My ASUS Z690 mobo hates Gen 4 NVMe drives:/

[drumn_bass]

Hi guys,

 

Just wondering if anyone have heard anything about this.

 

I have an Asus TUF Z690 D4 Plus mobo, with 12700K, 64GB DDR4 Ram and RTX 3070. And currently 2 1TB Western Digital SN750 Gen3 NVMe SSD drives.

 

I used to have a Samsung 980 Pro Gen 4 drive as my system drive. Used it for about 9 months. In October, I was considering a possibility of upgrading to a 13 gen Intel CPU, and got a BIOS update that supported it, from September 2022. Immediately after the update, I started getting random BSOD, usually when running games, virtual machines, or just having a lot of things going on. All the codes indicated issues were with system files, for example "crytical_process_died", or "system_service_exception". When the BSOD occurred, it just flashed for a split second and the system restarted. It never got a chance to record any dump files. I struggled for weeks trying to figure out what was causing it. At first blaming my WD SN750 drives, then realizing they actually work fine on their own, so I eventually removed the Samsung 980 Pro Gen4 drive, and used the 2 WD SN750 drives in my system for almost 2 months with no issues. At the same time, I moved the 980 Pro into my kid's system, as a system drive, and it's been working with no issues there. There was a new BIOS update for my mobo early December, I got it, and wanted to try a Gen4 drive again. I got a WD SN850x Gen4, and at first it appeared to work fine at first. Used it for about a week with the 2 WD SN750s, so thinking the issue was resolved with a BIOS update, I ordered and received a new Samsung 990 Pro Gen4, and proceeded to set it up as my system drive. And that same day, I started getting BSOD issues again. And to top it off, when I went back to the WD SN850x, I started seeing these BSOD crashed again, although not as often as I did with Samsung. I spent all weekend troubleshooting, reinstalling, and testing, and I will not be going over everything here, but let's just say the conclusion I arrived at, is that my Z690 mobo just doesn't like any PCIe Gen4 drives anymore.

 

I now have 2 WD SN750 Gen3 drives by themselves, and it runs fine again (and honestly, there is not a lot of difference between Gen3 and Gen4 drives for me, but the Samsung 990 Pro did feel a bit snappier, every time until it crashed with an epic BSOD). I figured out how to reliably reproduce the issue. Basically, I created a folder with a couple dozen different files, totaling around 15GB. When I start my system, with a Gen4 drive (Samsung, or WD) as a system or a secondary drive, and try to copy this folder to or from ANY drive, about mid-way through the transfer, the transfer speed drops to 0, and while the drive still appears in explorer, it becomes completely unresponsive. CrystalDiskInfo doesn't recognize it anymore, and the Task Manager still shows it, but it's read and write speeds stay at 0. If it is a secondary drive, it just remains in this "dead" state until I restart. If it's a system drive, it instantly throws out a BSOD and the system restarts. To clarify, this can happen to both Gen3 and Gen4 drives, but only if there is a Gen4 drive present in the system.

 

So, I can have these configurations with these results:

WD SN750 Gen3 + WD750 Gen3 - Runs fine

WD SN850x Gen4 + WD SN750 Gen3 + WD750 Gen3 - Crashes

Samsung 980 Pro Gen4 + WD SN750 Gen3 (1 or 2 of them) - Crashes

Samsung 990 Pro Gen4 + WD SN750 Gen3 (1 or 2 of them) - Crashes

Samsung 990 Pro Gen4 + WD SN850x Gen4 - Crashes

And finally, I couldn't confirm if a Gen4 by itself would cause crashes as I didn't have a way to use my test folder method, and didn't have time to set up and use the system until it crashed as otherwise it can be completely random.

 

The only way the system runs reliably is if there are no Gen4 drives installed. It's worth mentioning that all those Gen4 drives run absolutely fine in my home server system on B460 chipset.

 

I tried everything I could think of. Chipset drivers, power settings, BIOS settings, running Gen4 drives on Gen3 M.2 settings, and a bunch of other things that I can't even recall right now as at this point my head is spinning. I was literally on this every waking hour of the day since Friday. 

