Lines appear on display > computer locks up > reset > infinite beeping

[Tad Bittoomuch]

So, launching right in - I've recently started having this issue whereby my computer will inexplicably lock up and display white lines on the screen (See picture).  When I reset it by pressing the reset button on the case, it fails to post and continually beeps from that small speaker soldered onto the motherboard.  Infinitely.  Until I turn it off.

 

Doing so will result in the computer failing to post and beeping and power cycling.  The beeps it makes are short, and the interval between is even shorter.  Think of it as analogous to one of those reversing sensors in a car when you get too close to something.

 

What the computer is doing doesn't seem to have an effect on when these events will occur.  Sometimes I could just be staring at the desktop and it will lock up, or even editing a word document or watching a YouTube video.  It doesn't seem to matter.  Once I saw this happen when the computer was booting.

 

Things that may be relevant:

Operating System:  Windows 7 Professional 64bit
BIOS version: U1j  (I don't know either, that's just what AIDA64 tells me)

BIOS date: 20/10/2013

 

System Specifications:

CPU: Intel i5 3570k

CPU cooler: Corsair H60 120mm AIO

RAM: 32GB Kingston 1600MHz DDR3 (4x8GB)

Motherboard: Gigabyte GA-Z68P-UD3

Graphics: 2x Radeon HD 7970

Main SSD: Intel 535 series 480GB

Secondary SSD: Intel 535 series 480GB

Hard Drive: WD Blue 1TB

Power Supply: EVGA Nex650G1 (650W)

 

I should probably note that while a k skew, the 3570k is not overclocked, nor has it ever been.  I built the computer in 2012 and I've never had issues like this before.  The graphics cards are not overclocked either as they consume enough power and produce enough heat as it is.

Also, CPU temperatures are in the high 30s (°C) to low 40s (°C) when idling, and GPU temperatures are around 50°C at idle.

 

Thanks for having a look.  Much appreciated

[limegorilla]

Never seen this before, but maybe a GPU problem? Could you use the iGPU?

And when this started, had you cleaned the PC recently, or moved any components around? Are all the PSU cables connected correctly?

Do you have a on-mobo LED display for errors? Or can you look at your mobo manual to see if that gives an identification code to the beep sequence.

[VVoltor]

Some kind of hardware failure seems very likely. You'll have to do some testing to narrow it down.

Run memtest, to make sure it's not he RAM. Try using only 1 GPU, or maybe integrated graphics. Otherwise my money would be on the mobo gone bad.

 

Also, first thing I noticed looking at your hardware list, is you're running two power hungry cards on only a 650W PSU. Doesn't sound ideal

 

[Tad Bittoomuch]

23 minutes ago, limegorilla said:

Never seen this before, but maybe a GPU problem? Could you use the iGPU?

And when this started, had you cleaned the PC recently, or moved any components around? Are all the PSU cables connected correctly?

Do you have a on-mobo LED display for errors? Or can you look at your mobo manual to see if that gives an identification code to the beep sequence.

I was thinking that it could be graphics related.  I have a friend who said they had a similar issue (but without the beeping) whereby their display would periodically do the line thing.  Apparently it was due to GPU sag and they fixed it by tying a piece of string around a heat pipe and somewhere in the case.  It never happened again, or so I'm told.

I try to leave it alone - if it works then don't touch it, right?  I dust it, on average, once a month.

There are LEDs on the motherboard but they aren't really that helpful, and I did have a look at the motherboard manual and there's nothing about the beep sequence.

 

19 minutes ago, VVoltor said:

Some kind of hardware failure seems very likely. You'll have to do some testing to narrow it down.

Run memtest, to make sure it's not he RAM. Try using only 1 GPU, or maybe integrated graphics. Otherwise my money would be on the mobo gone bad.

 

Also, first thing I noticed looking at your hardware list, is you're running two power hungry cards on only a 650W PSU. Doesn't sound ideal

 

I did have a look at memtest, so while it was downloading I tried doing a memory stress test with AIDA64 for about an hour.  The test completed with no issues (compared to when I had a stick of RAM that I knew was bad, the AIDA64 test would fail within seconds).  I'll give the iGPU a go, see if that solves any issues.

As for the power issue, I might buy one of those wattmeter things to see how much power my system draws under full load, then I might see about getting a higher wattage power supply.

 

Thanks a lot, much appreciated

[gnomik]

If its a psus fault it would just power off, not do this weird stuff. It could just be the chipset, or damaged gpus, or dirty contacts.

[0x1e]

I would remove the gpus completely and use the igpu.

If it solves the issue then I would try one GPU at a time then take it from there.

[Tad Bittoomuch]

On 19/04/2018 at 10:35 PM, gnomik said:

If its a psus fault it would just power off, not do this weird stuff. It could just be the chipset, or damaged gpus, or dirty contacts.

I've cleaned the contacts of both the RAM and GPUs, also added some string so they won't sag.  This doesn't seem to make a difference, so I'm considering if I can get it to boot to Windows in safe mode I'll update the motherboard BIOS.  Not sure what the GPU bioses are, or even if they're the same.  Could the fact that the GPU bioses don't match be what's causing this issue?

 

On 19/04/2018 at 11:08 PM, 0x1e said:

I would remove the gpus completely and use the igpu.

If it solves the issue then I would try one GPU at a time then take it from there.

I removed both GPUs and tried with the iGPU - still the same problem.  I might have to reinstall Windows, even through I don't really want to.  Since the last time I've reinstalled Windows I've had this version for a couple of years and I like the way my filesystem works with all my libraries and pinned folders.

 

I ran Memtest+, which turned out that all the memory is in complete working order.  I also replaced the CMOS battery.

[TechyBen]

On 19/04/2018 at 1:35 PM, gnomik said:

If its a psus fault it would just power off, not do this weird stuff. It could just be the chipset, or damaged gpus, or dirty contacts.

Nope. If the voltage regulator fails, it could cause any type of fault, no? So AFAIK it would not just shut down. I've had a PC show no signs of problems, but crash in Windows 10 after 5 mins, and only a PSU swap fixed it.

 

[edit] Though I've no idea if any of the microcode updates for the spectre etc updates are causing crashes still?

[gnomik]

15 hours ago, TechyBen said:

Nope. If the voltage regulator fails, it could cause any type of fault, no?

If there is a serious issue like that in the psu, it would power itself off.

[Tad Bittoomuch]

Solved:  To all those wondering, the issue was apparently to do with the auxiliary hard drive preventing the computer from posting.  I unplugged it and the computer booted to Windows instantly, then I updated the motherboard BIOS and ddu'ed the graphics drivers.  I reinstalled the hard drive and the one graphics card, reinstalling the graphics drivers and then rebooted and reinstalled the other graphics card.

 

So far it's been working perfectly.  Glad I don't have to build a new PC, especially at the moment with the price of DDR4.

 

Thanks for all your help and suggestions.