Nvidia 4090 GPU stuck at PCIE1.1x16

WakeX

Member
Joined
Jun 16, 2023
Messages
100
Nope, only exact same as the last beta:
- AGESA ComboPI 1.0.0.7a updated with the required solution.

So far after punching in all my settings again, I've noticed the following two new entries in the BIOS:
Under OC there is a setting below Memory Context Restore called Memory Failure Retry Count which defaults to 1, nice to have for memory settings which only sometimes fail to train I guess.

Under ACPI Settings there is a new entry called CPU Fan Fail Warning Control which defaults to Disabled, which is also a good addition.
Not sure why it's set to disabled as default though, shouldn't the opposite the case and only need disabled if you're not running a fan/pump but some other external solution like LN2 or a chiller maybe?


The time spent on post code 15 (memory training) was also cut down massively on my settings (MCR auto, PD off) to around 32 seconds - Last BIOS time reported by task manager was way over 100 seconds easily on the prior beta BIOS, now it's consistently below 50 seconds. I wonder if I force MCR to enabled if it would affect anything and be even faster (or if I'd get BSODs like some reported with PD off), gonna try later.

Unfortunately the PCIe 1.1-bug still is present and showed up quite fast on the third and fourth reboot in a row, while on the fifth everything was fine again:
oW9Wr0w.png

Once it runs at 4.0 I don't notice any instability, crashes or the like - so I'm not sure why that happens, if the signaling was bad wouldn't it have issues at 4.0?

I've got a Crucial T700 5.0 4TB on the way, maybe even this week already which will replace the slow WD Blue SN550 3.0 250GB (which will go into an external case) in the slot closest to the CPU - wonder if changing that will impact this pesky PCIe 1.1-issue, if at all.


Oh yeah, almost forgot:
Trying to get into M-Flash to update the screen froze and I was greeted with a 00 on the post code.
I pressed the power button on the case first to try getting a shutdown but that just restarted the system and then it went into M-Flash - after which I flashed the new BIOS without a hitch (should've probably rebooted cleanly and try again but yeah...).
Odd behavior to say the least.
 
Last edited:
Ugh I have almost this exact problem. GPU goes to x1 pcie but in order to get it to go back to x16 PCIE i have to reboot several times. Sometimes 10 times, it eventually will show up x16 and work fine again. Not sure if this is the new BIOS or the GPU itself, haven't had a chance to test on another board or roll back the BIOS yet. I don't even know if it happened with the old BIOS because checking the pcie lanes is not something I do all the time. I've tried reseating gpu, cleaning it, etc.. B650E board .. Just curious if anyone else has tried the new AGESA 1.1.0.0 and had this issue?

The last update for my motherboard was the beta bios in October for AGESA 1.0.8.0 which had issues and there has been no bios updates since

As for this issue, where as with my 4090 it was happening constantly, with my 6650XT it has only happened twice and hasn't happened again since I last mentioned it so it definitely appears to affect AMD gpus completely differently to NVIDIA gpus

As for your issue, I'd suggest going back to the last bios you had that was stable and see if the issue persists, if it doesn't then try report it as an issue so it can be looked into for that bios version
 
Ugh I have almost this exact problem. GPU goes to x1 pcie but in order to get it to go back to x16 PCIE i have to reboot several times. Sometimes 10 times, it eventually will show up x16 and work fine again. Not sure if this is the new BIOS or the GPU itself, haven't had a chance to test on another board or roll back the BIOS yet. I don't even know if it happened with the old BIOS because checking the pcie lanes is not something I do all the time. I've tried reseating gpu, cleaning it, etc.. B650E board .. Just curious if anyone else has tried the new AGESA 1.1.0.0 and had this issue?

In my case, it was the newest BIOS for my MB 7D76vA8 as of this post. Went back to the previous one and all good again.

Do you have the B650 Edge WiFi MB? Couldn't find a B650e model.
 
In my case, it was the newest BIOS for my MB 7D76vA8 as of this post. Went back to the previous one and all good again.

Do you have the B650 Edge WiFi MB? Couldn't find a B650e model.
Tried older bios, still x1 .. put the 4090 in another , older machine which is only pcie 3.0, it works fine x16.. Put a gtx1050 in this board with the new bios and old bios works fine at x16 3.0..
This is a new board so there's no 'old known working bios' that I can try. It has never worked right since I got it a week ago. But if the gtx1050 works at x16 then the slot can't be toast? or the cpu? CPU works fine, literally the only single problem so far has been the 4090 only doing pcie x1 4.0. Very frustrating. Not sure what to do, could be the CPU ? So far i haven't gotten it to go back to x16 mode even with 20 reboots ugh..
I even tried a radeon 6400 (it's pcie 4.0 x4 ) and it works as x4 in the slot.. What's wrong with the 4090, i'm running out of things to try..
I put the 4090 in the chipset pcie x4 slot and it links up at x4 pcie 4.0 .......... So it works at x16 3.0 in my other computer, works at x4 pcie 4.0 and randomly works at x16 pcie 4.0 in the main slot, and once it's at x16 i can use it fine, i used it for days playing games and other stuff benchmarks, torture tests etc.. I didn't want to reboot :/ And radeon low profile works x4 pcie 4.0 in main slot.. and 1050 works x16 pcie 3.0 ... So it's got to be some sort of incompatibility with the 4090??
What AGESA does your bios have that works? i think this one is 1.0.0.7c , I tried 1.1.0.0 also.
 
