GIGABYTE Z690I AORUS ULTRA D4? problem

1

1_rick

Supreme [H]ardness
Joined
Feb 7, 2017
Messages
4,610
Bought motherboard in topic on 12/12, put an i5-12600K in it. Ran it that way for a few days, then put an EVGA 1070 in it. Last week I upgraded to an RX 6800, and a couple days later I moved the computer to another room. Since yesterday, I think, about 2/3 of the time when I try to wake the computer up from sleep, the monitors stay dark for a bit, then I get the bios boot logo. At first I thought maybe I'd put shut the computer down instead of putting it to sleep, but just now I checked the event viewer and it crashed. The System log shows about 10K (!) events from about 5:43 (I was on it watching youtube at that time and didn't notice anything weird. About 6:30 I went out to get some supper and there's one error saying "the system session has transitioned from 6 to 7", whatever that means, and then a bunch of stuff at 7:15, which is about when I turned it on, starting with "the OS started" and "Windows failed to resume from hibernate". Then there's a bunch of information messages, an "the system shutdown at 6:20 was unexpected" and a couple of errors that say stuff like "there was an event, but the description can't be found", and a warning saying "The description for Event ID 27 from source e2fexpress cannot be found." and mentioning the NIC (Intel I225-V). The PCI device ID below seems to be "12th Gen Core Processor PCI Express x16 Controller #1"

What's probably bad? Sounds like the CPU or the motherboard, based on the PCI VID. I haven't done any troubleshooting yet.

-----------------------------------
A corrected hardware error has occurred.

Component: PCI Express Root Port
Error Source: Advanced Error Reporting (PCI Express)

Primary Bus:Device:Function: 0x0:0x1:0x0
Secondary Bus:Device:Function: 0x0:0x0:0x0
Primary Device Name:pCI\VEN_8086&DEV_460D&SUBSYS_50001458&REV_02
Secondary Device Name:
 
1_rick said:
Bought motherboard in topic on 12/12, put an i5-12600K in it. Ran it that way for a few days, then put an EVGA 1070 in it. Last week I upgraded to an RX 6800, and a couple days later I moved the computer to another room. Since yesterday, I think, about 2/3 of the time when I try to wake the computer up from sleep, the monitors stay dark for a bit, then I get the bios boot logo. At first I thought maybe I'd put shut the computer down instead of putting it to sleep, but just now I checked the event viewer and it crashed. The System log shows about 10K (!) events from about 5:43 (I was on it watching youtube at that time and didn't notice anything weird. About 6:30 I went out to get some supper and there's one error saying "the system session has transitioned from 6 to 7", whatever that means, and then a bunch of stuff at 7:15, which is about when I turned it on, starting with "the OS started" and "Windows failed to resume from hibernate". Then there's a bunch of information messages, an "the system shutdown at 6:20 was unexpected" and a couple of errors that say stuff like "there was an event, but the description can't be found", and a warning saying "The description for Event ID 27 from source e2fexpress cannot be found." and mentioning the NIC (Intel I225-V). The PCI device ID below seems to be "12th Gen Core Processor PCI Express x16 Controller #1"

What's probably bad? Sounds like the CPU or the motherboard, based on the PCI VID. I haven't done any troubleshooting yet.

-----------------------------------
A corrected hardware error has occurred.

Component: PCI Express Root Port
Error Source: Advanced Error Reporting (PCI Express)

Primary Bus:Device:Function: 0x0:0x1:0x0
Secondary Bus:Device:Function: 0x0:0x0:0x0
Primary Device Name:pCI\VEN_8086&DEV_460D&SUBSYS_50001458&REV_02
Secondary Device Name:
Click to expand...
I have the same motherboard. Its been great.


Did you update the bios? The latest one is from January 12th.

When you switched to an AMD card, did you refresh/reinstall Windows?

Have you tried reseating all of the hardware? (GPU, RAM, CPU, etc.)

Do you have a spare power supply?
 
