Drivers & Software

[BlackJacque]

Hello.

ASUS B650E-F ROG Strix Gaming WiFi Motherboard  1807 BIOS
AMD Ryzen 9 7900X
G.Skill Flare X5 Series 32GB (2x16GB) DDR5-6000 RAM
Noctua NH-C14S
Crucial T700 M.2 2TB NVMe SSD
PowerColor AMD Radeon RX 7900 XT
SilverStone Technology Extreme 850R Platinum Cybenetics Platinum 850W SFX12V
Sligar Cerberus X

I bought the mobo, RAM, and CPU in a MicroCenter bundle.  I kept it too long before the build to return the parts to MicroCenter.

This machine is a heartache.  Random BSODs in Windows 11 making it unusable.  Fewer BSODs in Windows 10. CHKDSK would BSOD.  Windows Load was iffy.  DISM and SFC report corrupted DLLs after BSODs occurring after fresh Windows installs.  Fewer still BSODs not running the Power Color 7900 XT installed.  That is running on the CPU's GPU.  

I RMA'd the mobo back to Asus with the C:\ included for "Instability".  No joy.  There is no record in the Log that they ever booted into Windows.  They ran their motherboard H/W diagnostic and returned it with "Passed".

The RAM passes Windows Memory Test.  I've been running it without EXPO since the crashes started.

The SSD passes Crucial's SSD diagnostic.

I doubt the CPU is defective.

Chasing down the Memory Dumps the BSODs are moving around.  AMDKMDAG.SYS and Video Schedule Internal errors are the latest causes of BSODs. 

I've cleaned the video with DDU and re-installed the most recent AMD graphic driver only.  The best it gets is I'm now seeing application crashes (browser and J. River Media Center) that play videos.  

Oddly, the box will run a stress test for more than an hour without BSODing.  The problem appears to be video related?

Anyone with any experience with this MicroCenter Bundle and have any ideas on how I can get this box 'back on the rails' and using the Radeon 7900XT?

[knepich]

Have you tried to lower your cpu clock freq by say 100mhz?

With the latest amd drivers i had to revert to stock cpu settings, and my cpu pulls more voltage to be stable. If i run 23.5.2 drivers i can OC my cpu to great values but the newer drivers i need more voltage to be stable at stock speed. Else my gpu drivers fail and bsod with also leads to corrupt windows.

[BlackJacque]

>Have you tried to lower your cpu clock freq by say 100mhz?

No.

I'm running the 31.0.22023.1014  10//25/2033  AMD Driver now.

This being power-related makes sense, the way the bugchecks 'move around'.  That I upgraded to a 850W PSU from a 750W  Corsair SF750, and the problem persists, without slotting the external GPU makes me wonder, how much more wattage is needed?

[jrronimo]

I know your RAM passes the Windows RAM test, but that doesn't mean it's beyond suspicion. Try pulling out one of your two sticks and seeing if the crashes keep happening. If they do, try the other stick.

I usually also suspect PSU in situations like this. That you're getting the same error with two different units makes me think it's not likely the issue, but you never know. I threw your build into PC Part Picker here: https://pcpartpicker.com/list/ptYzkJ -- it gives an estimated wattage of 589W, but you usually want to have that be 80% of the load of the full PSU, so: (589/.8) = 736.25W -- so a 750W or better should suffice.

I see in your reply that you have an external GPU... Do you have an enclosure that you're putting the 7900XT into? That could also be the issue... Can you put the GPU in the motherboard directly just to test?

Make sure you have the latest BIOS installed: https://rog.asus.com/us/motherboards/rog-strix/rog-strix-b650e-f-gaming-wifi-model/helpdesk_bios/ 

And make sure you have the latest chipset drivers: https://www.amd.com/en/support/chipsets/amd-socket-am5/b650

Finally, what are the CPU temperatures when you crash? I like CoreTemp (https://www.alcpu.com/CoreTemp/ - click "Tools" > "Logging On"). The log will be located in c:\program files\coretemp as a .csv file.

When I work on a problem child like this, I always try to simplify as much as possible. Remove anything that isn't absolutely necessary.

You might even consider things like trying a different HDMI/DisplayPort cable, or even a different monitor if you have one available.

But my main hunch is the external GPU setup without being there to inspect it in person. (And if you're in Colorado, I'd be happy to do that! )

Good luck!

[jrronimo]

I'm also a little suspicious of that SSD since you're having so much trouble even loading Windows. Does it have a heatsink on it?

Crucial has drivers for it too that I think most people wouldn't normally think to install:

https://www.crucial.com/support/ssd-support/t700-support

Do you have a spare m.2 you could toss in just for troubleshooting? If not, you might snag like a 1 TB Samsung 980 Pro for $80. If it helps, RMA the Crucial and great! If it doesn't, you now have a 1 TB extra drive for games.

[jrronimo]

I know your RAM passes the Windows RAM test, but that doesn't mean it's beyond suspicion. Try pulling out one of your two sticks and seeing if the crashes keep happening. If they do, try the other stick.

I usually also suspect PSU in situations like this. That you're getting the same error with two different units makes me think it's not likely the issue, but you never know. I used a calculator to estimate the wattage of your build, and it gives an estimated wattage of 589W, but you usually want to have that be 80% of the load of the full PSU, so: 589 divided by .8 = 736.25W -- so a 750W or better should suffice.

I see in your reply that you have an external GPU... Do you have an enclosure that you're putting the 7900XT into? That could also be the issue... Can you put the GPU in the motherboard directly just to test?

Make sure you have the latest BIOS installed.

And make sure you have the latest chipset drivers.

Finally, what are the CPU temperatures when you crash? I like CoreTemp (click "Tools" > "Logging On"). The log will be located in Program Files slash CoreTemp as a csv file.

When I work on a problem child like this, I always try to simplify as much as possible. Remove anything that isn't absolutely necessary.

You might even consider things like trying a different HDMI or DisplayPort cable, or even a different monitor if you have one available.

But my main hunch is the external GPU setup without being there to inspect it in person. (And if you're in Colorado, I'd be happy to do that! )

Good luck!