AMD Zen Discussion Thread

Cuddlesworth · 11-02-2020 12:41pm

.G. wrote: »

You can submit the dumps on the Windows community forums and get them looked at.

So people taking a pot shot at it? Because I know one person who can actually read them properly(outside of looking at the basics) and to say its a skill set is a understatement.

.G. · 11-02-2020 6:13pm

Cuddlesworth wrote: »

So people taking a pot shot at it? Because I know one person who can actually read them properly(outside of looking at the basics) and to say its a skill set is a understatement.

When I googled my issues I came across a few pages which seemed to have Microsoft people responding to them, maybe I wasn't paying proper attention.

I don't see the issue though, people here have tried to help, the people there will too. I'll take any help I can get to be honest. I can't debug them properly I'm open anyone who apparently can, not sure what else anyone is supposed to do for help with debugs apart from learn to do it themselves!

Haven't had a BSOD yet since i did what I did they other day though. Still many crashes with Ubisoft games in the main, so I'll be hitting their tech dept first.

.G. · 11-02-2020 11:26pm

Jinxed it. BSOD tonight page fault in non paged area. Only hardware I haven't changed in the machine is the cpu, mobo and 550w PSU.

Might try an older bios version.

Cuddlesworth · 12-02-2020 10:56am

Stress test the cpu, watch the temps at the same time.

.G. · 12-02-2020 12:50pm

Cuddlesworth wrote: »

Stress test the cpu, watch the temps at the same time.

Thanks. With prime95 or something else?

I might put my gpu and psu in it for a while too, that will mean only the mobo and cpu left from the original build. It's not a great case for airflow though and my gpu is a 2080ti. His gaming temps are all fine in it with his 1080.

H3llR4iser · 12-02-2020 6:57pm

.G. wrote: »

Thanks. With prime95 or something else?

I might put my gpu and psu in it for a while too, that will mean only the mobo and cpu left from the original build. It's not a great case for airflow though and my gpu is a 2080ti. His gaming temps are all fine in it with his 1080.

Your case is a mystery at this point. I would recommend - but that's probably been said already - to run a stress test with the bare minimum components to get the system running and work up from there. TBH, everything seems to be pointing towards a defective mainboard - or even the CPU, but I'd look at the mobo first.

Any chance you could test the CPU on a different board?

Prime95 or OCCT would be good candidates, AIDA64 too.

.G. · 12-02-2020 8:36pm

Yeah I'm thinking mobo too or maybe just a sh!tty version of BIOS. MSI don't seem to be paying much attention to the gaming plus as regards BIOS updates. I've no way of testing the cpu on a different board without buying one but I may just have to do that to rule one or the other out. He has game crashes daily but BSOD not as often, once yesterday and today but not for a few days before that so I'd need the board for a few weeks to be sure of things.

smacl · 14-02-2020 2:12pm

Cuddlesworth wrote: »

So people taking a pot shot at it? Because I know one person who can actually read them properly(outside of looking at the basics) and to say its a skill set is a understatement.

To be fair, you can get quite a bit from the basics. Opening the debug file in a debugger will often give you a stack trace to the faulting module. If you've had a few BSODs and they all fault in the same module and the same address, there's a fair chance it is a driver issue. I've had this very recently with the latest version of Acronis where their security offering BSODs when I plug in my Garmin. I use VS2017 for debugging but WinDgb will also do the trick nicely. Once you know the faulting module and can figure a matching .DLL, EXE or similar executable you can check properties, or just uninstall and see if it fixes the issue. More often then not, you won't have the symbols to do a more detailed analysis though you can get many of them from the MS symbol server site. If you check a number of BSOD dump files and they're all happening at different locations, it is more likely an intermittent hardware issue.

If you do get a dump file pointing to a driver or kernel code, it can be really useful to the developer in pinpoint an issue and they'll thank you for passing it on. Not sure what MS does with dump files that indicate 3rd party code issues but not much at a guess.

