Even after replacing the CPU, the PCIe slot version is still stuck.

gjrtm343154402df

New member
Joined
Feb 23, 2025
Messages
5
My previous setup was:
9700X
X870 Tomahawk WiFi
G.Skill 64GB CL30 6000MHz
RTX 5080
FSP 1000W PSU

Everything was working perfectly fine. But after replacing my CPU with the 9800X3D, everything went completely wrong.

From the very first boot with the 9800X3D, I immediately noticed stuttering on the Windows lock screen—my mouse, keyboard, and display were lagging. It felt like the system was freezing every 3–5 seconds. As the issue worsened, the screen would go black, or I would get a Video Scheduler Internal Error blue screen, causing the system to restart. However, sometimes after a reboot, the system would work normally.

When the system was stuttering, I checked GPU-Z, and I noticed that the PCIe version was stuck at 1.1x16. Even under load, it remained locked at 1.1x16. In some cases, it was stuck at 2.0x16 or 3.0x16, causing severe lag. During these moments of lag, even when the system was idle, the GPU usage would spike abnormally. And when the system did work "normally," the PCIe version remained locked at 5.0x16, rather than dynamically adjusting between 2.0x16 (idle) and 5.0x16 (load) as it should.

Since all of these problems started only after installing the 9800X3D, I decided to swap back to my 9700X, and everything returned to normal. Even in GPU-Z, I could see that PCIe behavior was back to normal—2.0x16 in idle and 5.0x16 under load.

The 9800X3D has already been sent for RMA and replaced. and I have installed the latest version, 1A26 Beta, but the issue remains the same.
 

Attachments

  • 33.gif
    33.gif
    31.7 KB · Views: 543
Last edited:
This can be a common issue with these cards sometimes, see here, here and here for example, and this video (around 10 min mark):


So, it's not the CPU's fault, even though it may seem like that at first glance. It's probably more about slight lack of signal quality on the PCIe bus due to the way the Founders Edition is built, with a center PCB and then going to another PCB to connect to the PCIe slot. So it acts similar to a riser cable integrated into the graphics card, deteriorating the signal quality already. And then your 9700X just had marginally stronger signal quality to begin with, so it just about managed not to be affected by this issue.
 
Last edited:
This can be a common issue with these cards sometimes, see here, here and here for example, and this video (around 10 min mark):


So, it's not the CPU's fault, even though it may seem like that at first glance. It's probably more about slight lack of signal quality on the PCIe bus due to the way the Founders Edition is built, with a center PCB and then going to another PCB via a flat cable to connect to the PCIe slot. So effectively it acts as a riser cable integrated into the graphics card, deteriorating the signal quality already. And then your 9700X just had marginally stronger signal quality to begin with, so it just about managed not to be affected by this issue.
Thanks for your response! However, I have already tried those solutions, and the ASPM L0 setting didn’t really help me. Also, I’m not using a riser cable. The same issue seems to be most common on MSI boards, and this might eventually force me to replace my motherboard.
 
Also, I’m not using a riser cable.

I tried to word it a bit more accurate: The 5080 FE card itself is designed in a manner which is similar to using a riser cable. There is a board-to-board connector on those cards (on the 5090 FE too), so you not only have the contact resistance between the card's contacts and the PCIe slot (like on any normal GPU), you also have another connection on the card itself.

Here you see how it looks:

disassembly4.jpg


After taking off the triangular covers, underneath you see the PCIe board, fastened to the main GPU board with screws.
Screws taken out:

disassembly5.jpg


Underneath (extra connector highlighted):

disassembly7.jpg


PCIe 5.0 has immense requirements for signal quality, since the transfer rate was doubled once more. Any unwanted signal reflections or distortions can cause problems similar to what you're seeing. Also see here: https://www.igorslab.de/en/nvidia-g...t-geforce-rtx-4080-ti-mit-blackwell-genen/10/

There he also suggests as a potential workaround to clock the bus 1 or 2 MHz lower.
 
You are probably wondering why the 9800X3D triggers the issue and not the 9700X?
When the 9800X3D is plugged into the board It will impact the motherboard VRM differently, power supply, voltages lines..., even how it clocks up/down is a source for interference.

