NVIDIA Quadro RTX 4000

NVIDIA Quadro RTX 4000. The NVIDIA Quadro RTX 4000 delivers GPU-accelerated ray tracing, deep learning, and advanced shading in an accessible single-slot form factor.

The NVIDIA Quadro RTX 4000 features 36 RT cores to accelerate ray tracing, 288 Tensor cores to accelerate artificial intelligence, and 8 GB of GDDR6 memory to accommodate large datasets.

It’s equipped with three DisplayPort 1.4 connectors with HDR support and features the latest Virtual Link connector (which can also drive a DisplayPort panel via an included USB Type-C to DP adapter) to simplify connectivity with next-generation VR hands. Plus, with Quadro Sync II, you can deploy massive digital display solutions at higher density and lower cost.

Turing GPU Architecture

Based on a state-of-the-art 12nm FFN (NVIDIA FinFET) high-performance manufacturing process, customized for NVIDIA to incorporate 2304 CUDA cores, the Quadro RTX 4000 GPU is the most powerful computing platform for HPC, AI, VR, and graphics workloads on professional desktops in a single slot. The Turing GPU architecture enables the biggest leap in real-time computer graphics rendering since NVIDIA invented programmable shaders in 2001. It includes 13.6 billion transistors on a 545 mm2 die.

RT cores

New dedicated hardware-based ray-tracing technology enables the GPU for the first time to render cinematic-quality, photorealistic objects, and environments in real-time with physically accurate shadows, reflections, and refractions. The real-time ray-tracing engine works with NVIDIA OptiX, Microsoft DXR, and Vulkan APIs to deliver a level of realism far beyond what is possible with traditional rendering techniques. RT cores accelerate Bounding Volume Hierarchy (BVH) traversal and ray tracing functions using a reduced number of rays cast through a pixel.

Improved Tensor Cores

New mixed-precision cores designed specifically for deep learning matrix arithmetic, providing 8 times more TFLOPS for training compared to the previous generation. The NVIDIA Quadro RTX 4000 uses 288 Tensor Cores; each Tensor Core performs 64 floating-point FMA (fused multiply-add) operations per clock, and each SM performs a total of 1024 individual floating-point operations per clock. In addition, to support FP16/FP32 matrix operations, the new Tensor Cores have added precision INT8 (2048 integer operations per clock) and experimental INT4 and INT1 (binary) modes for matrix operations.

Advanced shading technologies

Mesh shading: Computer-based geometry pipeline to speed up geometry processing and removal on geometrically complex models and scenes. Mesh shading provides up to 2x performance improvement for geometry-related workloads. Variable Rate Shading (VRS): Gain rendering efficiency by varying the shading rate based on scene content, gaze direction, and motion. Variable rate shading provides similar image quality with a 50% reduction in shaded pixels. Texture Space Shading: Shading of object/texture space to improve the performance of work tasks that require a pixel shader, such as depth of field and motion blur. Texture space shading provides higher throughput with increased fidelity by reusing pre-shadowed texels for VR workloads that require a large pixel-shader volume.