Nvidia preheat RTX 40 series graphics card to confirm Ada architecture
CTOnews.com Sept. 15, according to Videocardz, today, Nvidia released a warm-up message, showing the "Bernoulli Digital engine Computing Chart", which was released by Ada Lovelace in 1843, which indirectly confirmed that RTX 40 will adopt the Ada architecture.
In addition, a post-it note appears in Nvidia's warm-up message, the number shown may refer to the specification of AD102 GPU-the chip size of 629 mm ², the number of transistors, 75.38 billion may represent its performance 2.08 times that of the previous generation.
CTOnews.com has learned that Nvidia has now announced a special live broadcast of GeForce Beyond on September 20, which is expected to introduce information about the next generation of GPU.
The following is the latest revelation of the RTX 40 graphics card:
RTX 4090
Equipped with AD102-300 GPU, enable 16384 CUDA cores, GPU frequency up to 2520 MHz, equipped with 24GB 21 Gbps GDDR6X memory, up to 1 TB / s bandwidth, 450W default TGP, maximum configurable TGP is 660W.
RTX 4080 16GB
Equipped with AD103-300 GPU, enable 9728 CUDA cores, GPU frequency up to 2505 MHz, equipped with 16GB 23 Gbps GDDR6X memory, TGP is 340W, can be configured to 516W.
RTX 4080 12GB
Equipped with AD104-400 GPU, enable 7680 CUDA cores, GPU frequency up to 2610 MHz, equipped with 12GB 21 Gbps GDDR6X memory, TGP 285W, can be configured to 366W.