 

I will probably just stick with Gen3 drives, but I wish I knew what the issue is. I saw similar issues reported on Reddit and Asus forums (but for some reason Asus forums appear to be offline right now). Also, some mentions of Gigabyte boards on Z690 chipset having similar issues. Usually with Samsung drives. Another question I have is if it sound like there might be grounds for a RMA on the mobo, but I'd hate to try to go through this process only for them to tell me they don't see any issues with it. It's still within a year since the purchase date. It's irritating as my system worked great with a Gen4 980 Pro for 9 months before the BIOS update "broke" it, and as far as I can tell, there is no acknowledgement of any issues like these from Asus.

[CommanderAlex]

Can you revert to the previous BIOS version?? Sometimes, newer BIOSes can introduce instabilities in certain configurations, mainly just RAM OC where it was stable before a BIOS update, reverting back resolves the instability. 

[drumn_bass]

1 hour ago, CommanderAlex said:

Can you revert to the previous BIOS version?? Sometimes, newer BIOSes can introduce instabilities in certain configurations, mainly just RAM OC where it was stable before a BIOS update, reverting back resolves the instability. 

@CommanderAlex

 

Ah. Excellent suggestion. However, yes, I did try that, and it was a no go. That September's BIOS did not allow a roll back to an earlier version, and I don't have a BIOS flash back feature on my motherboard. 

[CommanderAlex]

7 hours ago, drumn_bass said:

@CommanderAlex

 

Ah. Excellent suggestion. However, yes, I did try that, and it was a no go. That September's BIOS did not allow a roll back to an earlier version, and I don't have a BIOS flash back feature on my motherboard. 

Hmm, have you tried contacting Asus about this issue? See what they say or if they don't know to elevate the case to higher technical support. 

[drumn_bass]

@CommanderAlex

 

I'll think about it. Do you by chance have any experience dealing with them? In my experience, it's more productive to talk to other users on the forums or reddit, than trying to work with a manufacturer, but I might be wrong. I spent so much time on this, I'm not sure if I want to continue, or just stick with Gen3 drives and be happy. 

 

I have a theory, but I feel like I might be showing my ignorance if I go too deep into it. Can it be a PCIe lanes issue on the chipset? The specs state that Z690 has 28 PCIe 4 lanes, and again, I might be showing my lack of understanding here, but the GPU is on the 16 lane slot, so 12 lanes technically remain open. M.2 slots use 4 lanes each, so in theory, I should have no issues with at least 3 M.2 drives? It's probably not even related to the number of lanes, I'm just throwing it out there. I tried using the Gen4 drive in the M.2 / 1 slot, which is going directly to the CPU, and it wasn't recognized at all. Just tell me to shut up about it if I'm out of my element here.

 

There is a post, on Asus forums, that looks like it might be the exact issue, or very close, in any case, might have some clues in it, but I can't get to their forums for several days now, not sure if it's me, or they're truly down for maintenance for that long. I work as a community Manager on some company's forums myself, and I can't imagine being offline for more than maybe 4 hours, it'll be an emergency.

 

https://rog.asus.com/forum/showthread.php?126865-Hero-Z690-having-issues-with-getting-windows-errors-and-nvme-drive-not-working/page9

 

I hope it's ok to post a link here, maybe someone can try opening it to let me know if it works?

 

Searching around, I see similar (but not exactly the same) issues reported with Gigabyte and MSI boards, with Z690, so it's probably the chipset thing, and some manufacturers may just be better than others at working around it. A part of me thinks that maybe I should just get a newer Z790 board, but then if I go buy new hardware, I feel like I might as well part my system out and build a whole new one, and to be honest, I don't really want to make this kind of investment, again, since I just build a new computer less than a year ago:(

[CommanderAlex]

19 minutes ago, drumn_bass said:

I'll think about it. Do you by chance have any experience dealing with them? In my experience, it's more productive to talk to other users on the forums or reddit, than trying to work with a manufacturer, but I might be wrong. I spent so much time on this, I'm not sure if I want to continue, or just stick with Gen3 drives and be happy. 

