PC build, year and a half of hell! Few questions regarding my rights.

[Mikeadamwood]

Hi,

I'm a motion 3D/artist, and had a high(ish) spec custom PC built for GPU rendering by a local PC builder after deliberating between them and Scan. Decided to go to the local one as it would offer me the doorstep support I hoped I'd never need. Turns out I did, but it has not been the class of support I anticipated.

I don't want this post to be too long but basically it's been a year and a half of hell! The company have provided me support, but a year and a bit later they still haven't been able to get to the bottom of it.

Some of specs are from memory as the PC is currently at the store:
Intel Core i7-6850k 3.60GHz
ASUS 2011-V3 X99-E WS Motherboard
BeQuite CPU fan
Corsair RM1000x RMx Power Supply
Samsung 1TB M.2 SSD
Seagate 3.5", 3TB, SATA3, SkyHawk Surveillance Hard Drive
Thermaltake Black Core X5 EATX PC Gaming Cube Case
32Gb Corsair Vengeance, 4x 8GB DDR3, 2800Mhz (Had to guess this as it's not the same as the quote, and machine is with them so can't check)
2x EVGA GeForce GTX 1080 Hybrid

Bluescreens if left on while I'm at work, due to the nature of my work the machine needs to be on as things are rendering, but even if I'm not rendering it will decide to restart itself from a BSOD. I've tried everything now, and so tired trying to diagnose the problem..Just want a machine to allow me to do my work! I bought the Hybrid cards so they would keep the cards cool when rendering for days at a time, the purchased machine can't even make it through a day without shutting down.

Here is reliability monitor showing bluescreens everyday.

Backwards and forwards for over year, we'd think we have it down then it starts again. So recently they took my machine and gave me a temporary machine and drive copied my harddrive to it (only the C drive) they also moved over my D drive so I'd have my projects. On the loan machine I get the same issues, so I took it upon myself to format the C and start fresh with windows to see if it was a driver issue, after a few days I'm getting the same problem again where it restarts randomly when out.

No idea how this could happen on 2 machines with the only thing the same being my D drive with all my projects on. This drive causes me no issues in other areas so can't see how it could be this causing it.

Now 46 emails later, 5 trips over the year to the store for them to look at the machine. It's now been with them for almost 2 weeks and they haven't given me a diagnosis yet.

Now what would you guys do in my situation, which is causing me a huge amount of stress. I feel like just asking for a refund...I'd love to just start again, somewhere like SCAN, but not sure if I actually have a legal stance on it being over a year (even though the problems have been from date of purchase) Feel so stuck and fed up, and my personal work hasn't seen the benefit of what I bought this machine for.

Would really appreciate some advice.
Thank you in advance!

 

[SkyNetRising]

1st - you have to monitor temperatures of cpu and gpus.
2nd - you have to diagnose BSODs.

 

[gumbrellr]

Yeah random blue screens suck, I've dealt with more in the past year than the past ten! No idea where you stand legally but i'd look at the minidumps they usually offer some kind of clue. Nirsoft have a blue screen viewer which makes it easy to analyse them if you don't want to fath with win debugger. Otherwise anything common to the two machines hardware or software could be suspect.

 

[punkncat]

I tend to agree with the above poster, in that from what you have posted we have no information beyond your reliability issue. Taking a look at your build I would instantly suspect air flow in that case. Your mentioning it running right after updates kinda throws me though. Really need to know what the code is on the BSOD screen.


Aside from the technical...I would think that if you have had an open issue with this store this whole time, it's probably unreasonable to consider a refund. A darned good discount on a new build, a partial refund, etc. should be likely of consideration. In reality, if they are continuing to try and rectify the issue you really may not have a legal leg to stand in that they are trying to work with you to fix it in appearance anyway.

 

[Mikeadamwood]

1st - you have to monitor temperatures of cpu and gpus.
2nd - you have to diagnose BSODs.

1. I've monitored CPU and GPU temps, in fact I always do this as it's the whole reason I bought the Hybrid cards, they are always normal temperatures...