Oh, crap.

https://community.intel.com/t5/Proc...usands-of-WHEA-17-errors-a-minute/m-p/1343311

I get the same errors and believe it is a chipset driver issue for Z690 and Alder Lake. Fresh WIN 11 install, updated bios for ASUS Z690-A Gaming WIFI MOBO, 12900K, ASUS TUF OC 3090 GPU, Samsung 970 PLUS as Boot Drive in M.1 slot, and a Samsung 970 PLUS in 2nd M.1 slot.
Others experiencing the same thing. Only solution so far is putting GPU in 2nd PCI slot running at GEN 3 speeds (kinda beats the point of having a GEN 4 card). At this point I'm guessing this devolves into finger pointing between Intel and Motherboard manufacturers in regards to whose drivers are broke for this....insert Spiderman meme.
Click to expand...
I could try setting the slot to gen 3 speeds, although obviously with an ITX board there's only one slot.
 
chameleoneel said:
I have the same motherboard. Its been great.


Did you update the bios? The latest one is from January 12th.

When you switched to an AMD card, did you refresh/reinstall Windows?

Have you tried reseating all of the hardware? (GPU, RAM, CPU, etc.)

Do you have a spare power supply?
Click to expand...
So far all I've done is find the forum link posted in #3. I'm on the F6 bios, which is the current one. I did that before swapping cards.

Before I got the new card, about a week ago, I guess, Windows Update got stuck, so I did an in-place upgrade. When I got the new card I rebooted to safe mode and ran DDU before taking out the 1070. I have not yet tried reseating anything yet--that was going to be next.

According to the forum link I posted above, it was claimed to be a problem for a number of people with z690s, across all mobo makers.

The only other PSU I have is in another computer already. The one in this computer is a Corsair RM750 that I bought last summer or fall, and is bigger than Gigabyte's recommendation for the 6800. I can swap 'em if necessary but it'll be annoying.
 
chameleoneel said:
That's old. if it were a driver issue, it would be resolved by now. Do you have the latest drivers and windows updates?
Click to expand...
Latest Windows updates. Doing driver checks now. I have the latest BIOS, haven't yet checked anything else.

If it matters, I'm using Gigabyte's drivers, not the ones from nVidia, unless Windows Update updated them.
 
1_rick said:
Latest Windows updates. Doing driver checks now. I have the latest BIOS, haven't yet checked anything else.

If it matters, I'm using Gigabyte's drivers, not the ones from nVidia, unless Windows Update updated them.
Click to expand...
Intel is weird about chipset drivers nowadays. they have like 3 or 4 different things which you need to have. The chipset INF update utility doesn't even say it supports anything newere than Z170. But I run it anyway, when I'm doing al fresh intel build. Gigabyte has the HID filter and serial I/O driver posted. so grab those.

I also do Intel Management Engine

Then, I use the Intel® Driver & Support Assistant, to grab the newest drivers for the stuff it grabs.

And then let Windows Update take care of the rest. I assume Intel's chipset updates come through windows update nowadays. Because, the driver and support assistant doesn't handle that.

and of course, make sure you have the latest graphics drivers from AMD, for your RX 6800.
 
chameleoneel said:
Intel is weird about chipset drivers nowadays. they have like 3 or 4 different things which you need to have. The chipset INF update utility doesn't even say it supports anything newere than Z170. But I run it anyway, when I'm doing al fresh intel build. Gigabyte has the HID filter and serial I/O driver posted. so grab those.

I also do Intel Management Engine

Then, I use the Intel® Driver & Support Assistant, to grab the newest drivers for the stuff it grabs.

And then let Windows Update take care of the rest. I assume Intel's chipset updates come through windows update nowadays. Because, the driver and support assistant doesn't handle that.

and of course, make sure you have the latest graphics drivers from AMD, for your RX 6800.
Click to expand...
I just ran down the Gigabyte driver page and installed all of them, even the stuff where I already had what was current. I was going to ask about video drivers next, figuring the answer was probably yes, but for some reason a lot of sites seem to say to prefer whatever is on the video card's driver page for whatever reason. At any event, my video drivers are from October or something so I'll grab the latest now.