The last update for my motherboard was the beta bios in October for AGESA 1.0.8.0 which had issues and there has been no bios updates since

As for this issue, where as with my 4090 it was happening constantly, with my 6650XT it has only happened twice and hasn't happened again since I last mentioned it so it definitely appears to affect AMD gpus completely differently to NVIDIA gpus

As for your issue, I'd suggest going back to the last bios you had that was stable and see if the issue persists, if it doesn't then try report it as an issue so it can be looked into for that bios version
.

1701940893172.png
 
Because I'm a bit crazy i rebooted this thing probably 500+ times, changing each bios setting 1 by 1 that had anything to do with pcie or io voltage or things connected to pcie (and there's a LOT OF SETTINGS), I figured something was interfering with the pci lane maybe noise, tried everything setting bifurcation (which worked the card negotiated at x8), but nothing else helped UNTIL:

Set Thunderbolt Support to Disabled in the BIOS and boom 3 reboots in a row so far at x16. If anyone else is having pcie issues try this if your board has thunderbolt
 
Set Thunderbolt Support to Disabled in the BIOS and boom 3 reboots in a row so far at x16. If anyone else is having pcie issues try this if your board has thunderbolt
That's odd I don't know of any MSI AMD Motherboard that has Thunderbolt on them

Just what Motherboard do you have b650e does not say much just that not an MSI Board as there are no B650E boards
 
I was in this thread because It's a MSI 4090, the motherboard isn't MSI buit it's one of the only threads I found that had this issue. So it most definitely is something with the BIOS or board config that is causing weird pcie issues. I guess this thread technically should be in the GPU section and i'm in the MB section but at least if someone else looks it up they might get some answers :)
Hardware:
Asrock B650E Taichi Lite (thunderbolt support connected to pcie lanes on CPU, this is what lead me to try and disable it)
Ryzen 7800x3d
MSI 4090 Gaming X Trio
64G Corsair RAM , two sticks not 4
m2 ssd in chipset slot
nothing else installed for testing, nothing plugged into the thunderbolt port
tried disabling iGPU also with no effect

It seems thunderbolt connects with pcie 5.0 x4 to the cpu , which is where some boards normally have a 2nd m2 5.0 slot , so this could also cause trouble.
I tried off/on the gen5 redriver for the first m2 slot with no effect. Even removing the m2 in first slot no effect.
So my guess is it's something to do with the TB controller, OR the 2nd set of x4 PCIE 5.0 lanes on the CPU which a B650 non-E board could have, and does have. A lot of b650 have pcie 5.0 lanes to the m2 and or 2nd m2 slot, but not the x16.. The main difference in the b650 vs b650E seems to be that the main x16 lanes to the PCIe slots are pcie 5.0 instead of 4.0, but the two 4x 5.0 blocks on the CPU can either be 4.0 or 5.0 on regular b650 (optional by manufacturer).

Possibly the USB4 controller is causing some kind of timing /noise issues , or possibly anything connected to those pcie lanes. Could be something bad in the CPU I suppose.
I'm working with the motherboard manufacturer to get a permanent fix for the BIOS.
I talked to some other people who had this exact same board, and it had the same issue of x1 but only with 4090.
It seems like the TB controller takes a long time to 'detect' on PCIe, I even noticed this when opening CPU-Z, Hwinfo etc it would take several seconds at the PCI scan part where after disable TB it is like 1 second or less.
 
Last edited:
I was in this thread because It's a MSI 4090, the motherboard isn't MSI buit it's one of the only threads I found that had this issue. So it most definitely is something with the BIOS or board config that is causing weird pcie issues. I guess this thread technically should be in the GPU section and i'm in the MB section but at least if someone else looks it up they might get some answers :)
Can you list your hardware as the more info we have the better the odds of finding a fix?
 
Which slots should I have my NVME(s) in to ensure I have maximum power delivery to CPU and GPU?

I have them in 1 and 2 right now I think on the MSI ACE x670e
 
Last edited:
Just an update for MSI X670E CARBON WI-FI users, with the latest BIOS:

View attachment 181967
And SoC voltage to 1.23v, i am not getting the issue ever, someone with MSI X670E CARBON WI-FI and CPU 7XXXx3D have issues yet?

Also i have noticed that the SoC voltage is exactly what i set in the BIOS, with previous BIOS this value wasn't constant and with a positive offset of +0.01/0.02

Appears as though a bios update has finally been released for X670E ACE (7D69v1C1) so I'll update and see if that fixes anything for myself (Carbon Wifi appears to have a slightly newer bios version though?)

I personally haven't experienced X16 1.1 since the previous time I mentioned it here and I was given my money back for my 4090 because they didn't have stock to replace the GPU.

Since 4090s are scarce at the moment and have increased in price I'm going to keep an eye on the prices/stock and debate whether to purchase a new one, but I'd also prefer to wait and see if the bios updates fix the X16 1.1 for others first

Edit:

I'm also still experiencing this.. "Coil Whine" and from what I can tell it appears to be coming from the VRM area..? (Left of the CPU - Where the MSI Dragon Logo is, and also from around the motherboard IO) It appears to just randomly start happening much like the X16 1.1 issue and weirdly enough just like that issue rebooting my PC stops it until it eventually starts doing it again.
 
Oh I mean lanes then I guess?
Ahh, that makes more sense.
There will be no loss with all M.2 Populated and GPU in PCIE_1

Please note due to hardware limitations with the chipset M.2 running on the Chipset will run slower than the same speed M.2 On CPU
 
Experienced X8 1.1 on my 6650XT again today.

Unlike the 4090 the driver doesn't crash so the data of the GPU is still visible and as you can see the GPU appears to be running fine, just the fact it's locked at Gen 1.1 which results in the GPU performing terribly when in-game

4CzA2MX.png


zF1AyZo.png


Edit: Running Bios Ver 7D69v1C1
 
Last edited:
Back
Top