NVIDIA GeForce RTX 3090 Ti: full specs, price, performance details

NVIDIA's new flagship GeForce RTX 3090 Ti graphics card is reportedly right around the corner, with the suped-up card packing the GA102 GPU -- the flagship GPU and the fastest gaming GPU that NVIDIA has ever created -- alongside 24GB of ultra-fast GDDR6X memory.

2

The new GeForce RTX 3090 Ti continues on the GPU dominance that NVIDIA built with its current flagship RTX 3090, offering the same but also changed 24GB of GDDR6X memory. We still have the same 24GB of GDDR6X but it is both faster -- clocked at 21Gbps -- and has a much wider 384-bit memory bus that for the first time ever, has a GeForce GPU pushing over 1TB/sec of memory bandwidth.

NVIDIA is using the GA102 GPU with 84 SM units that provide a huge 10,752 CUDA cores -- up from the 82 SM units and 10496 CUDA cores on the RTX 3090 -- but there's also next-gen RT (Ray-Tracing) cores, Tensor cores, and brand new SM (streaming multi-processor units). The GA102 GPU inside runs at a base GPU clock of 1560MHz, while boost GPU clocks hit 1860MHz. The TDP has been massively increased at 100W

Micron has provided the GDDR6X memory once again, with NVIDIA's new GeForce RTX 3090 Ti having 24GB of GDDR6X clocked at 21Gbps -- but it's also on a 384-bit memory bus -- capable of just over 1TB/sec of memory bandwidth (well, 1008GB/sec memory bandwidth to be specific).

We are expecting NVIDIA's in-house GeForce RTX 3090 Ti Founders Edition to be the first PCIe 5.0 graphics card ever, with a single 16-pin PCIe power connector that will provide up to 450W of power to the card, while the PCIe 5.0 connector itself can handle up to 600W of power so dual-GPU setups using dual PCIe 5.0-capable GPUs would be consuming in excess of 1200W. Insanity, but man do I love it and I can't wait to play around.

• NVIDIA GeForce RTX 3090 Ti: 40 TFLOPs (FP32) / 78 TFLOPs (RT) / 320 TFLOPs (Tensor)
• NVIDIA GeForce RTX 3090: 36 TFLOPs (FP32) / 69 TFLOPs (RT) / 285 TFLOPs (Tensor)

2

Marvels of NVIDIA Ampere Architecture - 2nd Generation RTX

Enabling the blistering performance of the new RTX 30 Series GPUs and the NVIDIA Ampere architecture are cutting-edge technologies and over two decades of graphics R&D, including:

• New streaming multiprocessors: The building block for the world's fastest, most efficient GPU, delivering 2x the FP32 throughput of the previous generation, and 40 Shader-TFLOPS of processing power.
• Second-gen RT Cores: New dedicated RT Cores deliver 2x the throughput of the previous generation, plus concurrent ray tracing and shading and compute, with 78 RT-TFLOPS of processing power.
• Third-gen Tensor Cores: New dedicated Tensor Cores, with up to 2x the throughput of the previous generation, making it faster and more efficient to run AI-powered technologies, like NVIDIA DLSS, and 320 Tensor-TFLOPS of processing power.
• NVIDIA RTX IO: Enables rapid GPU-based loading and game asset decompression, accelerating input/output performance by up to 100x compared with hard drives and traditional storage APIs. In conjunction with Microsoft's new DirectStorage for Windows API, RTX IO offloads dozens of CPU cores' worth of work to the RTX GPU, improving frame rates and enabling near-instantaneous game loading.
• World's fastest graphics memory: NVIDIA has worked with Micron to create the world's fastest discrete graphics memory for the RTX 30 Series, GDDR6X. It provides data speeds of 1TB/s system memory bandwidth for graphics card applications, maximizing game and app performance.
• Next-gen process technology: New 8N NVIDIA custom process from Samsung, which allows for higher transistor density and more efficiency.