Still seeing the WHEA errors--after my last reboot I checked Event Viewer and there's thousands of them from just a few minutes ago.
 
run the Intel® Driver & Support Assistant

and then check windows update again, just to be sure.
 
I would also try loading the bios defaults. sometimes between bios updates, things can get stuck somehow. and re-loading the defaults can fix that. Its tough to explain, but I have seen such a thing fix issues, a few times.
 
chameleoneel said:
I would also try loading the bios defaults. sometimes between bios updates, things can get stuck somehow. and re-loading the defaults can fix that. Its tough to explain, but I have seen such a thing fix issues, a few times.
Click to expand...
Yes, I've seen that before. I don't think I've done anything but turn on XMP, so that fortunately only means there's not much to put back.
 
1_rick said:
Yes, I've seen that before. I don't think I've done anything but turn on XMP, so that fortunately only means there's not much to put back.
Click to expand...
No I don't mean user custom settings. I mean whatever low level changes from one bios to another. sometimes loading defaults (even if no settings were previously changed) and/or taking out the cmos battery, somehow clears things and fixes issues.
 
chameleoneel said:
I mean whatever low level changes from one bios to another. sometimes loading defaults (even if no settings were previously changed) and/or taking out the cmos battery, somehow clears things and fixes issues.
Click to expand...
Sorry, guess I wasn't clear. I was saying

1. yes, I have seen "load d efaults" mysteriously fix things, so I'll try it, and
2. fortunately, I didn't set a lot of things, so I don't have a lot to do once I wipe out my changes.

Also, thanks for your suggestions so far. Installing the latest Adrenalin 22 drivers from AMD's site now, and then will run Windows Update again and do the bios thing. Already ran the Intel support assistant, although all it wanted to do was update wifi and BT.
 
Samsung 970 EVO 500GB. IIRC it's not all that old, maybe a year?

I just rebooted after windows update + latest GPU drivers + bios "load optimized defaults" and fired up a game, and I'm still seeing the errors--dozens, if not hundreds a second. The event viewer said there were 12,500 entries and they were all between 9:48:04 and 9:50:00.
 
1_rick said:
Samsung 970 EVO 500GB. IIRC it's not all that old, maybe a year?

I just rebooted after windows update + latest GPU drivers + bios "load optimized defaults" and fired up a game, and I'm still seeing the errors--dozens, if not hundreds a second. The event viewer said there were 12,500 entries and they were all between 9:48:04 and 9:50:00.
Click to expand...
I wonder if your board is just bad. Maybe they had a bad run of some sub-components or something.

BTW, I have a 970 pro and 960 pro installed on this same board. As well as a budget Adata SP550 512GB Sata SSD from like 2015----which is actually my windows drive. (Installed W11 when I got this board).
 
chameleoneel said:
I wonder if your board is just bad. Maybe they had a bad run of some sub-components or something.
Click to expand...
You think it's more likely to be the mobo, or the CPU? The PCI Vendor ID was 8086, which is Intel, and the device ID is "12th Gen Core Processor PCI Express x16 Controller #1" (This was in the original post but it was kinda buried.)

I'd prefer not to but I can still play "reseat all the components", and even put the old GPU in if need be.
 
Try the reseat snd then the old card first. That controller, if valid, is on the CPU most likely.
 
Haven't had a chance to reseat yet--will do that this afternoon, but I did have enough time to go into the BIOS and disable the iGPU and set the video card slot to Gen 3. Suddenly, on reboot, no WHEA errors, just like other people were having. Of course, I don't know--yet--which thing made the problem go away so now I have to A/B test.

Doesn't seem to affect the games I play performance-wise but I don't like it and I don't consider it a permanent solution. Guild Wars 2 seemed to reliably cause thousands of those errors but isn't now. GPU-Z reports a GPU load of 90-99% when playing that game, though. Oof, too bad the game's not a big enough money maker to hire someone who can fix that.
 