.G. · 14-02-2020 4:38pm

smacl wrote: »

To be fair, you can get quite a bit from the basics. Opening the debug file in a debugger will often give you a stack trace to the faulting module. If you've had a few BSODs and they all fault in the same module and the same address, there's a fair chance it is a driver issue. I've had this very recently with the latest version of Acronis where their security offering BSODs when I plug in my Garmin. I use VS2017 for debugging but WinDgb will also do the trick nicely. Once you know the faulting module and can figure a matching .DLL, EXE or similar executable you can check properties, or just uninstall and see if it fixes the issue. More often then not, you won't have the symbols to do a more detailed analysis though you can get many of them from the MS symbol server site. If you check a number of BSOD dump files and they're all happening at different locations, it is more likely an intermittent hardware issue.

If you do get a dump file pointing to a driver or kernel code, it can be really useful to the developer in pinpoint an issue and they'll thank you for passing it on. Not sure what MS does with dump files that indicate 3rd party code issues but not much at a guess.

I've used Whocrashed to debug the last two BSODs and both are pointing at a Windows kernal error but the program can't be certain if it's caused by a third party driver or not and if so which one! It's certain it's a software issue though. Just had another BSOD but weirdly the dump file is missing.

I've ubisoft tech support trying to help at the moment but they haven't asked for dump files.

We're probably way off topic here as I suppose it's not really a AMD zen issue so if mods want to move all this to its own thread feel free!

Cuddlesworth · 14-02-2020 6:16pm

smacl wrote: »

To be fair, you can get quite a bit from the basics. Opening the debug file in a debugger will often give you a stack trace to the faulting module. If you've had a few BSODs and they all fault in the same module and the same address, there's a fair chance it is a driver issue. I've had this very recently with the latest version of Acronis where their security offering BSODs when I plug in my Garmin. I use VS2017 for debugging but WinDgb will also do the trick nicely. Once you know the faulting module and can figure a matching .DLL, EXE or similar executable you can check properties, or just uninstall and see if it fixes the issue. More often then not, you won't have the symbols to do a more detailed analysis though you can get many of them from the MS symbol server site. If you check a number of BSOD dump files and they're all happening at different locations, it is more likely an intermittent hardware issue.

If you do get a dump file pointing to a driver or kernel code, it can be really useful to the developer in pinpoint an issue and they'll thank you for passing it on. Not sure what MS does with dump files that indicate 3rd party code issues but not much at a guess.

I was just curious what level went into it on community forums. I thought I could read them well(a long time ago) but the only option was to teach yourself.

smacl · 15-02-2020 9:21am

Cuddlesworth wrote: »

I was just curious what level went into it on community forums. I thought I could read them well(a long time ago) but the only option was to teach yourself.

I'm coming at it mostly from the software side where Stackoverflow and Serverfault can be pretty good if a bit snarky and MSDN tending to throw up a lot of similar answers from MVPs who tow the Microsoft line a bit too much. I process crash dump files regularly, but they tend to be application level and I have the symbols (e.g. PDB files) and code knowledge to make sense of what is going on. Without symbols best you can do in my experience is find out faulting module and send the dump file on to them. A dump file relating to a driver or BIOS issue is going to be no use to Microsoft for example. The strategy I use these days is simply to send the dump file to all the suspected sources and see if one of them sticks.

Totally agree with the self learning thing but the community forums are a huge asset to achieving this.

.G. · 15-02-2020 9:34pm

Are they're any non MSI B450 or x470 boards out there that are Ryzen 5 ready out of the box? Or any that have a bios flashback facility.

Might buy another board just to see if it helps.

15-02-2020 9:40pm

I think a few of the higher end Asus boards might have flashback.

.G. · 15-02-2020 10:40pm

Probably be pricey too I suppose! Was hoping for a budget option since it may not be the issue anyway! All the Budget options with flashback are MSI.

savemejebus · 15-02-2020 11:51pm

.G. wrote: »

Probably be pricey too I suppose! Was hoping for a budget option since it may not be the issue anyway! All the Budget options with flashback are MSI.

Was your mobo a MAX version? If not then a maybe an x470 or b450 MAX would be a budget option. I get it's still MSI but at least you wouldn't need to worry about a bad bios flash.

.G. · 16-02-2020 12:07pm

No not a Max. I checked the last bios revisions on the other msi boards and they were all November like our one. Obviously there's loads of people using them without issue but it just worries me that I buy one and it solves nothing!

.G. · 16-02-2020 1:14pm

Just remembered CCL do a free bios update so that solves that. Have MSI and Ubisoft on the case so will see what they come up with first.

