In terms of performance, based on raw specs, the Titan RTX has 6 percent more CUDA cores, RT cores, and Tensor cores than the RTX 2080 Ti.
Nvidia has announced today the most powerful GPU the company has manufactured so far, and possible the most powerful GPU yet produced. The card features almost the same raytracing performance as the Ti, but increased deep-learning muscle with 576 multi-precision Tensor Cores resulting in a hefty 130 TFLOPS of deep learning performance. TITAN RTX - dubbed "T-Rex" by NVIDIA - delivers ray-traced and next-gen virtual reality (VR) graphics as well as high-performance computing.
These properties are rated to deliver up to 130 teraflops (TFLOPs) in deep-learning power and 11 GigaRays in terms of ray-tracing performance.
"The introduction of T-Rex puts Turing within reach of millions of the most demanding PC users - developers, scientists and content creators".
But here's the kicker, the new Titan is paired with 24GB GDDR6 memory at 672GB/s of bandwidth, significantly faster than the previous Titans. Of particular interest for data scientists will be the fact that the Titan RTX supports RAPIDS, the open-source libraries that integrate with popular data science workflows to boost machine learning use cases. Pairing two Titan RTX cards via NVLink will scale both memory and compute performance. The fully-armed TU102 GPU and high core clock make the Titan RTX the fastest single gaming card around for those who can stomach its price, but gamers aren't really this card's raison d'être.