The latest information on NVIDIA's GeForce RTX 5090 & RTX 5080 GPUs suggests that the designs will be finalized this month.
NVIDIA GeForce RTX 5090 & RTX 5080 To Feature "D" Variants For China, Designs Being Finalized This Month With Higher TGPs
Yesterday, it was reported that NVIDIA's next-gen GeForce RTX 50 "Blackwell" GPUs could feature power ratings as high as 600W, and there has also been speculation regarding a potential delay. However, Benchlife has different information, and according to them, the designs will be finalized this month.
There was indeed a 400W NVIDIA cooling module case earlier, but it is currently in the cancellation or suspension stage; as for the 600W cooling module case, from the beginning to Now, it has not stopped, and the module manufacturer has also confirmed that NVIDIA currently has 5 GeForce RTX graphics card cases in progress.
Perhaps, 400W and 600W are the maximum heat dissipation capabilities of the radiator. In fact, the TGP of GeForce RTX 5090 and GeForce RTX 5080 are 550W and 350W respectively.
If there are no surprises, the Blackwell GPU architecture GeForce RTX 5090/D and GeForce RTX 5080/D are scheduled to be officially launched in September. If it is as our informant told us, there will soon be more information on the Internet for reference.
The tech outlet reports that the GeForce RTX 5090 and GeForce RTX 5080 are planned for official design spec finalization in September. In addition to that, it is also reported that NVIDIA will prepare "D" variants for both cards, the GeForce RTX 5090D and GeForce RTX 5080D.
Previously, NVIDIA only released the RTX 4090D as a China-exclusive variant which complied with export regulations. This time, both cards getting the "D" treatment means that there is going to be a significant performance bump over the previous generation.
On the power consumption side, it is reported that the TGPs of the GeForce RTX 5090 & RTX 5080 are rated at 550W and 350W, respectively. The previous 600W and 400W figures seem to align with the maximum heat dissipation figure of the new coolers featured on this card.
- RTX 5090 -> 550W (RTX 4090 @ 450W) = +22% Increase
- RTX 5080 -> 350W (RTX 4080 @ 320W) = +9.3% Increase
If that's the case, then the NVIDIA GeForce RTX 5090 would have a 100W higher TGP than the RTX 4090 while the RTX 5080 will have a 30W higher TGP than the RTX 4080. That seems a bit more reasonable than the numbers reported earlier but still, the TGP shouldn't reflect the actual power consumption of these cards which should be lower in actual gaming scenarios and only reach the TGP limits in certain applications.
What Benchlife said is that 5080/D and 5090/D are scheduled to officially finalize the design in September, not officially launch in September. https://t.co/UIdvq5uuhp
— MEGAsizeGPU (@Zed__Wang) September 5, 2024
Furthermore, it is reported by Expreview that the NVIDIA GeForce RTX 5090D will be available in China at the end of January after a formal announcement at CES 2025. Meanwhile, NVIDIA is also planning to EOL the GeForce RTX 4090D which should enter the respective discontinuation phase by December 2024.
If the designs are already being finalized this month, then we should get something in the coming months such as a teaser for a special GeForce event. If not, well then we could stick with the 2025 rumor.
NVIDIA GeForce RTX 50 "Blackwell" GPU Configurations:
| GPU Name | GPCs | TPCs | SMs | Cores (Assuming 128 per SM) | Memory Configuration | GPU SKU (Top) |
|---|---|---|---|---|---|---|
| GB202 | 12 | 96 (8 Per GPC) | 192 (2 Per TPC) | 24576 | 512-bit GDDR7 | GeForce RTX 5090 |
| GB203 | 7 | 42 (6 Per GPC) | 84 (2 Per TPC) | 10752 | 256-bit GDDR7 | GeForce RTX 5080 |
| GB205 | 5 | 25 (5 Per GPC) | 50 (2 Per TPC) | 6400 | 192-bit GDDR7 | GeForce RTX 5070 |
| GB206 | 3 | 18 (6 Per GPC) | 36 (2 Per TPC) | 4608 | 128-bit GDDR7 | GeForce RTX 5060 Ti |
| GB207 | 2 | 10 (5 Per GPC) | 20 (2 Per TPC) | 2560 | 128-bit GDDR6 | GeForce RTX 5060 |
| AD102 | 12 | 72 (6 Per GPC) | 144 (2 Per TPC) | 18432 | 384-bit GDDR6X | GeForce RTX 4090 |
| AD103 | 7 | 40 (6 Per GPC) | 80 (2 Per TPC) | 10240 | 256-bit GDDR6X | GeForce RTX 4080 SUPER |
| AD104 | 5 | 30 (6 Per GPC) | 60 (2 Per TPC) | 7680 | 192-bit GDDR6X | GeForce RTX 4070 SUPER |
| AD106 | 3 | 18 (6 Per GPC) | 36 (2 Per TPC) | 4608 | 128-bit GDDR6 | GeForce RTX 4060 Ti |
| AD107 | 3 | 12 (4 Per GPC) | 24 (2 Per TPC) | 3072 | 128-bit GDDR6 | GeForce RTX 4060 |
About the author: A Software Engineer by training and a PC enthusiast by passion, Hassan Mujtaba serves as Wccftech's Senior Editor for hardware section. With years of experience in the industry, he specializes in deep-dive technical analysis of next-generation CPU and GPU architectures, motherboards, and cooling solutions. His work involves not only breaking news on upcoming technologies but also extensive hands-on reviews and benchmarking.
Follow Wccftech on Google to get more of our news coverage in your feeds.
