NVIDIA GeForce RTX 4090 gets 24 GB GDDR6X memory at 21 Gbps and 600 W TDP, RTX 4070 gets 12 GB GDDR6 memory at 18 Gbps and 300 W TDP

NVIDIA’s upcoming GeForce RTX 4090 and GeForce RTX 4070 graphics cards based on the Ada Lovelace GPU architecture have been further elaborated by Kopite7kimi in his latest tweet. These graphics cards will rock some seriously insane specs and performance compared to current Ampere variants.

NVIDIA rumored to equip GeForce RTX 4090 with 24 GB / 21 Gbps memory at 600 W, RTX 4070 with 12 GB / 18 Gbps memory at 300 W

In its tweet, Kopite7kimi focuses on two NVIDIA ‘Ada Lovelace’ graphics cards, the GeForce RTX 4090 and the GeForce RTX 4070. The leaker has already revealed the alleged specs of other cards, such as the RTX 4090 Ti and the RTX 4080 but it seems he was able to find more information about the two cards mentioned above. The main difference is that the GeForce RTX 4090 will be based on an optimized AD102 GPU core, while the RTX 4070 will use an optimized AD104 GPU core.

NVIDIA & AMD GPU price update for May 2022: GeForce graphics cards now 14% above MSRP, Radeon only 6% above MSRP

NVIDIA Ada Lovelace ‘GeForce RTX 40’ GPU Configurations

GPU Name GPCs / TPCs Texting per TPC / total CUDA cores L2 cache Memory bus
AD102 12 / 6 2 / 144 18432 96MB 384-bit
AD103 7 / 6 2 / 84 10752 64MB 256-bit
AD104 5 / 6 2 / 60 7680 48MB 192-bit
AD106 3 / 6 2 / 36 4608 32MB 128 bit
AD107 3/4 2 / 24 3072 32MB 128 bit

NVIDIA GeForce RTX 4090 / AD102 GPU but optimized

The NVIDIA GeForce RTX 4090 will also use the same AD102 GPU as the ‘Ti’ variant, but in a smaller configuration based on the rumor. The AD102 GPU will use 144 SMs, a 71% increase over the existing GA102 GPU and include a massive 18,432 CUDA cores. In addition to the number of SM, the Ada Lovelace GPUs will also have larger L2 cache sizes. Starting with the AD102 GPU, the flagship would come equipped with up to 96MB of L2 cache, an insane 16x more than the 6MB L2 cache on the GA102.

The graphics card will additionally rock 24 GB of GDDR6X memory clocked at 21 Gbps over a 384-bit bus interface and deliver a bandwidth of over 1.0 TB/s. Of the TGP numbers we’ve seenthe card has a TBP of 600W at reference and that goes up with the non-reference designs.

NVIDIA GeForce RTX 3090 Ti Custom ‘XOC’ BIOS increases power limit to nearly 1000W

As we saw with the GeForce RTX 3090 Ti, there are certain models that go beyond the 450W TBP (reference) and up to 516W (custom). The same will be the case with Ada Lovelace based maps. And while this requires a lot of refrigeration, it seems that AIBs are good and ready. We’ll probably see the use of more hybrid GPU coolers instead of traditional air coolers once we go above the 600W TBP limit. This represents a 33% increase in TBP over the RTX 3090 Ti.

NVIDIA GeForce RTX 4070 / AD104 at 300W

Finally, there are the NVIDIA GeForce RTX 4070 class GPUs that will use the AD104 GPUs. The AD104 GPUs will feature 60 SM units for a total of 7680 CUDA cores and a total of 48MB of L2 cache. The RTX 4070 series in particular will feature 12 GB of GDDR6 memory via a 192-bit bus interface and a TGP of 300W. Since these are GDDR6 chips, we’re likely to get speeds of 16-18 Gbps, which should give you a bandwidth of around 400 GB/s. This card would use the PG141-SKU341 PCB.

Now taking into account the 300W TDP, the GeForce RTX 3070 Ti already has a reference TDP of 290W, so you’re looking at a meager 10W increase, but at the same time the performance would be comparable to an RTX 3080 Ti-class graphics card that has a power of 350W. The 12 GB memory also sounds very appealing, as it is the same amount as on the RTX 3080 Ti. It all depends on the price of this card as the 70 class is one of the most popular among the high end graphics cards and for users it would be a huge upgrade to get the performance of the 3080 Ti class ($ 1199 US) from about $500-$600. .

The NVIDIA Ada Lovelace GPU family is expected to bring a generational jump similar to that of Maxwell to Pascal. It is expected to launch in the second half of 2022, but expect the offering and price to be similar to current cards despite NVIDIA Spending billions of dollars to get those good, good TSMC 5nm (4N) wafers

NVIDIA GeForce RTX 40 Series Graphics Card Lineup (Rumors):

Graphics Card GPU PCB variant: SM Units / Cores Memory / Bus Memory Clock / Bandwidth TGP Power connections
NVIDIA GeForce RTX 4090 Ti AD102-350? To be determined 144/18432? 48GB / 384bit 24 Gbps / 1.15 TB/s ~900W 2x 16-pin
NVIDIA GeForce RTX 4090 AD102-300? PG137/139 SKU330 >144 / 18432? 24 GB / 384 bits 21 Gbps / 1.00 TB/s ~600W 1x 16-pin
NVIDIA GeForce RTX 4080 AD103-300? To be determined >84 / 10752? 16GB / 256-bit 21 Gbps / 672 GB/s ~350W 1x 16-pin
NVIDIA GeForce RTX 4070 AD104-400? PG141-310 SKU341 >60 / 7680 12GB / 192-bit 18 Gbps / 432 GB/s ~300W 1x 16-pin

Which NVIDIA GeForce RTX 40 series graphics card are you most looking forward to?

Leave a Reply

Your email address will not be published.