.G. · 17-02-2020 7:35pm

MSI have thrown their hat at it and asked me to follow up with my retailer for recourse. Thankfully it was bought from Amazon who as we know are absolutely sound, they've offered a return for full refund so I'm taking the chance and I've bought the newer MAX version off them and hope for the best. At least I'll know soon enough if it was the motherboard after all! Should have the new one soon and will send the old one back when I get it.

cherryghost · 18-02-2020 3:33pm

Good old MSI. The worst customer support experience I've dealt with in the PC industry, by far.

wotzgoingon · 18-02-2020 11:39pm

Old news but just remembered it again. 3990X runs Crysis without graphics card.

Serephucus · 18-02-2020 11:46pm

Give it 5 years at the current pace, and it'll be running 1080/60 maxed out. :pac:

Cuddlesworth · 21-02-2020 11:52am

AMD putting the screws to intel, price drops to the 3k series on top of the super low prices on the last 2 generations.

.G. · 21-02-2020 4:31pm

Interesting. Would have expected them to hold off on price drops until the 4th gen drops.

.G. · 24-02-2020 10:44pm

New motherboard installed about 3 hours now and one game crash and a BSOD already :rolleyes::rolleyes::mad:

wotzgoingon · 24-02-2020 10:48pm

.G. wrote: »

New motherboard installed about 3 hours now and one game crash and a BSOD already :rolleyes::rolleyes::mad:

Not good. I don't know what is wrong. Must be memory. Are you overclocking?

.G. · 24-02-2020 11:02pm

Nope. Everything at stock in the BIOS. Interestingly, I put his Ram in my machine for a while and he had mine, and then swapped back again but since then I've had three BSOD's myself having not had any for months! My ram possible screwed by his machine?!

Every BSOD he gets is a different error code too.

wotzgoingon · 24-02-2020 11:12pm

I feel for you and it's a odd one. I honestly have had no issues with my system except for one time the bios reset itself but I think I know what caused that as I have a older second GPU in my system that I run a old version of Linux on and forgot to switch back to the 5700 XT to run my Windows 10.

.G. · 25-02-2020 12:35am

Ubisoft had no answers up to now still investigating. Mainly seems to happen when he's playing their games. Neither did msi so just at the stage now where we just have to put up with it. Had another BSOD after my earlier post. Considered buying another 3600 but just going to leave it now and hope it doesn't get worse.

savemejebus · 25-02-2020 11:09am

.G. wrote: »

Nope. Everything at stock in the BIOS. Interestingly, I put his Ram in my machine for a while and he had mine, and then swapped back again but since then I've had three BSOD's myself having not had any for months! My ram possible screwed by his machine?!

Every BSOD he gets is a different error code too.

More likely would be his bad ram corrupting your windows install. I think that i read about that happening before but that was from running unstable overclocked ram.

.G. wrote: »

Ubisoft had no answers up to now still investigating. Mainly seems to happen when he's playing their games. Neither did msi so just at the stage now where we just have to put up with it. Had another BSOD after my earlier post. Considered buying another 3600 but just going to leave it now and hope it doesn't get worse.

Have you tried a a clean build with no ubisoft games installed to see what happens?

z0oT · 25-02-2020 11:27am

.G. wrote: »

Ubisoft had no answers up to now still investigating. Mainly seems to happen when he's playing their games. Neither did msi so just at the stage now where we just have to put up with it. Had another BSOD after my earlier post. Considered buying another 3600 but just going to leave it now and hope it doesn't get worse.

I don't think anyone else has suggested this to you in this thread - MemTest should be your 1st port of call if you think you've a memory issue.
https://www.memtest86.com/

It runs from a bootable flash drive which has the advantage of taking the OS out of the equation & isolating the memory. Set it up and run it overnight, if you don't get any errors you should be good.

At least that would allow you to rule out the memory once and for all.

.G. · 25-02-2020 12:27pm

savemejebus wrote: »

More likely would be his bad ram corrupting your windows install. I think that i read about that happening before but that was from running unstable overclocked ram.

Have you tried a a clean build with no ubisoft games installed to see what happens?