I've had to contact ASUS about my TUF gaming 6900XT black screening. Talked with them once and did something they advised to do...didn't work and issue still occurred on another PC and contacted them again to initiate RMA. Was found defective and they exchanged a refurb unit back to me, works flawlessly since I've had it. 

 

I've read forum posts on here pertaining to RMA issues where a slight scratch on the rear I/O backplate ASUS denied them RMA support as it's "customer induced damage" causing the card to black screen. Different cards and other components they warranted to deny RMA. Running off on a tangent here. 

 

19 minutes ago, drumn_bass said:

I have a theory, but I feel like I might be showing my ignorance if I go too deep into it. Can it be a PCIe lanes issue on the chipset? The specs state that Z690 has 28 PCIe 4 lanes, and again, I might be showing my lack of understanding here, but the GPU is on the 16 lane slot, so 12 lanes technically remain open. M.2 slots use 4 lanes each, so in theory, I should have no issues with at least 3 M.2 drives? It's probably not even related to the number of lanes, I'm just throwing it out there. I tried using the Gen4 drive in the M.2 / 1 slot, which is going directly to the CPU, and it wasn't recognized at all. Just tell me to shut up about it if I'm out of my element here.

 

There is a post, on Asus forums, that looks like it might be the exact issue, or very close, in any case, might have some clues in it, but I can't get to their forums for several days now, not sure if it's me, or they're truly down for maintenance for that long. I work as a community Manager on some company's forums myself, and I can't imagine being offline for more than maybe 4 hours, it'll be an emergency.

Ok, so your CPU itself, i7-12700K, has 24 PCIe lanes; 20 lanes for PCIe slots/M.2 NVMe & 4 lanes directly to the chipset. 16 PCIe Gen 5 lanes & 4 PCIe Gen 4 lanes. The Z690 chipset has 28 lanes for I/O, Ethernet, USB, etc. 

Here's a block diagram of what I'm saying. 

 

The chipset is responsible for dividing up it's own 28 lanes that has to direct data onto the 4 lanes connecting it to the CPU. 

 

According to the motherboard's manual, M.2_1 is directly connected to the CPU lanes, therefore the CPU should (don't quote me here as some of those lanes are Gen 5 while your graphics card is Gen 4. This is one thing I have yet to grasp what is going on here what number of lanes are being used) be allocating the appropriate # of lanes here to both GPU and M.2 installed in this slot. 

 

M.2_2, M.2_3, and M.2_4 are all running thru the chipset lanes. 

[drumn_bass]

1 hour ago, CommanderAlex said:

I've had to contact ASUS about my TUF gaming 6900XT black screening. Talked with them once and did something they advised to do...didn't work and issue still occurred on another PC and contacted them again to initiate RMA. Was found defective and they exchanged a refurb unit back to me, works flawlessly since I've had it. 

 

I've read forum posts on here pertaining to RMA issues where a slight scratch on the rear I/O backplate ASUS denied them RMA support as it's "customer induced damage" causing the card to black screen. Different cards and other components they warranted to deny RMA. Running off on a tangent here. 

 

Ok, so your CPU itself, i7-12700K, has 24 PCIe lanes; 20 lanes for PCIe slots/M.2 NVMe & 4 lanes directly to the chipset. 16 PCIe Gen 5 lanes & 4 PCIe Gen 4 lanes. The Z690 chipset has 28 lanes for I/O, Ethernet, USB, etc. 

Here's a block diagram of what I'm saying. 

 

The chipset is responsible for dividing up it's own 28 lanes that has to direct data onto the 4 lanes connecting it to the CPU. 

 

According to the motherboard's manual, M.2_1 is directly connected to the CPU lanes, therefore the CPU should (don't quote me here as some of those lanes are Gen 5 while your graphics card is Gen 4. This is one thing I have yet to grasp what is going on here what number of lanes are being used) be allocating the appropriate # of lanes here to both GPU and M.2 installed in this slot. 

 

M.2_2, M.2_3, and M.2_4 are all running thru the chipset lanes. 

