Early this morning, Nvidia CEO Huang Renxun released a new RTX Ampere architecture, and three graphics cards RTX 3090, 3080 and 3070 in less than an hour. Now, Nvidia China has explained the “knowledge points” of the new graphics card, let’s take a look.
In general, the performance of RTX 3090, 3080 and 3070 GPUs compared with the previous generation Turing architecture products has been increased to 2 times, and the energy efficiency ratio has been increased to 1.9 times. Such a substantial performance improvement is inseparable from the improvement of the architecture and the upgrade of the process.
Nvidia said that “Ampere” is a miracle of architecture, bringing the second generation of RTX.
New streaming multiprocessor
The world’s fastest and most efficient GPU component has a processing capacity of 30 Shader-TFLOP/s, and the FP32 throughput is twice that of the previous generation.
Second generation RT Cores
The new dedicated RT Core has a computing capacity of 58 RT-TFLOPS, which is twice that of the previous generation. It also supports ray tracing, shading and calculation.
The third generation of Tensor Cores
The new dedicated Tensor Core has twice the throughput of the previous generation, and can run AI-driven technologies faster and more efficiently, such as NVIDIA DLSS, with a computing power of up to 238 Tensor-TFLOPS.
NVIDIA RTX IO
Realizes GPU-based fast loading and decompression of game resources. Compared with hard disks and traditional storage APIs, the input/output performance can be accelerated by up to 100 times . Combined with Microsoft’s new Windows version of DirectStorage API, RTX IO transfers the work of dozens of CPU cores to RTX GPU, improves frame rate, and achieves almost instantaneous game loading.
The fastest video memory in the world
NVIDIA cooperated with Micron to create the world’s fastest GDDR6X video memory for the RTX 30 series, providing graphics applications with a data transfer speed close to 1TB/s, maximizing the performance of games and applications.
Next-generation process technology
The new 8N NVIDIA custom process from Samsung enables higher transistor density and higher efficiency.
Nvidia’s public graphics card uses a brand new design, with highlights as follows:
Dual-axis flow cooling solution
The heat dissipation performance is increased by 2 times, and the amazing one-piece design. The graphics card has better heat dissipation and is quieter, bringing unparalleled performance to players and creators.
Exquisite mechanical design and circuit design
A more sturdy mechanical structure, a new low-profile leaf spring and a new 12-pin power interface, leave more space for components and heat dissipation, and are compatible with the 8-pin connector of the existing power supply through the included adapter.
HDMI 2.1
HDMI 2.1 brings higher bandwidth, allowing for the first time a single cable connection to an 8K HDR TV for ultra-high resolution gaming.
AV1 decoding
The first independent GPU to support the new AV1 decoder, allowing players to watch up to 8K HDR network video with only 50% less bandwidth than before.
Naijatechnews has learned that RTX 3070, 3080 and 3090 will be listed on October, September 17 and September 24, respectively, with prices of 3899 yuan, 5499 yuan and 11,999 yuan respectively.
ASUS, Colorful, Gengsheng, GALAXY, Gigabyte, Inzhong, MSI and Zotac will release standard and overclocked versions of GeForce RTX 3090, 3080 and 3070 GPUs in China. GeForce RTX 30 series GPUs will also be sold through products from Acer, Alienware, Asus, Dell, HP, Lenovo and MSI.
For more such interesting article like this, app/softwares, games, Gadget Reviews, comparisons, troubleshooting guides, listicles, and tips & tricks related to Windows, Android, iOS, and macOS, follow us on Google News, Facebook, Instagram, Twitter, YouTube, and Pinterest.