That's not a big gap, especially considering the spec difference between the two RTX 6000 series GPUs. The RTX Pro 6000 comes with 96GB of GDDR7 operating on a 512-bit interface, spread over 32 chips with 3GB of capacity each. The GPU also comes with 24,064 CUDA cores split across 188 SMs.
The RTX 6000D comes with 14% less memory and memory bandwidth, along with 20% fewer CUDA cores. It has 84GB of GDDR7 memory connected to a 448-bit bus and 19,968 CUDA cores spread across 156 SMs.
Geekbench's OpenCL benchmark represents just one workload, so take these performance results with a pinch of salt. Shader compute performance usually takes the smallest hit on Nvidia's neutered China-exclusive GPUs. AI performance usually takes a much larger hit, and it's that workload that the USA's export controls care about. (These D-series GPUs are designed to comply with the U.S. government's China-specific export regulations on GPUs.)
The RTX 6000D's favorable performance against the RTX Pro 6000 could have been compelling for customers if the price was right, and the card was launched in the right context. The RTX 6000D couldn't have launched into a less favorable environment. Not only did the Chinese government ban the card months after launch —it also purportedly faced tough competition from Chinese GPU manufacturers .
Get Tom's Hardware's best news and in-depth reviews, straight to your inbox.
Key considerations
- Investor positioning can change fast
- Volatility remains possible near catalysts
- Macro rates and liquidity can dominate flows
Reference reading
- https://www.tomshardware.com/pc-components/gpus/SPONSORED_LINK_URL
- https://www.tomshardware.com/pc-components/gpus/nvidia-rtx-pro-6000d-squeaks-ahead-of-rtx-5090d-in-geekbench-opencl-china-tailored-ai-card-still-performs-well-despite-regulatory-woes#main
- https://www.tomshardware.com
- Taiwan's $3.2 billion plan for 'AI island' with data centers, quantum hubs, and AI robotics labs faces risks — power and geopolitical headwinds may threaten cou
- Intel's next-gen Nova Lake will finally tackle AMD's Ryzen X3D, but only with pricey 'K' models — 144MB Big Last Level Cache response to 3D V-Cache will only co
- AI On: 3 Ways to Bring Agentic AI to Computer Vision Applications
- TSMC says advanced‑node capacity falls ‘about three times short’ of AI demand — company's wafer capacity is 'still not enough' as demand continues to rise
- These are the best Black Friday deals on budget Wi-Fi 7 routers — dual-band as low as $81, tri-band from $170
Informational only. No financial advice. Do your own research.