@CommanderAlex

 

Thanks for the explanation. I know people usually recommend installing your system drive into the M.2_1 slot, but I think it's location is physically not very great. It's right in between the CPU, and a very toasty GPU, so simply having an SSD in that 1st slot, makes it run about 20 degrees hotter than any other M.2 slot on the motherboard. While people believe having it connecting directly to the CPU should in theory improve responsiveness of the system, under heavier loads, the SSD can actually overheat and throttle, plus seeing it reach 80+ degrees makes me a bit uncomfortable, although I know these newer Gen4 SSDs can actually run even hotter than that and still stay within spec.

 

I might try the Samsung 990 Pro in the M.2_1 slot again. The last time I did, it wasn't recognized, so I need to look into that too. When I was changing PCIe generation settings in BIOS, the M.2_1 wasn't even listed:/ I might have to reset CMOS, and if it's still acting up, then it might indicate something is wrong with the motherboard for sure.

[drumn_bass]

A midway update.

 

So the M.2_1 slot recognized the Samsung 990 Pro Gen4 drive, and while it's still empty with unallocated space only, so far copying files (aka test folder scenario) did produce any errors. Currently, there are also 2 Gen3 WD 750s are in the M.2 slots 3 and 4. The system is on the slot 4. I will go ahead and clone the system disk to the Samsung (normally I do a clean install, but it's for testing purposes for now), and do some tests. I'll let you know what happens.

[CommanderAlex]

26 minutes ago, drumn_bass said:

@CommanderAlex

 

Thanks for the explanation. I know people usually recommend installing your system drive into the M.2_1 slot, but I think it's location is physically not very great. It's right in between the CPU, and a very toasty GPU, so simply having an SSD in that 1st slot, makes it run about 20 degrees hotter than any other M.2 slot on the motherboard. While people believe having it connecting directly to the CPU should in theory improve responsiveness of the system, under heavier loads, the SSD can actually overheat and throttle, plus seeing it reach 80+ degrees makes me a bit uncomfortable, although I know these newer Gen4 SSDs can actually run even hotter than that and still stay within spec.

 

I might try the Samsung 990 Pro in the M.2_1 slot again. The last time I did, it wasn't recognized, so I need to look into that too. When I was changing PCIe generation settings in BIOS, the M.2_1 wasn't even listed:/ I might have to reset CMOS, and if it's still acting up, then it might indicate something is wrong with the motherboard for sure.

Yeah I was concerned about putting an SSD in that slot too on my motherboard and a previous build I built back in 2019 on an MSI X299 Gaming Pro Carbon AC. I've monitored temps and they don't get too hot. I think mainly it's OK as long as you aren't doing heavy read/write sequences on the drive and have a ton of heat, as that alone drives SSDs temps up. 

 

I'm not sure how much of an effect it can do, but when you flashed the new BIOS, do you recall having XMP enabled at the time? I think the times I've done it on my board (different boards so I am unsure if it's the same) that tells me to not flash with XMP enabled, basically do a clear CMOS and then flash. 

17 minutes ago, drumn_bass said:

A midway update.

 

So the M.2_1 slot recognized the Samsung 990 Pro Gen4 drive, and while it's still empty with unallocated space only, so far copying files (aka test folder scenario) did produce any errors. Currently, there are also 2 Gen3 WD 750s are in the M.2 slots 3 and 4. The system is on the slot 4. I will go ahead and clone the system disk to the Samsung (normally I do a clean install, but it's for testing purposes for now), and do some tests. I'll let you know what happens.

Does that drive show up in CrystalDiskInfo? It should be reporting back info on the drive and health (Note: in terms of health, S.M.A.R.T. data is not always fool proof in indicating a drive is about to fail). 

[drumn_bass]

9 minutes ago, CommanderAlex said:

Yeah I was concerned about putting an SSD in that slot too on my motherboard and a previous build I built back in 2019 on an MSI X299 Gaming Pro Carbon AC. I've monitored temps and they don't get too hot. I think mainly it's OK as long as you aren't doing heavy read/write sequences on the drive and have a ton of heat, as that alone drives SSDs temps up. 

 

I'm not sure how much of an effect it can do, but when you flashed the new BIOS, do you recall having XMP enabled at the time? I think the times I've done it on my board (different boards so I am unsure if it's the same) that tells me to not flash with XMP enabled, basically do a clear CMOS and then flash. 