2. I've tried this via blue screen dump

Don't mean to sound rude, and honestly don't take this the wrong way but your answer was rather vague and not that helpful. This has been going on for over a year if you read my post, so routine checks like that have been done time and time again.

If I didn't provide enough information then please let me know and I can try to provide it you.

 

[Mikeadamwood]

Yeah random blue screens suck, I've dealt with more in the past year than the past ten! No idea where you stand legally but i'd look at the minidumps they usually offer some kind of clue. Nirsoft have a blue screen viewer which makes it easy to analyse them if you don't want to fath with win debugger. Otherwise anything common to the two machines hardware or software could be suspect.

I have used this, albeit more recently..Here's one of the dumps,

Unfortunately only an image, like I said my machine is in the store currently so pulling these all from emails. When I get the machine back, do you have tips on analysing these dumps with NirSoft or should I just post on here for someone to look at?

I tend to agree with the above poster, in that from what you have posted we have no information beyond your reliability issue. Taking a look at your build I would instantly suspect air flow in that case. Your mentioning it running right after updates kinda throws me though. Really need to know what the code is on the BSOD screen.

Aside from the technical...I would think that if you have had an open issue with this store this whole time, it's probably unreasonable to consider a refund. A darned good discount on a new build, a partial refund, etc. should be likely of consideration. In reality, if they are continuing to try and rectify the issue you really may not have a legal leg to stand in that they are trying to work with you to fix it in appearance anyway.

Thanks for the response and advice. If you mean the actual code on the blue screen it tends to be either IRQL NOT LESS OR EQUAL or KMODE EXCEPTION NOT HANDLED

I thought so much from the legal standpoint..Literally have zero rights at this point. I was trying to kill 2 birds with one stone with this post..Advice on where to turn next, might just be the cherry on top in starting again and ordering a build from Scan.

 

[punkncat]

Selling used PC parts can be a real crapshoot, but perhaps the bones of this system might be well served parted and elsewhere? I know you could make your purchase price on the video cards in a moment right now...well, at least on lower end cards. Those 1080 are pretty much a focused group.

My thoughts would be that IF you can afford to buy another system, do so. Get a prebuilt option, or something that an aftermarket builder sells a LOT of, with good support. So long as the builder for this machine agrees to work with you, has value...but at the same time you can't use a machine sitting in the shop and the value of those parts reduces every day.

 

[Mikeadamwood]

Selling used PC parts can be a real crapshoot, but perhaps the bones of this system might be well served parted and elsewhere? I know you could make your purchase price on the video cards in a moment right now...well, at least on lower end cards. Those 1080 are pretty much a focused group.

My thoughts would be that IF you can afford to buy another system, do so. Get a prebuilt option, or something that an aftermarket builder sells a LOT of, with good support. So long as the builder for this machine agrees to work with you, has value...but at the same time you can't use a machine sitting in the shop and the value of those parts reduces every day.

Sound advice, thank you.

I'm definitely thinking of biting the bullet, has gone on far too long for me to even consider this thing will ever be resolved. It's the biggest and worst investment I ever made.

 

[SkyNetRising]

I have used this, albeit more recently..Here's one of the dumps,

What driver is listed in column "Caused by Driver"?
Generally IRQL_NOT_LESS_OR_EQUAL or KMODE_EXCEPTION_NOT_HANDLED signals about driver issues, also can be caused by overheating.
Can you show screenshot from MSI afterburner hardware monitoring window, while system is performing high intensity work? Graphs important - cpu usage/temp, gpu usage/temp.

Spoiler

MSI afterburner hardware monitoring window looks like this:

 

[gumbrellr]

Yes what does the probably caused by field say?

Also,

Does it blue screen when you're using it or just when you go out and leave it rendering?

If you leave it doing nothing what does it do?

Is the hardware the same as the original machine?

The bugcheck suggests a driver error. You can try using driver verifier which is built into windows.

Have a read here.

https://www.kapilarya.com/how-to-use-driver-verifier-in-windows-10

Check you're ok with disabling the verifier before you enable it.

