NVIDIA’s next-gen data center GPU (accelerator) will draw over 1,000W of power. This was revealed by well-respected tipster @kopitekimi the other day. Codenamed “Hopper” after the American computer scientist Grace Hopper, the GH180 is expected to feature 3D stacking (TSMC’s CoWois packaging technology). It’ll be a heterogeneous chip featuring a combination of Grace (GH200) and Hopper (TH500). What exactly is the difference between Grace and Hooper remains unknown. Either way, Hopper should feature two identical chiplets, each featuring over 10,000 CUDA cores and around 128GB of HBM2e/HBM3 memory.
NVIDIA RTX 4080/4090 Specifications
If the AD102 includes a total of 18,432 cores, we can expect roughly 16,000 cores on the RTX 4080 and 18,000 on the RTX 4090. Considering that we’re likely to see the AD102 fabbed on Samsung’s 5nm/6nm node, the increase in transistor density should absorb the otherwise expected increase in die size.
With the next generation of GeForce graphics cards, we’re not only looking at phenomenal performance targets but also considerable increases to the TDP. According to Greymon and Kopitekimi, the Lovelace-based RTX 4080/4090 will draw as much as 500W of power under load. This is despite the use of one of the most advanced and efficient process nodes on the planet. However, running the numbers kind of adds up. The AD102 flagship is expected to feature 144 SMs/12 GPCs, a gain of 71% in terms of logic compared to the GA102. Even if TSMC’s N5 node is 30% more power-efficient than Samsung’s 8nm LPP node, we’re looking at an increase of at least 80% in hardware units. That should easily result in a power draw at least 30-50% more than the top-end (RTX 3080/3090) Ampere offerings.
According to a report from Igor, this will be facilitated by the new 12VHPWR power connector. As part of the next-gen PCIe Gen 5 standard, the PCI SIG has defined a new “High Power Connector” or H+ for graphics cards with a power requirement of up to 600W. The 12VHPWR connector is a major departure from the existing PCI Express 2×3/2×4 Auxiliary Power connectors. While the pins of the former have a spacing of 3.0 mm, the latter has a larger spacing of 4.2 mm. Therefore, they won’t be compatible with the same hardware and require changes on both the PSU and the GPU PCB.