Nvidia may have an even more monstrous GPU coming up

A new rumor suggests that Nvidia may be working on an even more power-hungry RTX 40-series graphics card than the flagship RTX 4090.

The company has reportedly built a test board with a total graphics power (TGP) of 900 watts, topping every previous expectation as to the card’s power requirements.

In fact, there is another full-fat AD102 SKU with 900W TGP, 48G 24Gbps GDDR6X, 2*16pin and higher frequency.
But no one knows whether it will become an actual product. Because the test board of AD102 has more than two 16pin connectors, so everything is possible.

— kopite7kimi (@kopite7kimi) April 27, 2022

Today’s round of rumors comes from kopite7kimi on Twitter, who is a well-known tipster in the PC hardware space, with a good track record when it comes to Nvidia predictions. This time, kopite7kimi talked about Nvidia allegedly preparing a graphics card that utilizes the full AD102 GPU.

The GPU, aside from the enormous 900-watt power draw, would have plenty of memory to match. Kopite7kimi pointed toward 48GB of GDDR6X memory with a 24Gbps bandwidth. The frequency would also be higher than the base RTX 4090, of course, but the leaker didn’t give any figures as to that. The GPU would also allegedly feature a dual 16-pin design.

These specifications put the rumored GPU far above what we’ve heard of the flagship RTX 4090 so far. Multiple sources now allude to the RTX 4090 having a 600-watt power requirement as well as 24GB memory. This strongly points to Nvidia making this 900-watt/48GB memory version a separate GPU. It could be a future RTX 4090 Ti, although that’s a pretty steep jump when compared to the current generation of GPUs.

After all, the gap between the RTX 3090 and the RTX 3090 Ti is nowhere near as big. In terms of performance, there is an increase in favor of the RTX 3090 Ti, but it comes at a much-increased power draw. If Nvidia isn’t turning the new monstrous GPU into an RTX 4090 Ti, it could be a TITAN graphics card instead — something Nvidia didn’t revisit in the current generation of GPUs.

RTX 4080 will use AD103 chips, build with 16G GDDR6X, have a similar TGP to GA102.
RTX 4070 will use AD104 chips, build with 12G GDDR6, 300W.
Both of them haven't started testing yet, but soon do them.

— kopite7kimi (@kopite7kimi) April 27, 2022

Aside from talking about the rumored RTX 4090 Ti, kopite7kimi confirmed some of the specifications of the RTX 4080 and RTX 4070. According to them, each model will feature a different GPU. The RTX 4080 would use the AD103 GPU with 16GB of GDDR6X memory and a similar TGP to the current-gen GA102. Meanwhile, the RTX 4070 would use the AD102 chip with 12GB of GDDR6 memory and a 300-watt power requirement. The GPUs haven’t entered the testing phase yet, but it’s likely to happen soon.

Nvidia’s next generation of graphics cards, dubbed Ada Lovelace, is set to release later this year. Until Nvidia officially confirms the specifications and models that it’s going to launch, everything is just speculation — but the little we do know seems pretty exciting.

Editors' Recommendations