If it's a problematic driver I would expect it to blue screen on boot. The minidump from this might give a better clue but in my mind it would at least confirm if it is a driver error or not.

The other option is to perform a clean boot and see if the issue re-occurs. Clean boot will disable all non Microsoft services/software so will tell you if it's something installed on the system rather than a driver.

http://support.logitech.com/en_us/article/228

If it's a temp issue I'd suspect a simple stress test would re-produce the issue.

It's a shame the system builder isn't systematically doing this for you. There is no silver bullet when it come these kind of things just a logical process to try and isolate the problem.

 

[Mikeadamwood]

Apologies for late reply saw your messages but haven't been able to jump on and reply till the end of the day.

What driver is listed in column "Caused by Driver"?
Generally IRQL_NOT_LESS_OR_EQUAL or KMODE_EXCEPTION_NOT_HANDLED signals about driver issues, also can be caused by overheating.
Can you show screenshot from MSI afterburner hardware monitoring window, while system is performing high intensity work? Graphs important - cpu usage/temp, gpu usage/temp.
MSI afterburner hardware monitoring window looks like this:

Thanks for the info! I should be getting the machine back this weekend so will grab afterburner once I do and get detailed reports of heat like you mentioned. I'd like to say the problems will be sorted but I highly doubt they will after all this time, very little confidence with the guys that built the machine. I have used other tools to monitor but not as detailed as the graphs within that software.

Yes what does the probably caused by field say?

Also,

Does it blue screen when you're using it or just when you go out and leave it rendering?

If you leave it doing nothing what does it do?

Is the hardware the same as the original machine?

The bugcheck suggests a driver error. You can try using driver verifier which is built into windows.

Have a read here.

https://www.kapilarya.com/how-to-use-driver-verifier-in-windows-10

Check you're ok with disabling the verifier before you enable it.

If it's a problematic driver I would expect it to blue screen on boot. The minidump from this might give a better clue but in my mind it would at least confirm if it is a driver error or not.

The other option is to perform a clean boot and see if the issue re-occurs. Clean boot will disable all non Microsoft services/software so will tell you if it's something installed on the system rather than a driver.

http://support.logitech.com/en_us/article/228

If it's a temp issue I'd suspect a simple stress test would re-produce the issue.

It's a shame the system builder isn't systematically doing this for you. There is no silver bullet when it come these kind of things just a logical process to try and isolate the problem.

Thanks for the detailed response! I'll have to get back to you with the DMP file info as I should get the machine back this weekend as I mentioned above, I have all the dump files on there but will be running stress tests to gather more data about these crashes if they persist, which I think they will given the history. As for the blue screening while on the machine, it has done quite a lot during use but usually it's when left, and often when I'm not even rendering..I'll know it's restarted as I will have apps I've opened which have closed while I'm out, often losing work which I have become wise to before leaving, so tend to double check I've saved everything. It's difficult to remember a time when the machine has ran for more than 10 hours straight...Where the machines at my work haven't been restarted for literally months and they are used for rendering all the time, we tend to just turn the displays off when leaving.

In terms of them stress testing, they are trying to isolate the problem, but feel as if the components in this build are really not what they usually deal with and don't feel all too confident at their approach.

Thanks again for responses, will get back to you this weekend!

 

[Mikeadamwood]

Quick question... The temp machine bluescreened this morning and the dmp file is 963mb. Is this a normal size for a crash file?

Also what's the best way to share the info from the dump file? Is there a way to export info from bluescreen view that isn't huge like this file?

 

[SkyNetRising]

Memory dump size depends on amount of ram in your system.

Set memory dump to small.

