That’s right, these new GPUs “generate” frames and insert the generated frame between the real frames, thus effectively increasing the frame rate. However, I think one of the main highlights of these new RTX 40 series GPUs is DLSS 3, or AI frame generation technology. The RTX 4080 is based on the new and better Ada Lovelace architecture, while the RTX 3080 Ti uses Ampere architecture. And yes, it doesn’t mean it will perform slower than the RTX 3080 Ti due to its lower core count. However, when you look at the RTX 3080 Ti’s specs, the RTX 4080 actually has fewer CUDA cores, Tensor Cores, and RT Cores than the RTX 3080 Ti. When we compare the RTX 4080’s specs to the RTX 3080, it looks a bit impressive, considering the core count increase. Finally, it has 16GB of GDDR6X memory, with an effective speed of 22.4Gbps, running on a 256-bit memory interface. It also comes with 304 4th gen Tensor Cores offering 780 Tensor-TFLOPS (with Sparsity) for AI processing and DLSS paired with 76 3rd gen Ada RT Cores capable of 113 RT-TFLOPS for powering next-generation ray-traced graphics. Since some of its SMs are disabled, there’s still room for an “RTX 4080 Ti”.Īccording to NVIDIA, the GeForce RTX 4080 has 9,728 CUDA cores, providing 49 Shader-TFLOPS of power for traditional rasterized graphics. The AD103 is an NVIDIA custom chip based on TSMC’s 4N process. Powering the RTX 4080 is an AD103 GPU with some of its SMs disabled. The new GeForce RTX 4080 is based on NVIDIA’s new Ada Lovelace architecture. GeForce RTX 4080 or RTX 3080/RTX 3080 Ti – Ada Lovelace or Ampere?
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |