OcUK System 'dead' - advice or return needed?

[Starwolf]

I purchased an OcUK Gaming Radiance Dawn system in April 2023, and it has been as reliable as one can expect of a Windows system - until recently.

I mostly use the system for AI Art with some light gaming. The AI Art makes heavy use of the GPU, and the UI runs in a web browser (I have used MS Edge for this, as I found it more reliable than Google Chrome) under Python, and Python started to crash regularly (hourly) with Event ID 1000, and sometimes I would see an error 0xC0000005 in the logs too. There would also be the occasional BSOD, and event logs would have the less than useful reason of BUGCHECK (sorry, I don't recall the IDs).

I would also get Event ID 1000 fails running games under Renpy.

Running DISM and SFC to check system integrity never revealed an issue.
Yesterday the problems became really bad, with even MS Edge failing with Event ID 1000 / Error 0xC0000005 continually. Having had enough of the issues, I decided to wipe the system drive and re-install Windows 11, so I downloaded the media creation tool to make a bootable USB stick with the OS on it, but when I ran the utility it would quit after a few minutes, and the Event log showed it failing with Error 0xC0000005.

I therefore used a different computer to create the USB stick.

1. The next step was to configure the machine to boot from USB in the bios and restart.
2. Windows installer ran from the USB stick and I told it to format the system disk and install the os
3. The 'Installing Windows' dialog appeared, and it copied Windows files (to the system disk)
4. It started the second step - Getting files ready for installation - but after a few percent failed with an Unexpected Error - unbelievably it proudly declared Error code: 0xC0000005

I have tried this a number of times and along the way created and used a different USB stick, but the error always happens during step 2, but not always at the same point - it is usually between 1% and 10% of the way through the step.

I am now at a loss as to how to proceed, and suspect I may have a hardware issue (the changing fail point during 'identical' install attempts suggested this). My initial thought was maybe the NVME storage or controller had issues, but I have seen threads listing similar symptoms on Reddit where the underlying cause turned out to be a bad CPU.

It is quite a high-end build, so I do not have any components that I can swap-in to do hardware checks.

Can OcUK suggest what I should try next, or is this a case of needing a return so that OcUK can check it out?

Thanks.

 

[Tetras]

Can OcUK suggest what I should try next, or is this a case of needing a return so that OcUK can check it out?

This is only a community forum, so I'm afraid you won't get a reply from OCUK themselves here, but from what you describe I also suspect there is a hardware fault and it needs to go back.

 

[Josh]

OcUK CS link is here - https://help.overclockers.co.uk/hc/en-gb - best to start a ticket with them and then will assist.

 

[Starwolf]

OcUK CS link is here - https://help.overclockers.co.uk/hc/en-gb - best to start a ticket with them and then will assist.

That page looked to me like it dealt with pre-sales stuff, but on further reading I think I found how to send a query like mine.

Thanks!

 

[Starwolf]

I have raised a ticket with OcUK and have an acknowledgement email...

 

[Josh]

I have raised a ticket with OcUK and have an acknowledgement email...

Excellent, if you want to attempt some troubleshooting, we do have community experts on the forums, but I'd wait for the shop to respond first.

 

[Starwolf]

I raised a ticket for the fault last Friday, so that's 4 days ago, including 2 full business days - the acknowledgement email said that OcUK try to look at tickets within 36 hours, but we are clearly beyond that now.
I will phone them in the morning, but does anybody happen to know what the current lead-times are for a response?

 

[tamzzy]

The last time I raised a webnote it took 4 business days to get a reply. I guess they are being snowed under. You could give them a call in the morning and they should deal with the RMA there and then.

 

[Starwolf]

The last time I raised a webnote it took 4 business days to get a reply. I guess they are being snowed under. You could give them a call in the morning and they should deal with the RMA there and then.

Oh dear - that is not exactly the best of service. I will give them a call tomorrow, as I hope they will agree that it is a hardware issue, requiring an RMA, so we can use the long response time to send the machine to them, rather than it just sit here doing nothing.

Thanks for the heads-up!

 

[ZXSpectrum]

Sounds to me like a memory or an SSD/IO error that is happening, generally i have found that issues with errors/blue screens during an installation that isnt very intensive use of the CPU is down to faulty memory or on as a long shot the SSD has problems with reading/writing data.

