NVIDIA’s China-specific GeForce RTX 4090D is set to launch in the coming weeks. As we get closer to the release, more and more details about the GPU are coming out. The latest bits from the rumor mill claim that the RTX 4090D will be un-overclockable. Additionally, the graphics card will allegedly have a lower TBP of 425W (down from 450W on the 4090) to accommodate the lower shader count.
Even though the headroom has decreased significantly, completely locking overclocking on consumer GPUs is rare. This can be done on a firmware or hardware level, and we usually see the former in action. The reduction in TBP is a secondary measure to ensure that the RTX 4090D performs worse than the stock 4090. Once the BIOS encryption is broken, removing these limits tends to be straightforward.
The GeForce RTX 4090D will leverage the AD102-250 GPU with a slightly lower core count than the 4090, but likely the same memory configuration. The GPU clocks are the same as the stock variant, topping at 2,520MHz.