If nothing works try another PSU, unplug things and move them around as all electrical devices connected will have some impact but equally the same goes for trying another motherboard or replacement GFX card, take your pick. These issues are always hard to resolve if nothing is actually wrong but only when in a particular combination. (bad design doesnt help)
 
Replace the motherboard with any motherboard from another manufacturer and the problem will go away, this is a PCI-E interface training problem and only occurs with MSI
 
Replace the motherboard with any motherboard from another manufacturer and the problem will go away, this is a PCI-E interface training problem and only occurs with MSI
I also think the motherboard is the problem. So, I purchased the Asus X870 TUF model, which is scheduled to be delivered tomorrow. I will never buy an MSI motherboard again.
 
I also think the motherboard is the problem. So, I purchased the Asus X870 TUF model, which is scheduled to be delivered tomorrow. I will never buy an MSI motherboard again.
Wondering if you have an update on this?

Mainly to help people who might have this issue in the future.

I had a similar issue but with a 4090, so it's not a 50 series issue, that is, if my issue sounds like yours was.

Upgraded from 7800x3d to 9800x3d, then booting the pc, MSI post screen would sometimes freeze for a second before entering windows lock screen, sound would be laggy/stuttering, everything is slowed down and stuttering.
I had to reboot the pc and hope it booted up fine, if it did, it would be fine until i turned off the pc again.
If i continued using the pc when it booted up in a laggy mess, it would error with the internal video scheduler error.

I worked out that if i did not have ANY nvidia drivers installed, it seemed to be fine, this was not an issue with the 7800x3d and the gpu is not faulty.

I have since replaced the x870 tomahawk, i have used the same pc on THREE different b650 motherboards and all work fine. Just posting this as it sounds like we shared a similar issue with the x870 tomahawk and i want to spread the word about this issue.
 
I've had my share of problems with the MSI X870 (tomahawk) and X870E(Carbon) boards and still do to some extent. Even though everything works almost fine now, I share the feeling that this will be my last MSI purchase, too expensive for too many headaches.
The common response I see is, "We're working on it, please be patient." But from my experience, they fix one thing and break another with each BIOS update. For example, with the Carbon model, the A25 beta BIOS was a pleasant surprise, while not perfect, but quite good. Then they released the non-beta A27 BIOS, which, in my opinion, is unstable and performs worse.
I don’t understand what happened to MSI. They used to make decent boards, but now even their $500+ models lack essential features like an external clock and come with tons of issues. You either have to tweak the BIOS (if possible) or wait months for fixes. For an enthusiast gamer and overclocker, exactly who these boards are aimed at, that's a complete failure.
Take the Carbon model as an example: it was released in October 2024, and five months later, we’re still struggling with one issue after another. I don’t know if this is incompetence or a fundamental hardware flaw that simply can’t be fixed.
 
To me it sounds that the board is simply faulty. And these things happen with electronics.
 
Yes, of course
No other manufacturer has problems like this, and the MSI x670 and x870 series just have electronics issues 😇
I replaced the CPU and RAM(with a different part number from the QVL list) to fix the PCIe stuck stuttering issue, but it didn't solve the problem. Some people even RMA'd their motherboards, but the issue still persisted, so I don't think it's a hardware problem. Everyone experiencing this issue is saying it's a BIOS problem, but MSI is not listening and keeps giving basic responses like "try changing the RAM" or "try disabling EXPO".
 
The best solution is to change the motherboard to another manufacturer
Hello
I had the same problem with my Asus TUF GAMING B650-PLUS WIFI (after changing my CPU from 7700x to 9700x).
I got very strong lagging, bsod error, bios selfmade restarts, fails post etc. I could not find a solution and returned the board as defective.

I should certainly not compare a cheaper Asus board with a more expensive MSI board, however the replacement mainboard, Asus ROG Strix B650E-E wasn't that great either: EXPO was not saved, audio issue, WIFI issue.

At present I have the MAG B850 TOMAHAWK MAX WIFI, which is also not bugs-free.

Which manufacturer have you changed to?
 
with my Asus TUF GAMING B650-PLUS WIFI
I got very strong lagging, bsod error, bios selfmade restarts, fails post etc
This is not the same problem (or rather the problems you mentioned) that is being discussed here
The specific problem is lags at Windows startup, if you look at the PCI-E version at times like this it will be below 4.0 and doesn't go up to 4.0 until you reboot the PC
So called PCI-E learning error
Which manufacturer have you changed to?
x870e Nova
 
Back
Top