GPU performance capped at zero in afterburner while core clock remains at 420mHz
GPU performance capped at zero in afterburner while core clock remains at 420mHz
I'm working with a 3090 FE and recently started optimizing my 7900X3D by installing the AMD Chipset installer. Since then, my power limit has remained stuck at 0%. Every time I adjust the slider or change the temperature settings, applying those changes always resets it back to 0%. I've looked into it and considered flushing the VBIOS, but I'm not familiar with the process and don't want to risk something. It's odd that the power limit isn't set to 100% by default. I also tried using GPU-Z to retrieve VBIOS details, but it displayed an error message saying the information couldn't be read.
I’d try a DDU on the 3090 and perform a new driver install after rebooting. It seems the chipset driver may have changed something related to your CPU’s IGP/driver, causing a conflict with the GPU driver you already have.
I’ll try it, this could seem like a joke but should I pick my IGP over the desecrate GPU? If I reinstall the driver, it should remove it and then reinstall again, right? Using a dedicated GPU when the driver gets removed might stop video output. I’ve never worked with DDU, any tips or guides would be super helpful. Thanks! Hope this helps me get back to Cyberpunk soon.
You've been working on this for about 5 to 6 hours. I used DDU, but even after setting up the base coat and installing the NVIDIA driver, my core clock remains stuck at 420mhz. Sometimes the power limit slider stays unlocked, but after checking the core voltage, it still doesn't work. My temperature slider seems stuck at zero no matter what I adjust. I can change the core and memory clock, but the slider just stays fixed. I've tried three different driver versions before, always cleaning them out and reinstalling, yet the issue persists. I'm starting to wonder if something inside the card is faulty, or if I'm losing my mind from trying so hard. Flashing the VBIOS might help, but I'm really worried I've wasted a lot of money on this.
You've encountered some persistent issues with your BIOS settings and memory clocks. It seems the problem might be related to the BIOS configuration itself, especially since changing drivers didn't resolve it. Try resetting your BIOS settings to default or consult the manual for specific troubleshooting steps. If the card still doesn't work, consider testing it with another motherboard or checking for hardware compatibility.
Did you ever discover a solution for this? I own a 4090 that experiences this problem occasionally during a restart or boot-up. Restarting a few times usually resolves it, but I've already spent over 4k on this build and don't want to have to reboot just to fix it. Clearly, games perform poorly when this occurs because the GPU power limit is stuck at zero. Every time it happens, it logs event ID 14 nvldmm in the event viewer. I've tried using ddu and installing a fresh Windows version, but the issue persists.
Same situation here, also 4090. I swapped out the motherboard after the PCI-e port failed (likely due to weight, even with support). Now it’s stuck at 855Mhz on the X570 board and the power limit is set to zero. Every save just resets it back. I’m using a cable mod riser, but I’m not sure if that’s the problem.
A lot of time has gone by. Through many troubleshooting steps, I realized the problem might be with my motherboard. There was a lot of drama with Ebuyer because I sent the motherboard for return and they bent the pins, trying to shift blame. I had to threaten legal action just to get a refund. A new motherboard arrived, but it worked until I reinstalled either the iCue or Lconnect. It seems one of those models might be causing the strange behavior—everything was fine before, then it started acting up after installation. I’m planning to uninstall them and see if anything changes.
Uninstalled iCue and Lconnect 3 again, but the problem remains. I’m updating to the newest drivers and hoping it fixes things, though I’m really frustrated. Everything worked fine the previous night, but using YouTube and installing iCue or something else seems to have caused the issue.
I've compiled a record of all my actions since the system started. Despite restarting multiple times, the problem kept coming back. I ended up reinstalling my RTX 3060, which cost me £700. Messages kept suggesting I install GCC, but I’d already added chipset and AMD drivers without success. I upgraded to Windows 10 Pro, activated the OS, and added iCue and Lconnect 3. Updated my DDR5 firmware via iCue—though I wasn’t sure RAM needed updates. Played FIFA 23, turned off Elgato Camera Hub and Opera GX at startup, watched YouTube, and tried to stream Jedi Survivor.