If you have access to another computer and a USB key, first thing i would do is create a bootable USB of memtest86 free (available here https://www.memtest86.com/download.htm) and boot from that and let it run through and see if there is any memory errors.

 

[Jay85]

Sounds to me like a memory or an SSD/IO error that is happening, generally i have found that issues with errors/blue screens during an installation that isnt very intensive use of the CPU is down to faulty memory or on as a long shot the SSD has problems with reading/writing data.

If you have access to another computer and a USB key, first thing i would do is create a bootable USB of memtest86 free (available here https://www.memtest86.com/download.htm) and boot from that and let it run through and see if there is any memory errors.

Does sound more like a failing drive tbh or could be something as simple as a failing Sata cable if its a SSD.

 

[Starwolf]

Sounds to me like a memory or an SSD/IO error that is happening, generally i have found that issues with errors/blue screens during an installation that isnt very intensive use of the CPU is down to faulty memory or on as a long shot the SSD has problems with reading/writing data.

If you have access to another computer and a USB key, first thing i would do is create a bootable USB of memtest86 free (available here https://www.memtest86.com/download.htm) and boot from that and let it run through and see if there is any memory errors.

Sounds reasonable, so I made a bootable USB and ran through the test as suggested. After a couple of hours (the machine has 32GB) the test completed successfully with no errors, so I think that eliminates a memory issue.

Thanks for the suggestion.

 

[Starwolf]

Does sound more like a failing drive tbh or could be something as simple as a failing Sata cable if its a SSD.

The Primary (OS) drive is a 1TB NVMe, but I also have a second drive (a 2TB NVMe) installed as a data drive.

I therefore tried installing an OS to the second NVMe, and it failed at a similar point, again with the 0xC0000005 error. In addition, I received the same error code while it was loading the installer from the USB stick (ie. before I had told it where to perform the installation), so I think that effectively excludes the drives as cause.

Thanks for the suggestion.

 

[ooStevo]

I've had HDD's going dodgy in the past. I couldn't figure out what was going on. Changing the Sata cable fixed it straight away. Maybe try that. I know you're using SSD but still.

 

[Starwolf]

I've had HDD's going dodgy in the past. I couldn't figure out what was going on. Changing the Sata cable fixed it straight away. Maybe try that. I know you're using SSD but still.

In my case, I have 2 NVMe drives, and the OS will not install to either, so I guess that the equivalent of a bad SATA cable would need to be an issue on the motherboard.

I suspect that the problem getting much worse in the same week that the story of sudden issues with Raptor Lake CPUs broke almost seems too much of a coincidence, so I would not be surprised if the machine comes back with a replacement CPU.

I have the RMA number now, and am currently awaiting pick-up.

 

[ooStevo]

If you're getting data error failures, it's the ram or storage. I've had Pc error issues in years gone by,and it was always ram or storage. Cpus don't really fail,and motherboards are robust as well. Saying that I did have a motherboard go up in smoke because I overclocked the fsb too much

 

[Starwolf]

If you're getting data error failures, it's the ram or storage. I've had Pc error issues in years gone by,and it was always ram or storage. Cpus don't really fail,and motherboards are robust as well. Saying that I did have a motherboard go up in smoke because I overclocked the fsb too much

The machine is going back to OcUK for repair, so we will find out what the issue is soon enough.

 

[ooStevo]

Well, let's see what happens I suppose.

 

[ED209]

The machine is going back to OcUK for repair, so we will find out what the issue is soon enough.

Keep us updated, would be good to know the outcome.

I hope your not without it for too long.

 

[Starwolf]

Update:

The machine was collected by DPD on Thursday 18th April.
I have POD of it arriving at OcUK early the following morning - Friday 19th April.
It is now a week later - in fact, 6 working days have passed, and I have heard no update from OcUK.

Does anybody know how long a system RMA is likely to take to process?

• Raised ticket - took 4 working days to respond.
• Agreed RMA - took 5 more working days before it was collected
• System back for repair - 6 more working days and counting...

Is there any way to get an update from OcUK without waiting until Monday and phoning support? I hoped that there would be a web portal that I can use to check the RMA ticket, but I cannot find one. The only online route that I can see is raising another ticket (which they don't encourage, and I fear it would take another 4 days before it is answered anyhow...)