Nvidia recently announced their lineup of RTX series GPUs which are embedded with their latest innovation — Turing architecture. The new Nvidia Titan RTX GPU is an ultra-high-end graphics card that is mainly focused on AI researchers and data scientists.

The Titan series from Nvidia has always been a work-grade professional GPU. Its predecessor, the Titan V was the previous generation top-of-the-line model. The Titan RTX is a more decked out version of their recently announced GeForce RTX 2080 Ti GPU, which in itself is a beast.

According to Nvidia, the Titan RTX is the fastest PC graphics card ever built and they are not wrong. This GPU can handle anything that’s thrown at it with its massive 24GB GDDR6 memory and support for Nvidia’s much-boasted ray tracing and Deep Learning Supersampling (DLSS) algorithm.

Also read: What is Dedicated GPU? How is it different from Integrated GPU?

So why is the TITAN RTX the fastest PC GPU?

Well, The Titan RTX GPU is embedded with 4,608 programmable CUDA cores for 3D graphics, 576 Tensor cores for AI and deep learning, and 72 specially designed RT cores for real-time ray tracing. It has a ray tracing performance which is up to 11 Gigarays per second compared to 10 on the GeForce RTX 2080 Ti, while deep learning performance throughput is rated at 130 Teraflops compared to 57 Teraflops.

The Titan RTX has the same 1350MHz base clock speed as the GeForce RTX 2080 Ti, but a higher boost clock speed of 1770MHz instead of 1635MHz.

The higher count in Teraflops makes a big difference in AI capabilities.  Nvidia has designed the RTX with AI and deep learning as their main priority.

How does it do with graphics memory?

Nvidia’s top-of-the-line RTX GPU has a hefty 24GB of GDDR6 RAM and has the same 14Gbps transfer speed as before. It runs on a 384-bit bus for increased throughput speed. The 12nm architecture processor of the TITAN RTX is embedded with 18.6 billion transistors. Its power consumption is rated at 280W.

It also has support for NVLink which basically interconnects two GPUs of the same kind and creates a single memory pool which creates enormous space for data.

How does it benchmark?

On 3DMark Firestrike, the TITAN RTX gets an overall score of 31,862 which is very impressive considering that it is single chip based GPU whereas the GeForce 2080 Ti when overclocked to the maximum, gets a score of 29,382.

So the TITAN RTX is a workstation grade GPU which is mainly focused on Artificial Intelligence and Deep Learning algorithms. Though it will crush any graphics intensive application run on it, it is not worth an upgrade over the 2080 Ti if gaming or content creation is your main purpose.

Yes, it is expensive at $2499 (or Rs 2,24,000 in India.)

Nvidia Titan RTX specs

• Architecture: Nvidia Turing
• Frame Buffer: 24 GB DDR6
• Base Clock: 1350 MHz
• Boost Clock: 1770 MHz
• Memory Clock: 7000 MHz
• Tensor Cores: 576
• CUDA Cores (single precision): 4608
• Graphics Processing Clusters: 6
• Texture Processing Clusters: 36
• Streaming Multiprocessors: 72
• Memory Data Rate: 14Gbps
• L2 Cache Size: 6144 K
• RT Cores: 72
• Memory Interface: 384-bit
• Total Memory Bandwidth: 672 Gbps
• Texture Rate (Bilinear): 510 GigaTexels/sec
• Transistor Count: 18.6 Billion
• Fabrication Process: 12nm FFN
• Connectors: 3 x DisplayPort , 1 x HDMI, 1 x USB Type-C
• OS Certification: Windows 7 64-bit, Windows 10 64-bit (April 2018 Update or later),Linux 64-bit
• Form Factor: Dual Slot
• Thermal Design Power (TDP): 280 Watts
• Thermal Threshold: 89° C
• Power Connectors: Two 8-pin

Also read: Here’s how you can choose which GPU is used by a game on Windows 10