It’s finally here, the moment thousands have been waiting for. Nvidia today unleashed its latest generation of flagship graphics cards with its Ada Lovelace architecture, the GeForce RTX 4090 and 4080, respectively.
The RTX 40 Series GPUs feature a plethora of innovations from the last generation:
- Streaming multiprocessors with up to 83 teraflops of shader power, roughly 2x more than the previous RTX 30 generation.
- Third-generation Ray-Tracing Cores with up to 191 effective ray tracing teraflops, 2.8x over the previous generation.
- Fourth-generation Tensor Cores with up to 1.32 Tensor petaflops – a whopping 5x over the previous generation using FP8 accelerations.
- Shader Execution Reordering (SER) that improves execution efficiency by rescheduling shading workloads on the fly to better utilize the GPU’s resources. As significant an innovation as out-of-order execution was for CPUs, SER improves ray-tracing performance up to 3x and in-game frame rates by up to 25%.
- Ada Optical Flow Accelerator with 2x faster performance allows DLSS 3 to predict movement in a scene, enabling the neural network to boost frame rates while maintaining image quality.
- Architectural improvements tightly coupled with custom TSMC 4N process technology results in an up to 2x leap in power efficiency.
- Dual NVIDIA Encoders (NVENC) cut export times by up to half and feature AV1 support. The NVENC AV1 encode is being adopted by OBS, Blackmagic Design DaVinci Resolve, Discord and more.
The most interesting feature we feel is the new third-generation RT cores. They deliver 2x faster ray-tracing intersection testing and include two new hardware units. An Opacity Micromap Engine that speeds up ray tracing of alpha-test geometry, and Micro-Mesh Engine that generates micro-meshes no the fly to generate additional geometric complexity without the performance and storage cost of such geometries.
GeForce RTX 4090 and 4080: The New Flagships
The RTX 4090 features 76 billion transistors and 16,384 CUDA cores. The RTX 3090 has 10,496 CUDA cores. This means the RTX 4090 has about 50% more CUDA cores than the RTX 3090. It achieves this while maintaining the same 450W power consumption. Like its predecessors, the RTX 4090, will feature 24GB of GDDR6X memory.
The GeForce RTX 4080 replaces the RTX 3080. It will launch in two configurations, 16GB and 12GB. The 16GB variant has 9,728 CUDA cores and 16GB of GDDR6X memory. The RTX 4080 12GB has 7,680 CUDA cores. This is an upgrade from the RTX 3080, which came in 12GB, 8960 CUDA Cores and 10GB, 8704 configuration.
Nvidia has significantly increased the CUDA cores in the 16GB variant of the RTX 4080 compared to the RTX 3080 12GB, by about 10%.
GeForce RTX 40 for Video Editing and Live Streaming with Nvidia Studio
The RTX 40 series will have new dual, eighth generation AV1 encoders that give improved performance for tasks such as video editing and live streaming. 3D artists can now render fully ray-traced environments with accurate physics and realistic materials, and view the changes in real time, without proxies.
Nvidia Broadcast, Nvidia’s streaming software development kit has got three updates, which include Face Expression Estimation, Eye Contact and quality improvements to Virtual Background.
GeForce RTX 4090 and RTX 4080 Pricing
The RTX 4090 is priced starting at $1,599. The RTX 4080 16GB is priced at $1,199 and the 12GB variant is priced at $899 respectively.
Watch out for variants from ASUS, Colorful, Gainward, Galax, GIGABYTE, Innovision 3D, MSI, Palit, PNY and Zotac.
Along with the unveiling of these graphics cards, Nvidia showcased the latest iteration of its DLSS technology, DLSS 3.0, read more about it here.
