NVIDIA’s next flagship graphics card will feature a boost clock that’s nearly 50% higher than the standard version of its current halo product for desktop gamers, according to the latest claims shared by leaker kopite7kimi, who tweeted some of the alleged specifications for the GeForce RTX 4090 and two of its lower-ended siblings this morning. The GeForce RTX 4090 is said to feature a base clock of 2,235 MHz and boost clock of 2,520 MHz, alluding to improvements of 60% (vs. 1,395 MHz) and 49% (vs. 1,695 MHz) versus its predecessor in the GPU frequency category, respectively. Coupled with an alleged CUDA core count of 16,384, the GeForce RTX 4090 is speculated to offer up to 90 TFLOPS of single-precision performance, a value that’s around twice as high as the GeForce RTX 3090. NVIDIA is expected to launch its GeForce RTX 4090 graphics card in October, followed by the GeForce RTX 4080 in November and GeForce RTX 4070 in December.
Continue to update.— kopite7kimi (@kopite7kimi) July 4, 2022
RTX 4090, AD102-300-A1, 16384FP32, 384bit 21Gbps 24G GDDR6X, 450W, base 2235 boost 2520 actual max >2750;
RTX 4080, AD103-300-A1, 10240FP32, 256bit 21Gbps 16G GDDR6X, 420W,
RTX 4070, AD104-275-Kx(x is a number)-A1, 7168FP32, 160bit 18Gbps 10G GDDR6, 300W.
Kopite also reiterates the previously shared specs on RTX 4080 and RTX 4070 SKUs. Each SKU is to feature a different GPU, either AD103 (RTX 4080) or AD104 (RTX 4070). It looks like RTX 4080 will feature a 256-bit memory bus and 16GB of GDDR6X memory clocked at the same speed as RTX 4090. Meanwhile, RTX 4070 which supposedly has 7168 CUDA cores according to the latest rumors, would have 160-bit wide memory bus and 10GB of GDDR6 18 Gbps memory.
NVIDIA is now expected to unveil its RTX 40 series around September or October this year. Initially only these three models are to be launched with mid-range RTX 4060 coming next year around CES 2023.