Does that drive show up in CrystalDiskInfo? It should be reporting back info on the drive and health (Note: in terms of health, S.M.A.R.T. data is not always fool proof in indicating a drive is about to fail). 

@CommanderAlex

 

The XMP is always enabled, and idk, it worked fine for many months before it started acting up after a BIOS update. So I just did what I said, cloned the system to the Gen4 drive in the M.2_1 slot, and as soon as it boots, I get the same BSOD again, "crytical_process_died". It lets me sign in to Windows, then runs for about 10 seconds, freezes (I assume that's when the drive stops responding) and finally throws the BSOD because it can't read any system files anymore. Removed the Gen4 drive, booted into the Gen3, and works fine again. I guess I could try a fresh copy of Windows on the Gen4 drive, but at this point, I honestly don't think it'll help (already did that so many times over the last few months). 

 

The good thing is that I was able to confirm the M.2_1 slot if still working:) (even though it doesn't make any difference as far as this specific issue I'm having)

[CommanderAlex]

13 minutes ago, drumn_bass said:

@CommanderAlex

 

The XMP is always enabled, and idk, it worked fine for many months before it started acting up after a BIOS update. So I just did what I said, cloned the system to the Gen4 drive in the M.2_1 slot, and as soon as it boots, I get the same BSOD again, "crytical_process_died". It lets me sign in to Windows, then runs for about 10 seconds, freezes (I assume that's when the drive stops responding) and finally throws the BSOD because it can't read any system files anymore. Removed the Gen4 drive, booted into the Gen3, and works fine again. I guess I could try a fresh copy of Windows on the Gen4 drive, but at this point, I honestly don't think it'll help (already did that so many times over the last few months). 

 

The good thing is that I was able to confirm the M.2_1 slot if still working:) (even though it doesn't make any difference as far as this specific issue I'm having)

So when you copied the OS drive on this Gen 4 drive in M.2_1, what did you do to system drive in slot 4?

[drumn_bass]

@CommanderAlex

 

Honestly, nothing, left it in there. But booted from the Samsung in slot 1. I'm sure it's not a good idea, but I've done it in my server/test bench PC many times when I wanted to clone disks, or back up to an image. 

[CommanderAlex]

11 minutes ago, drumn_bass said:

@CommanderAlex

 

Honestly, nothing, left it in there. But booted from the Samsung in slot 1. I'm sure it's not a good idea, but I've done it in my server/test bench PC many times when I wanted to clone disks, or back up to an image. 

Hm ok. I was gonna suggest possibly pulling it out of the system and see if the behavior changes. 

[drumn_bass]

@CommanderAlex

 

I'm gonna try the Gen4 990 Pto drive in the M.2_1 slot, with fresh install, and just one secondary Gen3 WD SN750 drive in M.2_4 slot. Just two drives. Clean slate.

 

I also remembered there is always some Asus driver that comes up in optional updates in Windows. I'm not sire what it even is, figured something for the motherboard, but I will try ignoring it too, just in case. In fact, I will probably skip all the optional updates this time. 

 

If it starts crashing again, I'm afraid it'll be it for me, I'll just send the 990 Pro back. I can't be spending the entire week messing with it:) I need to try to get some work done. 

[drumn_bass]

@CommanderAlex

 

Okay. Alex, thank you so much for all your help and great advice!

 

Like I said in my previous post, I had one final test to perform. Clean slate. Fresh Windows install. Blank secondary drive. No suspicious drivers. Just as vanilla as I can have it. Will it work reliably?

 

It did not. I got the BSOD "critical_process_died" within the first 3 minutes of use. It seems no matter what I try, my system just hates them Gen4 SSDs. Sure, there are probably still a lot more things I could have tested, but at this point, I honestly just need to get it to a working state and get back to life:)

 

So I will be sticking with my 2 Gen3 SN750 drives, and returning the Samsung 990 Pro and WD SN850x. Honestly, not that much of a difference anyway between Gen3 and Gen4 as far as I can tell. Heck, just a few years ago we were ecstatic about the speeds of SATA SSDs, and these Gen3 drives are already more than 10 times faster than those, so really... good enough!

