The GeForce RTX 5090, a next-generation flagship gaming GPU from NVIDIA powered by the company’s new Blackwell architecture, will feature over 8,000 more CUDA Cores than the GeForce RTX 4090, according to a list of alleged specifications that has been shared online. The news comes on the heels of a separate report regarding some of the GeForce RTX 50 Series’ cooling modules, which suggests TDPs of anywhere from 250 to 600 watts for the lineup and how AMD’s Radeon RX 8000 Series with RDNA 4 GPUs will not arrive until 2025 at the earliest.
NVIDIA GeForce RTX 5090 vs. GeForce RTX 4090 specs (rumored):
| Spec | RTX 4090 | RTX 5090 |
| Streaming Multiprocessors | 128 | 192 |
| CUDA Cores | 16,384 | 24,576 |
| Ray Tracing Cores | 128 | 192 |
| Tensor Cores | 512 | 768 |
| Boost Clock | 2.52 GHz | 2.9 GHz |
| L2 Cache | 72MB | 128MB |
| Memory Bandwidth | 1,008 GB/s | 1,532 GB/s |
The power demands of the GeForce RTX 50 Series, per Benchlife:
- “…according to information from the cooling module factory that we are familiar with, NVIDIA has been conducting relevant testing and verification of cooling modules for the GeForce RTX 50 series graphics cards based on the Blackwell GPU architecture, and has obviously started preparations for the GeForce RTX 50.”
- “Although there is no clear time point, there are currently about 4 plans in progress, with the highest wattage being 600W and the lowest wattage being 250W.”
- “Whether there is a chance to see NVIDIA launch GeForce RTX 50 series graphics cards in 2024 is still difficult to say clearly at this stage, but it is certain that AMD will not launch Radeon RX 8000 series graphics cards with RDNA 4 GPU architecture in 2024.”
Dextero wrote in its coverage:
…the RTX 5090 will end up being around 70% faster than the RTX 4090. Though, it’s unclear how, or where this figure might be replicated. It’s likely that the leaker is talking about pure, rasterized performance, with no DLSS bells and whistles to boost the performance up.