I didn't have any BSOD on my own machine with his ram in it, only getting them since I took my my ram back out of his machine and put back into mine! Not too concerned about that anyway, his is a much bigger issue. Havent tried to clean build without ubisoft but it's their games is all he plays. Might try it. When we built it yesterday I only installed windows, essential drivers, bitdefender and uplay for Seige and the division two Nothing else. Still crashed.

.G. · 25-02-2020 12:29pm

z0oT wrote: »

I don't think anyone else has suggested this to you in this thread - MemTest should be your 1st port of call if you think you've a memory issue.
https://www.memtest86.com/

It runs from a bootable flash drive which has the advantage of taking the OS out of the equation & isolating the memory. Set it up and run it overnight, if you don't get any errors you should be good.

At least that would allow you to rule out the memory once and for all.

Yeah was suggested and I've tried it. His ram passed with no issues reported over 13 hours of testing overnight. I haven't stressed the cpu yet, might do that with the new board in place now.

savemejebus · 25-02-2020 12:32pm

.G. wrote: »

I didn't have any BSOD on my own machine with his ram in it, inky getting them since I took my inw ram back out of his machine and put back into mine! Not too concerned about that anyway, his is a much bigger issue. Havent tried to clean build without ubisoft but it's their games is all he plays. Might try it.

You wouldn't necessarily have them immediately, data corruption can be weird. Edited as answered

.G. · 25-02-2020 12:33pm

savemejebus wrote: »

You wouldn't necessarily have them immediately, data corruption can be weird. running an overnight of memtest from http://www.memtest.org/ (the free unlimited version) as Zoot suggested (I know you ran the limited paid version before) is still a good plan too.

Yeah see above. I got access to the paid version and ran it over night with no issues found. Havent tried my own ram though, must check that.

Cuddlesworth · 25-02-2020 12:47pm

.G. wrote: »

Yeah see above. I got access to the paid version and ran it over night with no issues found. Havent tried my own ram though, must check that.

Boot a live linux OS and use the stresstestapp google created memory test.

H3llR4iser · 25-02-2020 2:37pm

.G. wrote: »

New motherboard installed about 3 hours now and one game crash and a BSOD already :rolleyes::rolleyes::mad:

Jeez .G. , this is the weirdest issue I've seen in a while.

If I tracked the issue and your troubleshooting till now correctly, you tried:

- Swapping out the ram sticks
- A different motherboard
- Swapping GPUs
- Stress testing the RAM

I've never seen a modern CPU cause so specific an issue, usually if a CPU is faulty it shows pretty obiviously, but it can't be excluded.

At this point, a CPU stress test is the next step - I'd go OCCT, Prime95 or Aida...or even set Cinebench to run for a long time (hours).

Lastly, two more "suspects" - have you tried a different storage device for the OS and game(s)? The fact the BSODs happen in game might be related to data I/O to the disk (e.g. loading/caching textures and models, pagefile).

I had a faulty SSD on a laptop in the past doing something similar - it'd work no problem for weeks, running everything fine, then suddenly BAM! - blue screen of death, usually when doing something like moving a large file or saving a big PSD. On reboot, all worked normally. Formatted the SSD, reinstalled everything from scratch, rinse repeat - a week or two ok, then BSODs.

Finally, how about the PSU?

.G. · 25-02-2020 5:18pm

Firstly thanks for all the replies. I'm not the only one who would like to find the source of this clearly

.

So as mentioned above,

I've tried:-

different ram and have stressed the original ram over night with no issues found
Now have a different mobo installed
Have also taken out both his storage devices and tried two completely different ones to no avail.

I haven't swapped out the gpu or the psu. The Psu is a 550w gold unit from seasonic. Going by online calculators that should be fine for a 3600 and a 1080. It's about a year and a half old I reckon.

I also haven't tried a different gpu as the only other one I have is my 2080ti and I don't think the psu will cope, maybe it will. Mainly though I don't want to take the chance that any dodgy hardware on his end possibly damaging my GPU.

I will stress the cpu as a next port of call using aida64, how long do we think I should run it for?

Will also try different storage setups as he always has uplay on his ssd along with windows but the actual games on the hdd, I'll put them all on the hdd and see what occurs. Will also try no uplay or ubisoft games installed at all and get him to play other stuff for a few days.

I've not got any idea about Linux though so that option is above my head.

.G. · 25-02-2020 5:41pm

.G. wrote: »

Firstly thanks for all the replies. I'm not the only one who would like to find the source of this clearly .

