenly quit like this one did!

Without RAM the system will not post so that is normal. It sounds more like the memory controller on the CPU may have gone bad by the symptoms. Try removing the CPU, checking for bent/damaged pins in the socket as well as warping of the CPU substrate and or burn marks. If all is well, carefully clean the bottom of the CPU with some Isopropyl alcohol and something like a makeup removing pad making sure not to leave any cotton on there then reinstall it and test again.

Being an X99 platform the RAM should work in any slot (1 module), if it does not then there is a problem, either with the board or the CPU memory controller. 

I am sorry to hear about your issue, it is one of the worst kinds of issues you can encounter because the issue could be caused by 2 components. There is nothing worse than replacing the wrong one and having the issue persist.

There is a small chance it could be the RAM but that is less likely. If the RAM was running at a higher than rated voltage it is quite possible it burned out. If you can get your hands on another system to test it in you should try that.

Thanks Xaltar.

One of the first things I did when the system crashed was take the CPU cooler off, remove the CPU, and check for damage.  There wasn't anything obvious to me.  I then put the CPU back on the board, put new thermal paste on, then the cooler.  That, unfortunately  had no effect.

Thanks for the empathy!  : )  This does suck.  

I was thinking that it is likely not the RAM itself because I have 4 sticks and have tried every combination of two in the appropriate slots on the MB and one time it allowed me to post, but then not again.

This MB is less than a year old, so under warranty.  Does ASRock have a technical support phone number?  The only thing I have found is a customer service number and a tech support web form.

Thanks for replying Wardog.

I did build this system back in March 2016, and it worked perfectly the entire time until the system suddenly and without warning just quit working.  It was as if someone had suddenly shut the electricity off to my computer.

http://www.newegg.com/Product/Product.aspx?Item=N82E16820231792

That is the memory kit I purchased.  

I don't *think* I overtightened the cooler.  There is really nothing to overtighten I don't think.  It's the cooler below, and it just has a lever that you rotate through 180 degrees to fasten the cooler to the brackets on the MB.  

https://www.amazon.com/gp/product/B00BSKY1M4/ref=oh_aui_detailpage_o03_s01?ie=UTF8&psc=1

So the two of you are thinking either MB or CPU?  I guess I should buy a new MB and see if that fixes it?

Good.  I was pretty sure the memory was compatible as well, so thanks for the double check.

I'd do the voltage thing if I could get it into BIOS.  Unfortunately, I can't even get to that point.  

The power supply was purchased brand new in March 2016.  The whole build used brand new parts and worked great without one problem until now.  I have an older 750W power supply that works that I am going to plug into the system today to eliminate the power supply as the source of the problem.  

If it ends up being the power supply, I'll post back.  Otherwise, I guess I'll buy a new motherboard and try that and post back in a few days when it arrives.

When I pulled the 4 sticks of RAM and went to various combinations of two sticks in an attempt to find what I though might be one bad stick, I did get the system to post to BIOS.  I exited BIOS and attempted to boot to Windows 10, and it looked and acted like it was going to work, but then i got this error message:

Of coruse, now I can't even get it to post to BIOS so I can't diagnose the USB driver error.  Not sure if that helps with the troubleshooting.

Wardog

I replaced the PSU with one that works, and I'm still getting similar behavior with the computer booting for about one second, then off, then on, then off, etc., every second or so.  After several cycles like this, it runs steady, but nothing happens except the fans running.  The system doesn't respond to the power button, either, when I try to shut it off by holding the power button down.  I guess I should be thinking about replacing the motherboard now?

http://forum.asrock.com/forum_posts.asp?TID=630&title=how-to-clear-cmos-via-battery-removal

If there is CMOS data corruption it could cause the issue, especially if it is the RTC. Also, have you tried switching to BIOS B?

Xaltar-

Thanks for responding.

No, I did not do a full CMOS clear.  I will read the thread and do one.  Also, I did not switch to BIOS B.  I don't know how to do that.  

Oh, good to know.  Thanks. Had no idea!

Tried resetting the CMOS data by using the button, using the jumpers, and removing the battery and unfortunately still the same problem.  I also tried BIOS B and that didn't work, either- same on/off stuff.  I now just started the 4 hour process via the link above.  Hopefully that works.

Much appreciate your time Xaltar and Wardog.

