Nvidia has announced the next-gen RTX 40 Series GPUs. After several rumors, they announced their latest products during their GeForce Beyond event. RTX 4080 and RTX 4090 are now official. Moreover, both are powered by the latest Ada Lovelace architecture. All RTX 40 series GPUs will have Nvidia ShadowPlay support to capture gameplay at up to 8K at 60 frames per second.
Ada Lovelace Architecture
The Ada Lovelace architecture is named after an English Mathematician. Adding on, it is the third generation that powers both RTX 4080 and RTX 4090. The Ada Lovelace architecture is based on a new 4nm architecture by TSMC. Moreover, the major areas of improvement are across its shader cores, RT cores, and Tensor cores. It also brings major leaps in ray-tracing and DLSS 3 (Deep Learning Super Sampling).
This is the third generation of Nvidia’s Deep Learning Super Sampling technology. DLSS 3 has major improvements when it comes to further increasing frame rates. It overcomes the CPU limitations and allows the GPU to render entire frames.
Nvidia Reflex is also integrated through DLSS 3. It allows the GPU and CPU to work together and reduces the latency overall of the system. In addition, it will double the system responsiveness. Moreover, the new Ada Optical Flow Accelerator allows movement prediction through DLSS 3. Deep Learning Super Sampling technology will allow you to achieve 4 times the performance in games in comparison to traditional rendering techniques.
RTX 4090 & 4080 Specifications
The RTX 4090 comes in only one variant as opposed to the two 4080 models. It will pack 24GB of GDDR6X memory. It is claimed to be 2 to 4 times faster than the RTX 3090 Ti. Moreover, it will consume the same power as the previous generation. Furthermore, a power supply of at least 850W based on a PC with a Ryzen 5900X processor will be needed. RTX 4090 has 16,384 CUDA Cores, a base clock of 2.23GHz that boosts up to 2.52GHz, 1,321 Tensor-TFLOPs, 191 RT-TFLOPs, and 83 Shader-TFLOPs.
RTX 4080 comes in two models – one with 12GB GDDR6X RAM and the other has 16GB of GDDR6X RAM. Nvidia claims that it is also 2-4 times faster than the previous generation RTX 3080 Ti. The 12GB model includes 7,680 CUDA Cores, 7,680 CUDA Cores, a 2.31GHz base clock that boosts up to 2.61GHz, 639 Tensor-TFLOPs, 92 RT-TFLOPs, and 40 Shader-TFLOPs. While the 16GB model includes 9,728 CUDA Cores, a base clock of 2.21GHz that boosts up to 2.51GHz, 780 Tensor-TFLOPs, 113 RT-TFLOPs, and 49 Shader-TFLOPs of power.
Pricing & Availability
RTX 4090 will be available on the 12th of October and will be priced at $1599. On the other hand, the RTX 4080 12GB variant will be priced at $899 and the 16GB will cost you $1199. Both will be available by November.