So as mentioned above,

I've tried:-

different ram and have stressed the original ram over night with no issues found
Now have a different mobo installed
Have also taken out both his storage devices and tried two completely different ones to no avail.

I haven't swapped out the gpu or the psu. The Psu is a 550w gold unit from seasonic. Going by online calculators that should be fine for a 3600 and a 1080. It's about a year and a half old I reckon.

I also haven't tried a different gpu as the only other one I have is my 2080ti and I don't think the psu will cope, maybe it will. Mainly though I don't want to take the chance that any dodgy hardware on his end possibly damaging my GPU.

I will stress the cpu as a next port of call using aida64, how long do we think I should run it for?

Will also try different storage setups as he always has uplay on his ssd along with windows but the actual games on the hdd, I'll put them all on the hdd and see what occurs. Will also try no uplay or ubisoft games installed at all and get him to play other stuff for a few days.

I've not got any idea about Linux though so that option is above my head.

Actually he has uplay and the ubi games on the HDD. Windows is on the SSD. Pretty sure we had them all on the SSD during a previous install of Windows and it still crashed but I'll try.

As for aida64. It has options to stress The GPU, Ram and storage devices as well as the CPU, FPU and cache. I'm ticking all those boxes yeah?

H3llR4iser · 25-02-2020 7:11pm

.G. wrote: »

Firstly thanks for all the replies. I'm not the only one who would like to find the source of this clearly .

So as mentioned above,

I've tried:-

different ram and have stressed the original ram over night with no issues found
Now have a different mobo installed
Have also taken out both his storage devices and tried two completely different ones to no avail.

I haven't swapped out the gpu or the psu. The Psu is a 550w gold unit from seasonic. Going by online calculators that should be fine for a 3600 and a 1080. It's about a year and a half old I reckon.

I also haven't tried a different gpu as the only other one I have is my 2080ti and I don't think the psu will cope, maybe it will. Mainly though I don't want to take the chance that any dodgy hardware on his end possibly damaging my GPU.

I will stress the cpu as a next port of call using aida64, how long do we think I should run it for?

Will also try different storage setups as he always has uplay on his ssd along with windows but the actual games on the hdd, I'll put them all on the hdd and see what occurs. Will also try no uplay or ubisoft games installed at all and get him to play other stuff for a few days.

I've not got any idea about Linux though so that option is above my head.

The PSU should be able to take a 2080TI at stock - my system with a 3900X, a 2070 Super AND a 1070 draws 505W at peak from the wall while EVERYTHING is at maximum usage (rendering using CUDA + CPU). While gaming (only on the 2070, of course) it goes nowhere near that value and hovers around 275W.

HOWEVER... I understand the reluctance to risk a card that costs as much as a decent second hand citycar

Maybe test his card on your system? Not the best of scenarios, but better than nothing...

.G. · 25-02-2020 7:25pm

His GPU came from my system when I upgraded, the perks he gets from me being a tech head

It was fine when I had it granted that doesn't mean it still is. Stress testing is ongoing right now, hour and a half in with no problems and max temp 78c on the CPU and 75c on the GPU.

z0oT · 25-02-2020 11:10pm

.G. wrote: »

His GPU came from my system when I upgraded, the perks he gets from me being a tech head

It was fine when I had it granted that doesn't mean it still is. Stress testing is ongoing right now, hour and a half in with no problems and max temp 78c on the CPU and 75c on the GPU.

If it makes you feel any better - here's some anguish I've had in the last few hours. :pac:

I picked up an R5 2600 for my server as an upgrade from the R3 2200G I currently have in it. A bit over €100 - a steal. It should have been just a simple swap, pull out the 2200G, and stick in the 2600 & Windows Server would just boot up as before, but nothing is ever easy. :pac:

The system hung a little upon startup, but that's to be expected with new hardware. Windows booted and here I am thinking great! But then after a few seconds I get a BSOD (Critical Process Died). Reboot and after that I can't get into the BIOS anymore.

So I try removing the battery to clear the CMOS (only way to do it on the board I have), I get into the BIOS, then it freezes upon disabling CSM.

