top
2020/10/15 08:35:47 Augmented realityAR

COVID-19's efforts in virtual space are accelerating!

COVID-19's efforts in virtual space are accelerating!


This year, the efforts for virtual space by new viruses have been accelerating worldwide. Even if COVID-19 has subsided this time, it is said that a new virus will occur once every few years in the future. In the future, what I used to take for granted has become an era where I have to think about a new life. Therefore, what is currently attracting attention is the change to a lifestyle with a new sense of the world using virtual space. This time, we would like to introduce some of the government's attempts.

ハロウィーンは仮想空間の“渋谷”を楽しんで 区が啓発 東京
Enjoy the virtual space "Shibuya" on Halloween and enlighten the ward Tokyo
October 1, 2020 17:04

While Shibuya Ward in Tokyo is calling for refraining from coming to the city during this month's Halloween period, an event to enjoy Halloween will be held on the Internet using a virtual space that reproduces the city of Shibuya. became.

Shibuya Ward is calling for refraining from coming to Shibuya to prevent the infection of the new coronavirus, as many young people gather around Shibuya Station every October on Halloween.

Under these circumstances, KDDI and Shibuya City Tourism Association will be holding an event where popular artists and comedians can enjoy online live performances in the virtual space town "Virtual Shibuya" officially recognized by Shibuya Ward during Halloween from 26th to 31st of this month. Announced that it will be held.

This means that you can take in a dedicated app, operate your avatar from your smartphone, etc., and participate in events for free while freely going back and forth in the online city of Shibuya such as scrambled intersections.

Ken Hasebe, Mayor of Shibuya Ward, said, "This is an initiative where you can enjoy Shibuya without actually coming to the city of Shibuya and participate in cutting-edge virtual events. I want many people to enjoy it."

 

* Reference news

米国時間の10月5日に開幕、10月9日まで開催中のGPU Technology Conference 2020より。RTX A6000とA40を公開するNVIDIAのJensen Huang(ジェンスン・フアン) CEO


GPU featured articles for developing virtual space!
The United States NVIDIA announced the GPU “NVIDIA RTX A6000” for workstations and the GPU “NVIDIA A40” for servers / data centers on October 5 (local time). Both are based on the Ampere architecture. The RTX A6000 will begin shipping cards in mid-December, and the A40 will begin rolling out on-board systems in early 2021.


From GPU Technology Conference 2020, which starts on October 5th, US time and continues until October 9th. NVIDIA CEO Jensen Huang unveiling RTX A6000 and A40

NVIDIA RTX A6000
The "NVIDIA RTX A6000" is a new GPU offered for workstations that replaces the traditional Quadro RTX. It seems to be based on the GA102 core of Samsung architecture manufactured by Samsung's 8nm process, which is the same as the GeForce RTX 30 series for consumers, but the number of CUDA cores, operating clock, and VRAM capacity have been increased compared to GeForce RTX 3090. The single-precision floating-point arithmetic performance is said to reach a maximum of 40 TFLOS.

The main specifications currently released are as follows.

GPU core: Ampere (Samsung 8nm)
Number of transistors: 28.3 billion
Die size: 628.4mm2
Number of CUDA cores: 10,752
Number of Tensor cores: 336
Number of RT cores: 84
Memory type: 48GB GDDR6 ECC
Memory bandwidth: 768GB / sec (384bit connection)
Power consumption: 300W
Bus interface: PCI Express 4.0 x16
Video output: DisplayPort 1.4 x 4 systems

NVIDIA RTX A6000

NVIDIA A40
The "NVIDIA A40" is also based on the GA102 core, and its specifications are close to the RTX A6000, but the floating-point arithmetic performance is slightly lower than the RTX A6000, so it is thought that the upper limit of the operating clock is lower than the RTX A6000. The main specifications of the A40 currently released are as follows.

GPU core: Ampere (Samsung 8nm)
Number of transistors: 28.3 billion
Die size: 628.4mm2
Number of CUDA cores: 10,752
Number of Tensor cores: 336
Number of RT cores: 84
Memory type: 48GB GDDR6 ECC
Memory bandwidth: 696GB / sec (384bit connection)
Power consumption: 300W
Bus interface: PCI Express 4.0 x16
Video output: DisplayPort 1.4 x 3 systems
Unlike the RTX A6000, which is offered in the form of a general graphics card with a cooling fan like the GeForce GPU, the A40 uses a passive cooling card with only a heatsink. This is because the A40 for servers / data centers is supposed to be installed in a large-scale system with a separate cooling mechanism. On the other hand, the A40 also has three DisplayPort outputs, which can meet the needs of screen output.


NVIDIA A40

* This article is information at the time of publication and may differ from the latest one. Please note.
* For the latest information on the new coronavirus infection, please also check the information published by public institutions such as the websites of the Ministry of Health, Labor and Welfare, the Cabinet Secretariat, and the Prime Minister's Office.