As NVIDIA was getting ready to release the GeForce RTX 50 Series of GPUs, codenamed “Blackwell,” a power supply manufacturer unintentionally revealed the power settings of every SKU. Users can configure their systems online and receive power supply recommendations by using Seasonic’s power supply wattage calculator. This implies that in order to handle the enormous diversity of components, the system frequently becomes overstocked with CPU/GPU SKUs. The next GeForce RTX 50 series, which includes the top-tier RTX 5090 GPU, is what we got this time. This SKU will launch with the GeForce RTX 5050 and is anticipated to have a 100 W TDP. The TDP of the RTX 5060, its larger sibling, is increased to 170 W, which is 55 W more than that of the RTX 4060, the “Ada Lovelace” model from before.

Located in the middle of the stack, the GeForce RTX 5070 boasts a 220 W TDP, which is 20 W more than the Ada generation. NVIDIA developed the GeForce RTX 5080 and RTX 5090, with 350 W and 500 W TDP, respectively, for higher-end SKUs. This also indicates a 30 W increase for RTX 5080 and a 50 W increase for RTX 5090 in TDP from the Ada generation. It’s interesting to note that NVIDIA intends to update the PCIe 6.0 CEM protocol and use a 16-pin 12V-2×6 connector to unify the power connection mechanism across the entire family this time. We find it fascinating that the “Blackwell” generation’s power consumption have increased across the SKUs, and we’re interested to see if the performance benefits will be sufficient to balance efficiency.