piggybank

I have a desktop computer giving problems, following info.

The last board failed CPU and connections to the CPU by corrosion – I am near the sea.

I bought a new motherboard, new CPU, new 8 GB x 2 memory, new case, and new power supply. Old hard drive, old DVD drive. Genuine programs Windows 10 64 bit is up-to-date, MS Office and others. New WI-Fi keyboard and mouse. No video card fitted I am using on-board video. Motherboard and CPU have a one year guarantee as of 26 November 22.

Gigabyte Technology Co. Ltd. H61M-DS2 3.0 (Intel Core i3-2120 CPU @ 3.30GHz)

Problem. Computer will not allow ReadyBoost; the computer shuts itself down unexpectedly after three hours use and more often twice a day crashes with black and white squared raster over the monitor, sometimes with load noise sometimes quiet. It may shut down after a few minutes and has had red and white raster on the monitor.

New keyboard has three letters that now will not do upper case; all other keys will do upper case.

Intermittent vertical yellow line on monitor screen that is on for some uses of the computer other times not.

Checks.

All programs up to date, all drivers up to date. Windows memory test shows OK. Reliability Monitor shows shutdown as unexpected but no reasons. Going back weeks on the Reliability Monitor several programs affecting the shutdown so these were removed from the start-up sequence. The CD from a previous version of the board does not give a video driver for the newer version of the board. Cannot do a Western Digital test of the hard drive as the computer crashes in the long time needed to do a full scan. Quick tests claim the drive is good.

Choices

I reckon I have tried everything. What’s the best option just change the motherboard or dig deeper into the software as the source of the problem?

Jhieminga

The only things I can think of:

1. Check the rating of your power supply, is it sufficiently powerful to run everything you've got hooked up to it?
2. As most of your stuff is new, you shouldn't be getting failures there but there may still be a component that failed straight out of the box of course. I would disconnect anything that's old, try to get it running on the new stuff first and see if that works without these issues turning up. Create a bootable USB with Linux on it or something like that and run from that so that you can avoid the old HDD. See if you can use a keyboard/mouse with wires to avoid connection issues there.

Others may well have better advice...
Edited to add: how stable is your power supply from the outlet? If the input power is wonky, you may get strange effects although a computer PSU may be able to deal with some variations.

andytug

As has been said could be power supply fluctuation, also - Is the loud noise fan noise? Overheating of the CPU/GPU (may be combined on same chip) can also cause shutdowns. A temp mointoring program may show this up.

Also worth downloading CCleaner from Piriform (free) and run the cleaner and the registry cleaner bits.

rans6andrew

I had a very similar problem on a machine I built up from new parts but ran Ubuntu rather than Windoze. I had crashing at random intervals and failed to find any specific cause. Over the next couple of years I replaced almost every part eiother by buying new or by exchanging with the equivalent parts from another computer. Nothing made any difference except underclocking reduced the incidence a bit. When I had exchanged everything except the case and the PSU I had enough spares to build a second system. As a last resort I replaced the PSU and the problem went away. I knew the fix was a good un when I reversed the underclocking and started to experiment with overclocking........... I nearly made the machine fast enough to make up for two years of slow/intermitent working and I had a spare machine ready just in case.

Rans6..........................

terrain safe

The Processor was launched in 2011 and the Motherboard had its last driver update in September 2015. It may be new to you but it either has been used elsewhere or has been sat on a shelf for a very long time. Check the thermal paste between the heatsink and the top of the processor. If windows 10 is up to date then there may well be a driver problem so check device manager as well.

Clay_T

There are some utilities in Windows that might help identify the problem.

Event Viewer under Administrative Tools in the Control Panel.
(...no idea where Microsoft hides Event Viewer in Windows 10.)

Look through the Windows Logs there. There are probably hundreds of messages in them.
Clear the logs, wait for the problem to happen again then check the logs.

Boot into the BIOS and have a look around there. There's a section that shows temps, voltages, etc.
Should be able to see if the CPU temp is in line.
Careful poking around there. Changing the wrong thing and saving the setting can really bork the system.
Best option is choose "Exit without saving" (unless you know what you're doing).

Thrust Augmentation

Sounds like it could be PSU or motherboard - seen very similar issues with an new MB that had faulty capacitors from the get-go.

atakacs

Does it also happen if you boot into some live linux from an external USB (just to get make sure it is not OS related but a hardware issue) ?

piggybank

Thank you for the helpful input, some items I have done and others to try.

1/ Power supply I reckon is OK as when I get the rastered screen the previous page is still there in the back ground. But will keep that on my check list. I will connect a voltage stabiliser I have in stock for one thousand watts.

2/ CC Cleaner is used a few times a week on registry and cleanout. I set the warnings for temperatures lower than the defaults.

3/ I will go to the Gigabyte site for updates. When the CPU was changed new thermal paste was used as a new fan was fitted at that time.

4/ Device manager says all drivers are OK. Ref the BIOS, it opened itself up tonight so I set a larger memory for the internal video. It was 64 MB and now 128 MB.

Event Viewer may be the Reliability Monitor, in Win 10. Not much info in that just ‘your confuser shut down unexpectedly’.

Once again thanks for the input, I will check these answers out.

golfbananajam

Not sure about that, see https://www.freecodecamp.org/news/ev...and%20warnings.

piggybank

That's interesting. I am looking at it now. Thanks

piggybank

The board appears to be 'new'. Its popular here in Indonesia by price and gives enough performance. The shop had a stack of these boards in boxes. Price is OK at GBP 75 for the new board and a new CPU. The legacy plugs on the back are useful to people that still need the 21 pin connector.. I have had three of these of the many years they have been sold here. That price included a new case and power supply.

piggybank

Thank you all for the useful input. The memory that was bought 4 months ago had failed despite passing the Windows Memory Test. Lifetime guarantees mean nothing here in Bali. I have a stack of failed memories and video cards. Other than refitting the good memory I refitted the video card after putting new heat sink compound on the chip as I was concerned about temperatures. Its running at 52C at present, which is OK as we are generally at 26 to 28C at night. The computer has run for 10 hours so it looks like the problem is gone.