Anyway, my boot drive is one of those Samsung VNAND M2 SSD drives that sits on the motherboard.  This drive has my OS, and all of my files and such.  I have a mirror image of my system from about a week ago on my mechanical storage drive, so I'm not going to lose much if things go pear shaped and I have to do a full system restore.  

Question, when I install this new MB, which is the same exact make/model of this seemingly dead one and I install my boot drive on the new MB, will it boot just like the same day the system failed, or am I going to have to go through additional steps because the boot "process" (for lack of a better term) is going to "see" a different MB and force me to go through additional steps?

Thanks.

Thanks wardog.  Will do.  Unfortunately I'm heading out of the country for a few days so won't be able to play with this problem until Wednesday.  Will report back with what's going on if you're still able to be around and so that if this happens to anyone else, they'll have some steps and (hopefully!) a final resolution.  

Parsec, thanks for responding.

Yes, the failure was sudden, without warning, without error messages, in what was a flawless system since being built last March.

I think the USB error was a Windows 10 error message.  When the system first failed and I thought I had a RAM stick problem, I removed 2 of the 4 sticks and started swapping them around trying to find which stick wouldn't allow the system to boot.  I was able to get sticks 1&2 to boot to BIOS, but saw that the BIOS was only recognizing 4GB of RAM, even though I had 8GB.  I then exited the BIOS and attempted to boot into Windows.

While booting into Windows 10, I got some sort of initial error message, Windows 10 did not shut down properly or similar, so it was going to do a repair or a special boot of some sort.  That's when I got the USB error message- while trying to boot into Windows 10.

I unplugged all USB items from the case except  my mouse.  The keyboard is in a serial port.  I tried rebooting, but at that point I couldn't even get the system to post to BIOS anymore.  

When I get back, the MB will have sit unplugged and without a battery for 4 days.  Before I install the new MB, I will try some more reboots and I will unplug the USB cables from the MB itself and see if that works.  

Thanks for the advice.  Will report back in a few days.

Parsec and Wardog-

Not sure if you guys are still around, but I purchased a new MB intalled it, and booted up the system.  No change.  It sounded promising at first with the fans running and some activity from the mechanical hard drive, but nothing else.  The monitor doesn't even come out of standby.  It won't post.

I thought maybe the video card was bad, so I took my kid's GEFORCE GTX 750 and plugged it in, no change.  Tried a different monitor.  No change.  

Did a CMOS reset with the reset button on the back of the motherboard, no change.

Not even sure where to go from here......

After a night's sleep has partially erased the frustration clouding my head, the ONLY thing I can figure is that I have a bad CPU or the new MB I purchased is also bad.  

I should have written in previous posts that after the initial failure, the monitor has never even come out of standby except once or twice after the initial failure when I was moving around the sticks of RAM to determine if it was a RAM problem.  After that, the system never POST'ed again nor did the monitor even come out of standby.

I meant PS/2 port with the keyboard.

Cleared CMOS with the button, computer unplugged, power button held in 10 seconds while computer unplugged, then pressed clear CMOS button.

I googled what happens when a CPU goes bad, and the behavior is very similar to what I am experiencing.  I bit the bullet and bought a new identical CPU ($400!) and it is arriving tomorrow.  

Last night, I did try many of the items you suggested except I did not try unplugging all the SATA drives AND the M2 drive.  I will do that tonight and report back.  I'm also going to put my multi-meter to each pin on every plug on my power supply and check for correct voltage.  

To add to the continuing saga.....

Put a brand new CPU in......same as before.  Computer turns on, fans turn, some mechanical hard drive activity heard, sounds promising, then nothing.  System just hangs with fans turning and no POST to BIOS.  Monitor doesn't even come out of standby.

Baffled.  I've tried two MB's, two CPUs, two PSUs, two graphics cards.  Unless all 4 sticks of my RAM went bad simultaneously or there is something wrong with my case, I am just utterly befuddled.

I'm heading out of town for a few days again, but when I get back I'm pulling everything out and putting it on cardboard and just getting back to basics.  If you have any other ideas, I'm all ears......

Have a good weekend.  Thanks for the help.

Yup, the monitor and video card are both definitely good. 

