Nvidia will launch its first few cards from the RTX-50-series next week at CES, including an RTX-5090, RTX-5080, RTX-5070 Ti, and RTX-5070. The 5090 will be of particular interest to performance-obsessed, money-is-no-object PC gaming fanatics since it’s the first new GPU in over two years that can beat the performance of 2022’s RTX 4090.
According to leakers, the 5090 will have a higher maximum power draw than the 4090 due to its improved performance and slower advances in chip manufacturing. VideoCardz reported that the thermal design power (TDP), which is the power consumption of the RTX 4090, will be increased to 575 W. The RTX 5080 will also see its TDP increase to 360 W from 320 W on the RTX 4080 Super.
The RTX-5090 is also close to the maximum power draw possible over a single 12-VHPWR plug, which can deliver up to 600 W (though, once you add the 75 W available through the PCI Express slot of your motherboard, the maximum power draw is a little higher at 675 W).
The fact that these cards have a higher peak power consumption does not mean they will always consume more power when gaming than their 40 series counterparts. Their performance could be so good that they still could be very efficient cards when it comes to performance per watt.
If you’re thinking about upgrading to an RTX-5090, and these specs are accurate then you may also need to upgrade your power supply. Nvidia recommends a minimum 850 W power supply to support the GPU while still leaving enough power for the rest the system. Nvidia’s recommendation for the minimum power supply for the 5090 is 1,000 W. This additional 125 W bump would suggest that Nvidia recommends a 1,000 W.
Nvidia’s CES keynote is scheduled to begin at 9:30 pm Eastern/6.30 pm Pacific on January 6, currently.