 

[smorizio]

as a tech i can see one pc having issue but a replacement pc with the same issue points to something going on at your work. i would start and have your local power compnay check the main line going into your building and the meter make sure there a good ground and there no surging and power is good on all bus bars in your fuse panel. have a local electrtion if you have alumimim wires to make sure there all locked down. non copper wires you need to check that there locked in every few years. have them check the amp load on the outlet that pc having issues make sure there no large motors from friges or ac on the same line. if there a lot of pc or printer on that line have someone run a new lines to kep the amp load withing amp loads. if your using a battery back up and or cheap ac splitter look at getting one of the good comercial pc ac splitters thta show load and temps. on the pc itself make sure the mb has the newest bios file. check with the store and the power supply vendor make sure the power supply is one of the newer haswell supported unit. the older non haswell power supplys the 12v leg when haswell goes into it high end sleep mode the load on the 12v rail is so low the older power supply will shut off or cause errors. a quick test of this is turn off onboard bios power savings and turn off power savings in windows. just have the pc turn the monitor off after x amount of time. also in the bios and windows turn off power savings for the pci bus. as your using a m2 ssd if the pci bus goes to sleep it may be causing issues with the drive. contact asus for there updated ram that been tested on this mb. see if the shop has two 16g dimms. or 4 other dimms that are on the qal list. (have them put in ecc ram and not non ecc gaming ram. see if the ram was the issue. also have the shop and yourself turn off power saving for the ssd and hard drives. with ssd if the power not clean to them there will bsod. have the shop check all the drives firmware make sure there up to date. the last issue is old software and windows 10. some old software like nortons and mcafee used to place scripts in windows to run scans when pc goes to sleep. these old scripts and windows 10 did not like each other. on returning the pc or getting your money back look at your state lemon laws. all states have them. if the vendor wont return your money after you gone though everything that you think is the issue. you may have to file small claims and use your states lemon laws. most times you win if you can show the pc been in the shop more then it been working.

 

[Mikeadamwood]

Memory dump size depends on amount of ram in your system.

Set memory dump to small.

Thanks will make sure I do that! Wondered why my dumps were allways struggling to write. This definitely clears that one up!

as a tech i can see one pc having issue but a replacement pc with the same issue points to something going on at your work. i would start and have your local power compnay check the main line going into your building and the meter make sure there a good ground and there no surging and power is good on all bus bars in your fuse panel. have a local electrtion if you have alumimim wires to make sure there all locked down. non copper wires you need to check that there locked in every few years. have them check the amp load on the outlet that pc having issues make sure there no large motors from friges or ac on the same line. if there a lot of pc or printer on that line have someone run a new lines to kep the amp load withing amp loads. if your using a battery back up and or cheap ac splitter look at getting one of the good comercial pc ac splitters thta show load and temps. on the pc itself make sure the mb has the newest bios file. check with the store and the power supply vendor make sure the power supply is one of the newer haswell supported unit. the older non haswell power supplys the 12v leg when haswell goes into it high end sleep mode the load on the 12v rail is so low the older power supply will shut off or cause errors. a quick test of this is turn off onboard bios power savings and turn off power savings in windows. just have the pc turn the monitor off after x amount of time. also in the bios and windows turn off power savings for the pci bus. as your using a m2 ssd if the pci bus goes to sleep it may be causing issues with the drive. contact asus for there updated ram that been tested on this mb. see if the shop has two 16g dimms. or 4 other dimms that are on the qal list. (have them put in ecc ram and not non ecc gaming ram. see if the ram was the issue. also have the shop and yourself turn off power saving for the ssd and hard drives. with ssd if the power not clean to them there will bsod. have the shop check all the drives firmware make sure there up to date. the last issue is old software and windows 10. some old software like nortons and mcafee used to place scripts in windows to run scans when pc goes to sleep. these old scripts and windows 10 did not like each other. on returning the pc or getting your money back look at your state lemon laws. all states have them. if the vendor wont return your money after you gone though everything that you think is the issue. you may have to file small claims and use your states lemon laws. most times you win if you can show the pc been in the shop more then it been working.

Thanks a lot for the detailed response! I'll be sure to send this info over to the builders to look over, of course taking out the legal part.

The electric line had ran past my mind, as the house I live in hasn't been rewired for a long time and is ready to be (but of course this is a huge job) I'd always laughed this off as it feels like I'm, clutching at straws, but given your advice I'll consider the speculation as somewhat valid and worth looking into.

Thanks again!