 

Anyway. It's been fun.

 

Thanks again man.

[CommanderAlex]

41 minutes ago, drumn_bass said:

@CommanderAlex

 

I'm gonna try the Gen4 990 Pto drive in the M.2_1 slot, with fresh install, and just one secondary Gen3 WD SN750 drive in M.2_4 slot. Just two drives. Clean slate.

 

I also remembered there is always some Asus driver that comes up in optional updates in Windows. I'm not sire what it even is, figured something for the motherboard, but I will try ignoring it too, just in case. In fact, I will probably skip all the optional updates this time. 

 

If it starts crashing again, I'm afraid it'll be it for me, I'll just send the 990 Pro back. I can't be spending the entire week messing with it:) I need to try to get some work done. 

 

22 minutes ago, drumn_bass said:

@CommanderAlex

 

Okay. Alex, thank you so much for all your help and great advice!

 

Like I said in my previous post, I had one final test to perform. Clean slate. Fresh Windows install. Blank secondary drive. No suspicious drivers. Just as vanilla as I can have it. Will it work reliably?

 

It did not. I got the BSOD "critical_process_died" within the first 3 minutes of use. It seems no matter what I try, my system just hates them Gen4 SSDs. Sure, there are probably still a lot more things I could have tested, but at this point, I honestly just need to get it to a working state and get back to life:)

 

So I will be sticking with my 2 Gen3 SN750 drives, and returning the Samsung 990 Pro and WD SN850x. Honestly, not that much of a difference anyway between Gen3 and Gen4 as far as I can tell. Heck, just a few years ago we were ecstatic about the speeds of SATA SSDs, and these Gen3 drives are already more than 10 times faster than those, so really... good enough!

 

Anyway. It's been fun.

 

Thanks again man.

Man that's a real bummer. Sorry we couldn't get it to work in your system but it was worth a shot. 

[drumn_bass]

@CommanderAlex

 

I'll keep some updates going in here in case you're curious, because after a few days of no issues with only 2 Gen3 WD drives, today, I finally decided to go ahead and download a few games onto my game drive, and after a while, it stopped responding, just like it happened before. So I'm suspecting now it's not only affecting Gen4 drives, but it can be some kind of bandwidth or temperature issue, and Gen4 drives are just able to get it to that point where something stops responding quicker. Looks like with Gen3 you really have to push it, and I was writing about 150GB of game data to it for a while until it finally happened.

 

So for now, I just shuffled the drives around to different M.2 slots, in case it's a specific slot not working right with a specific drive thing. And I also have a third Gen3 drive that I wasn't using, so I will probably put it in too to see if having 3 of them make it reach its limit quicker. Just to test that theory.

 

In the meantime, I will follow your advice and go ahead and reach out to Asus Support. Just to see what they say. Looks like their Forums are back up, so I'll go harass some people there too.

[drumn_bass]

There are definitely some posts on Asus ROG forums that describe very similar issues, some guys are mentioning random reboots with new Gen4 drive (or a second Gen4 drive added). One said he RMAed his board and the new appears to be fine, so I will definitely be reaching out to Asus support as soon as possible. Another mentioned getting BSOD with a new drive until he disabled AI Overclocking (I don't use that). In either case, it seems related, but with no definitive patterns, just random errors with Gen4 drives, or 2 drives in the system. So far.

[drumn_bass]

@CommanderAlex

 

Hm. Interesting. So I tried chatting with Asus support, and after entering my serial number, got this message:

 

"Thank you for your being a valued ASUS customer. In the unlikely event that your computer requires warranty service, please contact Best Buy for support at 1-800-433-5778 or visit https://www.bestbuy.com/geeksquad. For general Self-Service or obtain general product information, or to register your product visit: https://www.asus.com/us/support."

 

Have you heard of this before? If I'm understanding it correctly, they're delegating warranty service to Best Buy? (I did buy the board from Best Buy). I'm not sure if this should make me feel better or worse about it, lol.