Reboot - no video, nothing. Clear the CMOS again pull out everything from the board except 1 DIMM & the CPU fan, get into the BIOS once more upon a reboot, flash to an older version using the IPMI, no video upon boot once again. Re-Flash to a newer version, reboot - this time I get video but it hangs at initializing once again.

So after going around in circles for a while, in an act of desperation, I put the 2200G back in once more and everything boots up just as it was before. After plugging everything back into the board everything is back to normal - all my network drives, services - everything.

Then I think was it bad contacting with the CPU socket - so I stick back in the 2600 once more, hangups & freezes again. Replace it with the 2200G once more and everything is fine again.

So now I'm wondering did I get a bad CPU. It'll be the 1st time in my 14 or so years at this madness that it'll have happened to me. I still have my old B450 motherboard, so I might give the new R5 2600 a shot in that to test it out there.

Either way I might just return it anyway - Seriously disappointing!

.G. · 26-02-2020 7:55am

z0oT wrote: »

If it makes you feel any better - here's some anguish I've had in the last few hours. :pac:

I picked up an R5 2600 for my server as an upgrade from the R3 2200G I currently have in it. A bit over €100 - a steal. It should have been just a simple swap, pull out the 2200G, and stick in the 2600 & Windows Server would just boot up as before, but nothing is ever easy. :pac:

The system hung a little upon startup, but that's to be expected with new hardware. Windows booted and here I am thinking great! But then after a few seconds I get a BSOD (Critical Process Died). Reboot and after that I can't get into the BIOS anymore.

So I try removing the battery to clear the CMOS (only way to do it on the board I have), I get into the BIOS, then it freezes upon disabling CSM.

Reboot - no video, nothing. Clear the CMOS again pull out everything from the board except 1 DIMM & the CPU fan, get into the BIOS once more upon a reboot, flash to an older version using the IPMI, no video upon boot once again. Re-Flash to a newer version, reboot - this time I get video but it hangs at initializing once again.

So after going around in circles for a while, in an act of desperation, I put the 2200G back in once more and everything boots up just as it was before. After plugging everything back into the board everything is back to normal - all my network drives, services - everything.

Then I think was it bad contacting with the CPU socket - so I stick back in the 2600 once more, hangups & freezes again. Replace it with the 2200G once more and everything is fine again.

So now I'm wondering did I get a bad CPU. It'll be the 1st time in my 14 or so years at this madness that it'll have happened to me. I still have my old B450 motherboard, so I might give the new R5 2600 a shot in that to test it out there.

Either way I might just return it anyway - Seriously disappointing!

This stuff can be frustrating as hell! I feel your pain.

As an update on my end we ran aida64 for two hours with no issues. Within 5 minutes of ending the test he was un installing one of the ubisoft games, he hit cancel while it was doing it cos he thought it might delete his game saves and he got instant BSOD.

Eventually finished the process properly, have uplay and The Division 2 now installed on his ssd rather than the hdd. No BSOD yet but one game crash.

Kimbot · 26-02-2020 2:26pm

.G. wrote: »

This stuff can be frustrating as hell! I feel your pain.

As an update on my end we ran aida64 for two hours with no issues. Within 5 minutes of ending the test he was un installing one of the ubisoft games, he hit cancel while it was doing it cos he thought it might delete his game saves and he got instant BSOD.

Eventually finished the process properly, have uplay and The Division 2 now installed on his ssd rather than the hdd. No BSOD yet but one game crash.

Are the hdds brand new?

H3llR4iser · 26-02-2020 4:38pm

Kimbot wrote: »

Are the hdds brand new?

curious as well about that - this sounds more and more like a data I/O issue...or something so weirdly wrong with the CPU that only shows up during specific operations.

.G. · 26-02-2020 5:35pm

not brand new no but also came from my machine and were fine in it.

Kimbot · 26-02-2020 6:54pm

.G. wrote: »

not brand new no but also came from my machine and were fine in it.