Re-enabling the iGPU didn't cause the errors to come back, so it appears dropping the PCIe bus to Gen 3 is a stable workaround. Look at these GPU stats, though, during a world boss fight in a like 10 year old MMO. In Minecraft the clock tends to sit aroung 500MHz.

v
1645044236411.png
 
OK, pulled the video card, RAM, and CPU. Cleaned and repasted the CPU & HSF, put everything back, set the CPU PCIe link speed to "auto", powered up, ran a game, no errors, but GPU-Z said it was running at 3.0, not 4.0, and CPU-Z said 8GT/s link speed, while it was capable of 16. Power down, change from "auto" to "Gen 4", boot up, fire up Event Viewer first thing, boom, dozens of WHEA-Logger warnings a second. Haven't even got a game running, just Edge with this one tab, the Event Viewer, and GPU-Z.
 
Have you overclocked the cpu by changing the base clock? That could destabilize the bus when it's pushed to the limit. No pcie riser cables, right?
 
1_rick said:
OK, pulled the video card, RAM, and CPU. Cleaned and repasted the CPU & HSF, put everything back, set the CPU PCIe link speed to "auto", powered up, ran a game, no errors, but GPU-Z said it was running at 3.0, not 4.0, and CPU-Z said 8GT/s link speed, while it was capable of 16. Power down, change from "auto" to "Gen 4", boot up, fire up Event Viewer first thing, boom, dozens of WHEA-Logger warnings a second. Haven't even got a game running, just Edge with this one tab, the Event Viewer, and GPU-Z.
Click to expand...
Without another mobo, its impossible to know if its the CPU or the board.

With a gun to my head, i would say its the board. Only because I've rarely seen a CPU which truly had issues.
 
  • Like
Reactions: AP514
like this
Nobu said:
Have you overclocked the cpu by changing the base clock? That could destabilize the bus when it's pushed to the limit. No pcie riser cables, right?
Click to expand...
No overclock (beyond whatever multi-core enhancement may be set by default in the bios), and no riser cable.
 
I been digging around on this a bit. The issues seems pretty common. And seems to be speicifically associated with running the board in PCI-E 4.0 mode with a graphics card. (I haven't seen anyone with say, a 3.0 GPU and a 4.0 SSD say something about this issue. But it could still cause it).

I haven't had any problems with this board. But.....I also realized, I don't have a 4.0 GPU! However, I will in a few days, when Best Buy allows me to go pick it up.

Sucks because this is the only DDR4 ITX board, besides a junk one from asrock. Asus went ahead and made their B660 ITX board in DDR5 only.

Hopefully gigabyte sorts this out and allows people to get RMA replacements. It seems like a hardware issue and not something which they can fix with a bios update.
 
So, if you have a Gen3 card and attempt to run it at Gen4 speeds, if it actually tries to run, you're effectively overclocking the card. That'd explain why it was having bus errors and other issues.
 
Nobu said:
So, if you have a Gen3 card and attempt to run it at Gen4 speeds, if it actually tries to run, you're effectively overclocking the card. That'd explain why it was having bus errors and other issues.
Click to expand...
No. what i'm saying is, it seems that all of the people with problems, have PCI-E 4.0 gpus.

And indeed, the OP didn't have problems until they upgraded from a GTX 1070 to an RX 6800
 
chameleoneel said:
the OP didn't have problems until
Click to expand...
That is correct. To be fair, my 6800 runs fine in 3.0 mode in the games I play, and with great game rates. Others have had worse luck, where the problem is only reduced.

By the way from what I've seen since my last post, this is not specific to this board or even gigabyte. Apparently, the F6 bios was supposed to correct the problem, but it doesn't.
 
1_rick said:
That is correct. To be fair, my 6800 runs fine in 3.0 mode in the games I play, and with great game rates. Others have had worse luck, where the problem is only reduced.

By the way from what I've seen since my last post, this is not specific to this board or even gigabyte. Apparently, the F6 bios was supposed to correct the problem, but it doesn't.
Click to expand...
Trying my board out with this 6600 xt tonight
 

Attachments

  • 20220223_175052.jpg
    20220223_175052.jpg
    313.6 KB · Views: 0
  • 20220223_175103.jpg
    20220223_175103.jpg
    247.5 KB · Views: 0
  • 20220223_175130.jpg
    20220223_175130.jpg
    161.2 KB · Views: 0
  • 20220223_175157.jpg
    20220223_175157.jpg
    219.7 KB · Views: 0
chameleoneel said:
Trying my board out with this 6600 xt tonight
Click to expand...
That's apparently a PCIe 4 x8 board. Be interesting to see what happens--is it the Gen 4 or the full 16 lanes?