I am going to buy a couple of DDR4 sticks when I get home and try that to cover the remote(?!) possibility that all 4 of my relatively new DDR4 memory sticks failed simultaneously or nearly so.  

I will report back when I get home in a few days.

Thanks

Yes, I definitely kept notes and numbered the individual RAM sticks on the sticks themselves so I could keep track of what worked and what didn't.  I know sticks 1 & 2, and only those 2 sticks, allowed me to post to BIOS.....once.  I have been unable to post since then, even when using those same numbered sticks, but admittedly I haven't been resetting the CMOS every time I tried booting with each of the 4 sticks in pairs and individually.

I am going to buy a couple of new sticks of DDR4 to rule out simultaneous failure of all 4 of my memory sticks.

Thanks for sticking around.  Will report back when I get home.

https://www.amazon.com/gp/product/B00MMLUYPQ/ref=ox_sc_act_title_1?ie=UTF8&psc=1&smid=ATVPDKIKX0DER

OK, looks like you're talking about this stick. I have one of these sticks on the way.  Also shows as compatible on the ASRock website.  

Thanks.

Parsec and Wardog

Thanks!  I'm reassembling now.  Just want to get a clean boot to Windows, do a system image, then I am going to start dinking around with the individual sticks again and see if I can find the bad one(s).  

So you have had ALL 4 sticks fail on a matched 4 stick kit?  You'll have to explain that one to me as that is beyond my comprehension.

Parsec, Wardog, Xaltar-

Your help was much appreciated.  Obviously it was some sort of RAM issue, although I am still a bit befuddled........

I have the original 4 RAM sticks, I'll call them RAM 1, RAM 2, RAM 3, RAM 4 as i currently have then numbered.  They're 4GB each.

I booted up my computer with RAM 1 & RAM 2 in the A1 & B1 slots, Windows 10 said I had 4GB of RAM intalled when I have 8GB.  A bad stick I guess?

I booted up again, with only RAM 1 in A1, Windows 10 says I had 4GB of RAM installed.  Great.

I booted up again, with only RAM 2 in A1, Windows 10 says I had 4GB of RAM installed.  OK, I guess.  I thought this would be the bad stick since RAM 1 booted normally.

I booted up again, installed RAM 3 and RAM 4 in A1 & B1.  Windows 10 said I had 8GB of RAM.  Great.

I booted up again, left RAM 3 and RAM 4 in A1 & B1, put RAM 1 & RAM 2 in C1 and D1, and Windows 10 says I have 16GB of RAM.

So what's going on here? 

Contrary to popular thought, not all slots are equal on a motherboard. It's basic electronics really, each slot has a different length of circuit traces between it and the respective components it connects to. So while a module might not work correctly in say slot A1 it may work just fine in say slot D1 if slot D1 is closer to the RAM's power circuitry. Longer traces = greater resistance = signal degradation. I hope that makes sense....

That said, the RAM module causing the issue is still either faulty or not quite to spec with regard to voltage. I would test the system thoroughly, power off and on, disconnect from the wall for an hour or so and power on again from a proper cold boot a few times and then, if the RAM consistently shows up correctly in OS begin testing it with Memtest. If everything checks out then you should be good to go.  

Yes, that makes perfect sense.  Like wiring an electrical circuit for your home.  If you have a long length of wiring, you'll have greater resistance and can have voltage drop unless you use the appropriate guage of wiring.  I did't realize something like that would come into play with a MB.

I'll run MEMTEST on the sticks tonight.  

Thanks again for all of your help.  Much appreciated.

It definitely could be.  I will look in my BIOS and see if it's a particular slot that's giving me a problem if the memory issue shows up again.

Thanks for the videos.

Similar thoughts have been going through my head.  I've replaced everything with new, so I'm wondering if the RAM was OK all along.  I have this $400 CPU sitting here under warranty as was wondering whether I should even bother RMA'ing it, but I think I will as it could have been the CPU all along for all I know.

He-he.  No worries.  I come from an aviation background.....70-80% of aviation accidents and incidents are due to human error.  The statistics are similar with automobile accidents.  Along those lines, I would bet 70-80% of the questions you field are due to human error, and it is certainly possible that I didn't seat the RAM properly.  When I insert them, I always try to hear/feel the "click," but maybe the clicks weren't enough.

Thanks again for all of the help.  It was much appreciated.