Gaming GeForce RTX 4090 600W TBP and Workstation RTX L6000 375W TBP

NVIDIA’s Ada Lovelace GPUs, such as the flagship AD102 for GeForce RTX 4090 and RTX L6000 graphics cards, will be among the most power-hungry graphics cards ever. Rumors have already mentioned the insane power consumption of the next-gen lineup, but Moore’s Law is Dead reiterated, and that’s indeed the case in his latest video.

Gamers and workstation users get ready! Next-generation NVIDIA AD102 GPUs powering GeForce RTX 4090 and RTX L6000 consume up to 600W TBP in reference design

From what we know so far, the NVIDIA AD102 GPU will be a powerful monster, with a reported TBP of 850W. But as NVIDIA tries to optimize it, the TBP numbers will eventually drop. Therefore, the latest TBP figure for NVIDIA’s flagship graphics card is set at 600W (total board power). Two cards will come in flagship configurations, the gaming-focused GeForce RTX 4090 and the workstation-focused RTX L6000. These names have not been confirmed, but they will be the successors to the RTX 4090 and RTX A6000 based on the GA102 Ampere core.

NVIDIA reportedly begins testing its fastest next-generation GPU, the AD102, for GeForce RTX 4090 graphics cards with 24 Gbps GDDR6X memory

NVIDIA GeForce RTX 4090 ‘AD102 GPU’ Gaming Graphics Card, 600W TBP

According to Tom, the NVIDIA GeForce RTX 4090 has a TBP of 600W at reference, which will go up with the non-reference design. As we’ve seen with the GeForce RTX 3090 Ti, some models have TBP in excess of 450W (reference) and up to 516W (custom). The same goes for cards based on Ada Lovelace. While this requires a lot of cooling, it looks like the AIB is ready. Once the 600W TBP limit is exceeded, we may see more hybrid style GPU coolers being used instead of traditional air coolers. This would mark a 33% increase in TBP over the RTX 3090 Ti.

As for the graphics cards themselves, NVIDIA seems to want to keep the gaming crown on its own, and the GeForce RTX 4090 could be the first RTX 40-series graphics card to launch as early as Q3 2022. With AMD focused on launching Navi 33 first, it looks like NVIDIA will remain on the Game of Thrones until Navi 31 launches, and due to its MCM design, this is expected to deal a serious blow to the green team in the efficiency department.

NVIDIA GeForce and AMD Radeon graphics cards are now 20% below MSRP, good for gamers, but uncertainty hangs over the crypto space

NVIDIA CUDA GPU (rumored) Preliminary:

NVIDIA GeForce RTX L6000 ‘AD102 GPU’ Workstation Graphics, 375W TBP

The NVIDIA workstation lineup will also be updated with the new AD102 GPU cores, while the top RTX L6000 or RTX L8000 will use the top-of-the-line AD102 GPU. It is expected to have a TBP of 320 and a maximum of 375W. According to MLID, the current prototype is being tested using the EPS 8-pin connector, although we will likely see a final variant using the new PCIe Gen 5 (16-pin) connector.

Due to the card’s much lower TBP design compared to the RTX 4090, the dual-slot and blower-style cooler can still sustain it. But that’s still a massive 25% increase in power consumption for the new graphics card.

Having a flagship GeForce RTX 4090 with a 600W TBP means NVIDIA is likely to make the GeForce RTX 4080 run at a higher TBP as well. Previous claims pointed to a 450W TBP, the same as the current RTX 3090 Ti graphics card.

Additionally, this power-hungry graphics card will make 1000 and 1200W power supplies the new standard, which is where the new ATX 3.0 platform comes into play. The new ATX 3.0 standard will help deliver sustained power to next-generation PCIe Gen 5 cards, which are expected to feature “Power Excursions” of up to 1800W. So, if you want to try out the upcoming graphics card, a 1000-1200W power supply will be your best option.

NVIDIA Ada Lovelace and Ampere GPU Comparison