Either way, be interesting to see what happens.
 
1_rick said:
The System log shows about 10K (!) events from about 5:43
Click to expand...
So just to be clear, you opened up Event Viewer and in the Overview and Summary page, it says there were 10,000 "error" or "critical"?


If that's the case then.....so far it seems ok for me. I tested Dark Souls 3 at 1080p and 1440p. I also tested the Superposition benchmark. And I watched a couple of youtube videos.

Indeed, the 6600 XT only uses an 8x multiplier. So if its running in PCI-E 4.0 ---- its 4.0 8x.

If you put it into a board and/or with a CPU which is only capable of 3.0 or force a 4.0 board to run in 3.0 ---- it runs at 3.0 8x

I dunno if that's simply low enough to avoid errors. or if my board is simply ok.
 
Last edited:
chameleoneel said:
So just to be clear, you opened up Event Viewer and in the Overview and Summary page, it says there were 10,000 "error" or "critical"?
Click to expand...
No, I went into the System log itself, right after a reboot, and saw it say ~15,000 events at teh top. Then I paged down a few time and estimated 20-50 WHEA 17s a second, and then scrubbed through the entire thing, seeing nothing but WHEA 17s until the very bottom, where there were a few hundred (estimate) messages that were regular startup stuff. The range of timestamps from end to beginning of those 15,000 total messages was only a few minutes. I will admit that's not the same as spending however long it would take to verify my estimate that nearly all of the messages in the count at the top of the window were WHEA.

IIRC the actual WHEAs show up as Informationals or Warnings, and the text when you click on one says "corrected error". They've all fallen off now, so I'd have to reboot into the BIOS and set it back to Auto/Gen4 to be more specific. I just noticed there's a filter option so if I wind up doing that I'll try to do the filter to get a better estimate.

chameleoneel said:
Indeed, the 6600 XT only uses an 8x multiplier. So if its running in PCI-E 4.0 ---- its 4.0 8x.

If you put it into a board and/or with a CPU which is only capable of 3.0 or force a 4.0 board to run in 3.0 ---- it runs at 3.0 8x

I dunno if that's simply low enough to avoid errors. or if my board is simply ok.
Click to expand...
Well, I hope for you it's the latter but the former should be OK (as long as the board doesn't degrade with time, of course, alhough it's probably too early to tell if that's happening. Nobody seems to have reported anything like that that I've seen.

I would see this happen both at the Windows desktop and in Guild Wars 2, and IIRC not when playing Minecraft. It happens that I play the former in windowed fullscreen and the latter just as a maximized window. No idea if that matters, at least not without testing.
 
D-EJ915 said:
Seeing tons of whea errors reminded me of people having this same sort of issue on EVGA boards before. Seems like they released a bios which changed something to resolve it.

https://forums.evga.com/z590-FTW-WHEALogger-Warning-PCI-Express-Root-Port-m3439132.aspx

Not sure if this is related or helps but just thought I'd mention it.
Click to expand...
That's the same error, but a different BIOS generation (590 vs 690), and the poster says the error appears to be in the PCIe root hub, whereas with the 690s, Event Viewer (at least in my case) shows a PCI VID/PID for the CPU itself.

Gigabyte's released at least one BIOS intended to fix this for my mobo, but it hasn't worked for a lot of people. Hopefully they (and the other mfgs) will get it eventually.
 
Optimum Tech’s most recent video in which he moves to a 12900k platform on a Gigabyte z690i board has this issue as well. RTX 3090 and Whea errors running pcie 4. The issue goes away under 3.0 set in bios. Seems like the board or bios certainly

For what it’s worth I’m in a 12700k , Tuf D4 z690 and pcie gen4 running fine with a reference RX6800.
 