I'm leaning towards one of the hdds been corrupt in a weird way, you had the OS on 1 and data on the other. I would go with a complete reformat of 2 drives.
Then as a test I would use 1 drive and put the OS, drivers, uplay and 1 game that mostly causes the bsod on it. When all is up and config turn it all off and do the exact same thing with second drive so they are both identical. Then go and test 1 drive at a time and see if either cause the bsods but you will probably need to put a good bit of testing time into each setup over 2/3 days and document the results and see how it looks.

z0oT · 26-02-2020 9:56pm

z0oT wrote: »

If it makes you feel any better - here's some anguish I've had in the last few hours. :pac:

I picked up an R5 2600 for my server as an upgrade from the R3 2200G I currently have in it. A bit over €100 - a steal. It should have been just a simple swap, pull out the 2200G, and stick in the 2600 & Windows Server would just boot up as before, but nothing is ever easy. :pac:

The system hung a little upon startup, but that's to be expected with new hardware. Windows booted and here I am thinking great! But then after a few seconds I get a BSOD (Critical Process Died). Reboot and after that I can't get into the BIOS anymore.

So I try removing the battery to clear the CMOS (only way to do it on the board I have), I get into the BIOS, then it freezes upon disabling CSM.

Reboot - no video, nothing. Clear the CMOS again pull out everything from the board except 1 DIMM & the CPU fan, get into the BIOS once more upon a reboot, flash to an older version using the IPMI, no video upon boot once again. Re-Flash to a newer version, reboot - this time I get video but it hangs at initializing once again.

So after going around in circles for a while, in an act of desperation, I put the 2200G back in once more and everything boots up just as it was before. After plugging everything back into the board everything is back to normal - all my network drives, services - everything.

Then I think was it bad contacting with the CPU socket - so I stick back in the 2600 once more, hangups & freezes again. Replace it with the 2200G once more and everything is fine again.

So now I'm wondering did I get a bad CPU. It'll be the 1st time in my 14 or so years at this madness that it'll have happened to me. I still have my old B450 motherboard, so I might give the new R5 2600 a shot in that to test it out there.

Either way I might just return it anyway - Seriously disappointing!

Figured out my issue.

With the R5 2600 installed, I swapped the Crucial DIMMs for a Corsair Vengeance LPX DIMM from my gaming desktop - and everything worked just fine. BIOS loads just fine, Windows boots & all my services & network drives are fine once again.

So it would seem the issue is the Crucial memory & R5 2600 pair. That memory has been a bit funny in the past. My old i7-6700k for instance would freeze in the BIOS if XMP was enabled with it.

A successful few hours but I guess I need new memory now.

.G. · 26-02-2020 9:58pm

Kimbot wrote: »

I'm leaning towards one of the hdds been corrupt in a weird way, you had the OS on 1 and data on the other. I would go with a complete reformat of 2 drives.
Then as a test I would use 1 drive and put the OS, drivers, uplay and 1 game that mostly causes the bsod on it. When all is up and config turn it all off and do the exact same thing with second drive so they are both identical. Then go and test 1 drive at a time and see if either cause the bsods but you will probably need to put a good bit of testing time into each setup over 2/3 days and document the results and see how it looks.

I've got two spare HDD's and one spare SSD I've tried all of them at different points in this process and it happens with all of them, they can't all be corrupt imo. The process has been one SSD for windows matched with one of the HDDs. So I've tried his SSD with his HDD, I've tried my spare SSD with both my HDDs one at a time and then his one. Have tried different sata cables too.

Both the SSD's are only 128GB so can't put anymore than windows and the Division 2 on it. He still hasn't had a BSOD since we put uplay and the one uplay game on the SSD but the game itself still crashes. He's playing steam games today with no issues so far and they are on the HDD.

It really seems to be some weird quirk between his particular system and uplay. Granted I bet I've jinxed it now and the steam games will crash:D

.G. · 26-02-2020 10:34pm

Just came across a decent deal for a 1TB M2 drive so I've bought that. Will get rid of all the other drives and just use that when it comes and do lots and lots of praying

jonerkinsella · 07-03-2020 12:01pm

z0oT wrote: »

Figured out my issue.

With the R5 2600 installed, I swapped the Crucial DIMMs for a Corsair Vengeance LPX DIMM from my gaming desktop - and everything worked just fine. BIOS loads just fine, Windows boots & all my services & network drives are fine once again.

So it would seem the issue is the Crucial memory & R5 2600 pair. That memory has been a bit funny in the past. My old i7-6700k for instance would freeze in the BIOS if XMP was enabled with it.

A successful few hours but I guess I need new memory now.

Send the modules back to Crucial under warranty. There good to deal with.

AMD Zen Discussion Thread

Comments