Nvidia updates GPU lineup with faster and more power efficient RTX 30 series2 min read . Updated: 02 Sep 2020, 12:01 PM IST
- Nvidia CEO Jensen Huang announced the new series of graphics processing units at a virtual event
- The GeForce RTX 30 series fuses programmable shading, ray tracing and AI for developers to create entirely new worlds, Huang said
Gamers looking for an upgrade to any of the Nvidia RTX 20 series graphics processing units (GPUs) will soon be able to choose from the new lineup of more powerful RTX 30 series GPUs which were announced yesterday at a virtual event by CEO Jensen Huang.
Based on the new Ampere architecture, the RTX 30 series includes three new products—GeForce RTX 3090, RTX 3080 and RTX 3070.
The base variant, RTX 3070, starting at ₹51,000, comes with 8GB of GDDR6 RAM. According to Nvidia, it is 60% faster than its predecessor, RTX 2070. The mid-variant RTX 3080 will start at ₹71,000 and comes with 10GB DDR6X RAM. It is up to 2x faster than the RTX 2080. The top-of-the-line RTX 3090 will cost ₹1,52,000. It has a three-slot, dual-axial, flow-through design that is up to 10x quieter than the Titan series RTX and keeps the GPU cooler by up to 30 degree celsius. It has a 24GB GDDR6X RAM and is 50% faster than Titan RTX, claims Nvidia.
"GeForce RTX 30 series delivers our greatest generational leap ever. NVIDIA RTX fuses programmable shading, ray tracing and AI for developers to create entirely new worlds," Huang said in a statement.
What separates the Ampere architecture from the 12nm-based Turing architecture of the predecessor is the 8nm-based process from Samsung which enables higher transistor density and more efficiency.
The other highlights of the new architecture are the second generation RT cores which promise 2x throughput over the previous generation with 58 RT-TFLOPS of processing power.
Further, the third generation AI-based Tensor Cores with 238 TFLOPS of processing power, make the new GPUs 2x faster and more efficient in terms of running AI-based solutions Like Nvidia DLSS (deep learning super sampling) which renders fewer pixels and then uses AI to create higher resolution images.
The company recently released DLSS 2.0 with an improved deep learning neural network that boosts frame rates in the game. It also provides gamers the performance headroom which they require to enhance ray tracing settings and increase output resolutions. All that requires more processing capability which the new GPUs are meant to meet those needs.
To improve the memory capacity in the RTX 30 series, Nvidia has collaborated with Micron to create the GDDR6X discrete graphics memory which gives data speeds of up to 1TB/s for applications that depend on graphics cards. This should enhance game and app performance.
Huang also announced that the PC version of the highly popular online multiplayer shooting game Fortnite will now support RTX real-time ray tracing and can take advantage of global illumination, ambient occlusion, reflection and shadows in real time. In a nutshell, this means more realistic visuals during Fortnite gameplay.