Asus Z690 Hero + RTX 5090 requires PCI Express Gen.3 / 4 for stability

[darkcg]

Hello, I'm experiencing a weird situation with my Asus Z690 Hero Motherboard and my new RTX 5090 card.
The system goes black screen as soon as it reaches the desktop (sometimes even before, when loading the operating system). Basically it is unstable and it won't allow you to make any work at all. This happens not only in Windows but also in Linux (Ubuntu 22.04 LTS) so there is nothing specific to the operating system, I think.

But I discovered that putting the PCI Express from "Auto" to Gen.3/4 fixes the issue, I've put the system under stress tests using 3DMark and everything seems to be fine. The only problem is I'm losing a good 10%/5% of performance with this setting alone.

Any suggestion? Do you think it's a problem with the motherboard BIOS?

EDIT: Of course I have the latest BIOS version (4101) and the latest nvidia hotfix driver (572.75), both on Linux and Windows.
The card is a Zotac RTX 5090 AMP Extreme Infinity. I can't do other tests with other models right now, but a new Gainward RTX 5090 Phantom is arriving, so I'll do other tests.

EDIT 2: Latest nVidia 576.28 doesn't change anything with Gen.5, still receiving no video signal at boot with Gen.5 enabled, this is a video showing what's happening:
https://www.youtube.com/watch?v=yMydcRKGbp0

EDIT 3: This article seems to claim the issue is within the motherboard firmware:
https://www.guru3d.com/story/troubleshooting-rtx-5090-black-screen-failures-switch-to-pcie-gen-40/

Detailed technical analysis of the phenomenon reveals that the issue is closely associated with the data transfer protocols inherent to PCIe 5.0. When gurus install the latest drivers on a system equipped with a PCIe 5.0 motherboard, the interaction between the driver software and the hardware interface appears to cause an abrupt loss of display output. In many of the documented cases, the graphics card ceases to function once the driver installation reaches a critical point, implying that the high-speed signaling requirements of PCIe 5.0 may exceed the compatibility parameters set by the current firmware on some motherboards.

[SneakyLittleman]

OK...I guess testing was quick. PCIE4 is not stable for me either (much more stable than PCIE5 but it's still a crashfest) - downgrading to PCIE3...we'll see.

[darkcg]

Are you using a Z690 Hero? Or another motherboard?

[SneakyLittleman]

Hey there, I come with good news - also using a Z690 Hero:

Just turn off all ASPM features in the Bios (and gating option) - I can now run my AORUS 5090 in PCIe Gen4 mode without crashes. Phew.

I got the tip when google for Z690 Hero + 4090 - the previous generation had the same problem with this mobo. Asus support was useless and answered with AI chatbots...

Goo luck!

[darkcg]

@SneakyLittleman 
Unfortunately, that is not a very good option. Disabling all the features in the ASPM settings will consume a significant amount of energy. Furthermore, it will negatively impact performance, as processors like the Intel Core, which use hybrid architectures, rely on deactivating on-chip components to boost the performance of cores involved in computations. Disabling all the settings will lead to higher energy consumption and degraded performance.
If disabling the ASPM options resolves your issues, try re-enabling them and adjusting the power settings in the NVIDIA drivers to 'Max.' This approach solved my issues, and, oddly enough, the PC remained stable even after resetting the settings to their defaults.

[SneakyLittleman]

ok, another good catch - stable so far. I will try a high fps scenarios - this is when my system crashes the most.

[darkcg]

If using the power settings in the nvidia driver set to max works for you, try the next step: bring them to normal leaving the PCI Express Gen.4 activated. I don't know why, but this remained stable once and for all and I don't need Gen.3 anymore. I know, there is no logic but worked for me.

[breddy]

i have same problem with Asus ProArt Z690-CREATOR WIFI and Gigabyte GeForce RTX 5080 WINDFORCE OC SFF 16G:

UEFI BIOS booting occurs, you can see the logo on boot, but Windows 11 boots to a black screen. This happens every two or three days if pcie gen5 is enabled

I switched the slot to pci-e gen4 and disable PCI-E Native Power Management and ASPM (and fastboot). After this all works as expected on pci-e gen4 speed

It looks very much like the problem is with the motherboard

[lemonhead247]

Can confirm I have same issue. Z690 Hero. Had 4090 working fine, Gigabyte 5090 instability city. Will try tips here, had moved PCIEx to 5. Didn't think going down was the answer but I'll try anything at this point.

[darkcg]

@lemonhead247
Please let us know how it goes. The annoying part here is a total lack of communication from Asus reps.

[lemonhead247]

Moved from 5 to 4 PCIEx and things appear stable so far.  Played a round of COD, watched youtube videos and mp4s.  Usually if I have issues it's in the first minute after booting.  Seems solid.  My heart still skips a beat whenever the screen goes blank. I also was concerned about loss of performance, but after watching this and reading the article, i think i can live with it till next upgrade. https://www.youtube.com/watch?v=L1NPFFRTzLo