[CommanderAlex]

10 hours ago, drumn_bass said:

@CommanderAlex

 

Hm. Interesting. So I tried chatting with Asus support, and after entering my serial number, got this message:

 

"Thank you for your being a valued ASUS customer. In the unlikely event that your computer requires warranty service, please contact Best Buy for support at 1-800-433-5778 or visit https://www.bestbuy.com/geeksquad. For general Self-Service or obtain general product information, or to register your product visit: https://www.asus.com/us/support."

 

Have you heard of this before? If I'm understanding it correctly, they're delegating warranty service to Best Buy? (I did buy the board from Best Buy). I'm not sure if this should make me feel better or worse about it, lol.

I never heard that before. Did you purchase the motherboard from a Best Buy?? I live in the U.S. but I've read online here users from E.U. have to go through the retailer they purchased their parts from for RMA, it's not a direct consumer-manufacturer RMA like it is here in the U.S. so that's the only thing that comes to my mind. 

[drumn_bass]

@CommanderAlex

 

Yes, I bought it at Best Buy. I contacted them (via chat), and the agent confirmed that they are authorized to handle warranty issues, but, he told me I have to take my entire computer to the Geek Squad, to see if they can diagnose an issue, and he couldn't tell me if there would be any charges for their services. Honestly, I don't really want to do that, yet, however it might not be too bad since I know now how to fairly reliably replicate the issue, so I might be able to demonstrate it. I just don't trust leaving my computer with them:( I'm going to try something else first. 

 

I ordered another one of these motherboards, the same exact one, from BB. I'm hoping that it's either going to have an earlier BIOS version, or maybe just generally better hardware (in case there were some issues with early batches). I will do my tests, but no BIOS updates unless I run into some other stability issues. The theory here is if another one of these boards works, I will go ahead and try that warranty route. And if it starts giving me the same trouble, then I will just have to get rid of it and get something else. 

 

I did a test with 3 Gen3 drives, and just like I suspected, it gave me a BSOD almost instantly, right after I tried copying my test folder from one of the drives to the system drive. Removed the third drive, restarted, and the same folder copied just fine, and no crashes so far. So I have to revise my findings a bit, it's not a Gen4 NVMe drive that causes the issue, but looks like it just can't handle a combination of Gen4 + Gen3, or 3 or more Gen3 drives at the same time. There were in fact several similar reports on Asus forums with other Z690 boards, but none of the possible solutions people suggested worked for me. Looks like any brand of SSDs can cause it, but most people reported issues with Samsung drives, so for some reason, they definitely hate Samsung most:) 

 

If I have to get another motherboard, I know it's a cliché at this point, but it probably won't be Asus. To be fair, I have another Asus board in my server, a B460, and it is as reliable as they get, but something is not right with their newer ones, it seems. I always had good luck with MSI, they just take longer to boot, and I hated that, but the PC almost always sleeps anyway. 

[IkeaGnome]

Before you did the BIOS update, did you update Intel ME? 

Also, have you tried the newer Sata drivers for that board that came out in August?

https://www.asus.com/motherboards-components/motherboards/tuf-gaming/tuf-gaming-z690-plus-wifi-d4/helpdesk_download/?model2Name=TUF-GAMING-Z690-PLUS-WIFI-D4

BIOS and Firmware>scroll down to Intel ME and install the install utility.

Then Drivers and Tools>Pick your OS>Sata>Grab the newer of the two and install.

It's a new step in Intel BIOS updating that I'm wondering if a lot of people are missing. 

[CommanderAlex]

31 minutes ago, drumn_bass said:

@CommanderAlex

 

Yes, I bought it at Best Buy. I contacted them (via chat), and the agent confirmed that they are authorized to handle warranty issues, but, he told me I have to take my entire computer to the Geek Squad, to see if they can diagnose an issue, and he couldn't tell me if there would be any charges for their services. Honestly, I don't really want to do that, yet, however it might not be too bad since I know now how to fairly reliably replicate the issue, so I might be able to demonstrate it. I just don't trust leaving my computer with them:( I'm going to try something else first. 

