According to a well-known leaker, more details on the technical specifications of NVIDIA GeForce RTX 4090, GeForce RTX 4080, and GeForce RTX 4070 graphics cards have been leaked on the web. The new reports concern mainly the AD102 system clocks. Check out the rumored specs of the new NVIDIA GeForce RTX 4000 Series video cards below.
NVIDIA GeForce RTX 4090, GeForce RTX 4080 & GeForce RTX 4070 Specifications Leaked
For as long as anyone can remember, there has been hardly any information about the specifications of NVIDIA GeForce RTX 4000 series graphics cards, which will debut in a few months. Now, a new leak from the user kopite7kimi indicates the latest news of the RTX 4090/4080/4070 technical specifications has added some details compared to the last news for your reference.
The new data from insider kopite7kimi indicates that the NVIDIA GeForce RTX 4090 has a base frequency of 2235 MHz and a boost of 2520 MHz with an actual gaming clock of 2750 MHz.
The GeForce RTX 4090 is supposed to have a CUDA 16384 core configuration, so it simply means that at the stated clock speed of 2750 MHz, the GPU’s compute performance can reach 90 TFLOPS under a single-precision workload. This is almost exactly twice as high as the GeForce RTX 3090.
The insider also talks about the past specifications of the GeForce RTX 4080 and RTX 4070 – each model should have its own GPU, it’s either AD103 (RTX 4080) or AD104 (RTX 4070).
So, according to reports, it seems that the RTX 4080 will feature a 256-bit memory bus and 16GB of GDDR6X memory clocked at the same speed as the RTX 4090. On the other hand, the RTX 4070, which according to the latest rumors allegedly has 7168 CUDA cores, will have a 160-bit bus memory and 10GB GDDR6 18Gb/s memory.
It is most likely that NVIDIA will introduce its GeForce RTX 4000 series around September or October of this year. Initially, only three of these models are planned to be released, with an intermediate GeForce RTX 4060 coming next year, with an announcement at CES 2023.
Continue to update.
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.
— kopite7kimi (@kopite7kimi) July 4, 2022