Multiple GPUs | Users Grapple with NVIDIA Driver Issues
Edited By
Sofia Zhang
A group of tech enthusiasts is raising flags over GPU detection problems on their systems. Users report that despite owning multiple high-end NVIDIA graphics cards, only one GPU is fully recognized across various applications, igniting confusion and frustration.
Currently, an owner of a Mega Z790 ACE motherboard with dual GPUs — an RTX 4080 Ti and a RTX 5060 Ti — claims that the NVIDIA SMI command line only identifies the 4080 Ti as active. Despite the 5060 Ti showing up in device manager and NVIDIA's control panel, it remains inactive in software such as Blender and ComfyUI. "Only CUDA 0 is active in the multi-GPU node," the user stated, highlighting the persistent limitations.
The Issue at Hand
Users are perplexed: why do only certain GPUs function across software? The challenges seem to revolve around driver updates, with suggestions flowing from community forums.
Driver Update Needed: One comment notes the necessity to update to support CUDA 12.8 or higher, specifically for Blackwell GPUs like the 5060 Ti. This could be critical for resolving the detection issue.
"You may need to update your drivers to support CUDA 12.8 or higher,” a community member advised, pointing to potential software limitations.
Community Sentiment
The feedback on forums highlights a mixed sentiment. Some users are dissatisfied with the performance of their dual-GPU setups, while others remain hopeful that forthcoming driver updates will resolve these problems.
Key Highlights:
🔧 Driver Updates Required: Many emphasize updating to CUDA 12.8 for proper GPU functionality.
🎮 Software Limitations: Users express frustration over limited compatibility in popular programs like Blender and ComfyUI.
📈 Community Support: Users are pooling knowledge, sharing insights and troubleshooting methods.
What’s Next?
As the situation unfolds, it raises questions about NVIDIA's drive support for newer GPUs. Will the company act quickly to address these issues? Users await updated drivers, hoping to unlock the full potential of their systems.
In tech circles, this situation resonates with users experiencing similar frustrations. The varying levels of support and functionality across systems continue to be a divisive topic. Advocates for better support are growing, pushing for more robust driver solutions.
Stay tuned as developments arise, and users seek effective resolutions to harness their hardware's capabilities.
A Forecast for the Future of GPU Support
There’s a strong chance that NVIDIA will quickly roll out updated drivers to resolve the detection issues seen with certain high-end GPUs. Experts estimate around 70% probability that within the next month, we could see a patch that addresses the CUDA 12.8 requirements mentioned by users. This urgency stems from both the mounting complaints and the competitive pressure from rival card manufacturers. If NVIDIA fails to act timely, they risk losing customers to alternatives that fully optimize multi-GPU setups.
Echoes from the Past: The Windows Vista Launch
In a twist reminiscent of the Windows Vista launch in 2007, where the OS faced backlash for compatibility issues with existing hardware, we see a similar plight unfolding. Just as users grappled with outdated drivers and a lack of support, today’s GPU enthusiasts find themselves in a similar quagmire. The complexities of adapting cutting-edge technology often clash with legacy support, revealing a cycle where innovation can momentarily outpace the infrastructure designed to support it. This parallel serves as a reminder that even giants of technology can stumble in the race to stay ahead.