Ok.  Yeah if you bring over the computer, you'll want backups as they won't be held responsible for loss/corrupt data as you'll be signing a waiver for that and to be logged out of any emails and accounts you don't want them to have access to. I've had BB Geek Squad extended warranty plans for my laptops and that's a typical paper you sign for them to proceed working on the computer. You'll definitely be paying a diagnostic fee for them to check out the computer.

31 minutes ago, drumn_bass said:

 

I ordered another one of these motherboards, the same exact one, from BB. I'm hoping that it's either going to have an earlier BIOS version, or maybe just generally better hardware (in case there were some issues with early batches). I will do my tests, but no BIOS updates unless I run into some other stability issues. The theory here is if another one of these boards works, I will go ahead and try that warranty route. And if it starts giving me the same trouble, then I will just have to get rid of it and get something else. 

 

I did a test with 3 Gen3 drives, and just like I suspected, it gave me a BSOD almost instantly, right after I tried copying my test folder from one of the drives to the system drive. Removed the third drive, restarted, and the same folder copied just fine, and no crashes so far. So I have to revise my findings a bit, it's not a Gen4 NVMe drive that causes the issue, but looks like it just can't handle a combination of Gen4 + Gen3, or 3 or more Gen3 drives at the same time. There were in fact several similar reports on Asus forums with other Z690 boards, but none of the possible solutions people suggested worked for me. Looks like any brand of SSDs can cause it, but most people reported issues with Samsung dri

ves, so for some reason, they definitely hate Samsung most:) 

You can check it out with another motherboard like it and see if issues occur.

31 minutes ago, drumn_bass said:

f I have to get another motherboard, I know it's a cliché at this point, but it probably won't be Asus. To be fair, I have another Asus board in my server, a B460, and it is as reliable as they get, but something is not right with their newer ones, it seems. I always had good luck with MSI, they just take longer to boot, and I hated that, but the PC almost always sleeps anyway.

I've always had an MSI board and this time I decided to give Asus a chance since this is one of 2 original X570 motherboards that don't have a chipset cooler...X570S boards came after I purchased this board and were just as expensive. This board is pretty expensive too and ngl, I loved the features on my MSI board more than this like fan control and OC menu. 

[drumn_bass]

@IkeaGnome

 

This is actually excellent advice, as yes, it seems to be very important, and based on some comments on the Asus forums, a lot of people are missing this step. I have to admit, I did miss it myself when I updated BIOS in October. Now with the latest BIOS, I did that. Updated ME firmware and verified it was on the latest version in BIOS, before performing a BIOS update. So I'm not sure how much difference it can make at this point, but I guess I can try a CMOS reset? I feel like I did that, but I don't remember exactly at which stage of this whole thing. Do you think it's worth a shot? To clarify, it currently has the latest ME firmware, and BIOS version, as well as the latest Intel Chipset and ME drivers in Windows. 

 

I haven't tried a SATA driver manually as I don't use any SATA drives, and wasn't thinking about that. I can give it a try. Won't hurt anything.

 

@CommanderAlex

 

I'm not too worried about data, in fact if I'm going to let anyone touch my computer, I will probably wipe it clean and have a fresh Windows install on it. I'm used to having everything in the cloud or on my home server, so my actual workstation PC can be wiped and OS reinstalled at any point, and I can be back up and running within a couple of hours.

 

Yes, I'll be getting an exact same board on Friday. It's a bit of a pain, but I will go ahead and swap the mobos and see how it goes, then decide on my next steps. I had no issues at all with an MSI board, with both Ryzen (B550) and Intel (Z690), the only thing that was frustrating me with them was a slower boot time. On both systems, it took about 45 seconds to boot into Windows, and it was longer than my old 4770K based PC, so I decided to try Asus and then just liked the board, while everything worked. I never had any good luck with Gigabyte, and haven't tried any other brands, whatever one or two there might be.

[drumn_bass]

@IkeaGnome

 

Ok, so the SATA driver is the RST driver. Yes I tried it, but I got a message about it not detecting supported hardware, however that was a driver directly from Intel. Now with the package from Asus, I used the Asus installer, and it worked. So I'll do some testing and report back.