1_rick said:
Bought motherboard in topic on 12/12, put an i5-12600K in it. Ran it that way for a few days, then put an EVGA 1070 in it. Last week I upgraded to an RX 6800, and a couple days later I moved the computer to another room. Since yesterday, I think, about 2/3 of the time when I try to wake the computer up from sleep, the monitors stay dark for a bit, then I get the bios boot logo. At first I thought maybe I'd put shut the computer down instead of putting it to sleep, but just now I checked the event viewer and it crashed. The System log shows about 10K (!) events from about 5:43 (I was on it watching youtube at that time and didn't notice anything weird. About 6:30 I went out to get some supper and there's one error saying "the system session has transitioned from 6 to 7", whatever that means, and then a bunch of stuff at 7:15, which is about when I turned it on, starting with "the OS started" and "Windows failed to resume from hibernate". Then there's a bunch of information messages, an "the system shutdown at 6:20 was unexpected" and a couple of errors that say stuff like "there was an event, but the description can't be found", and a warning saying "The description for Event ID 27 from source e2fexpress cannot be found." and mentioning the NIC (Intel I225-V). The PCI device ID below seems to be "12th Gen Core Processor PCI Express x16 Controller #1"

What's probably bad? Sounds like the CPU or the motherboard, based on the PCI VID. I haven't done any troubleshooting yet.

-----------------------------------
A corrected hardware error has occurred.

Component: PCI Express Root Port
Error Source: Advanced Error Reporting (PCI Express)

Primary Bus:Device:Function: 0x0:0x1:0x0
Secondary Bus:Device:Function: 0x0:0x0:0x0
Primary Device Name:pCI\VEN_8086&DEV_460D&SUBSYS_50001458&REV_02
Secondary Device Name:
Click to expand...
Gigabyte (finally) confirmed this WHEA bug in mid-January, but only in a tiny thread: it should be acknowledged publicly. I've not seen any new updates since. :(

Gigabyte claimed it was related to the mini-ITX size (??) even though other ATX motherboards like the ASUS Z690-A and ASUS Z690 Maximus Hero are having almost identical issues.

Hi all,
Regarding the recent complaints that we received on the Z690I motherboards we truly apologize for the issues that you are experiencing.
We are still investigating the best way to make it as less annoying for end-users as possible.
The newer bios versions that can be downloaded from our website will improve the PCI-e stability.
Z690I AORUS ULTRA DDR4: F6 bios version
Z690I AORUS ULTRA (DDR5): F5 bios version
(They will be visible in few hours when the website is updated)
The PCI-e 5.0 eye diagram passes all the tests and there is no impact on the graphics performance. Also, it is designed according to the industry’s standards but it seems that in combination with the small form factor size there is still some interference that could not be detected in benchmarks and it causes WHEA errors.
Our team is still working on this issue and we will let you know for any further updates.
Thank you for your support.
Click to expand...
 
ikjadoon said:
Gigabyte (finally) confirmed this WHEA bug in mid-January, but only in a tiny thread: it should be acknowledged publicly. I've not seen any new updates since. :(
Click to expand...
True. I'm the first person I've seen with the problem with an AMD card.

I meant to say--last night I noticed a new Microsoft INF driver, dated 2/23 or so, on the mobo download page. I didn't really expect it to help, but I installed it anyway, and when I set the slot back to Auto, the WHEA errors came back, so the INF didn't help.

ikjadoon said:
Gigabyte claimed it was related to the mini-ITX size (??) even though other ATX motherboards like the ASUS Z690-A and ASUS Z690 Maximus Hero are having almost identical issues.
Click to expand...

Also, as has been mentioned, it's not confined to Gigabyte, either.
 
My board does have "coil whine".

Anytime something is processing, even the lightest thing, it buzzes and makes electrical sounds. Even something as dumb as holding your mouse over a game in steam, and then steam cycles through some small screen shots. As each screenshot cycles, it will go "bzzz, bzzzt, bzzt, bzzzzt, bzzzt,".
 
You must log in or register to reply here.
Back
Top