4 times faster than the previous model: Nvidia introduced its new artificial intelligence super chip

class = “sc-cffd1e67-0 iQNQmc”>

Nvidia CEO Jensen Huang introduced the new generation chips. Nvidia sees the system, called Blackwell, as “leading a new industrial revolution” through artificial intelligence (AI). When it comes to AI training, the system is four times more powerful than the current generation Grace Hopper.

Nvidia’s computer systems dominate AI training in data centers. The group also wants to expand its role in creating content using artificial intelligence. Huang emphasized Monday at the GTC developer conference in San Jose that the Blackwell system is 30 times better than Hopper. Nvidia also has new software for this, which can also be used via interfaces over the cloud.

Faster and more ecological

Huang believed that in the future, most content will not be prefabricated from storage, but AI software will create this content in a new way based on the current situation. Nvidia developed the computer system for this future. For example, instead of looking at data from somewhere, you will be able to talk to buildings via chatbot.

For example, Grace Hopper and the chatbot ChatGPT can be trained in three months with 8,000 Nvidia chips and 15 megawatts of power consumption, Huang said. With Blackwell you can do this with 2000 chips and 4 megawatts of electricity at the same time.

ChatGPT is the chatbot from OpenAI, the developer company that sparked the current excitement about artificial intelligence a year ago. Such software is trained with large amounts of information and can use this as a basis to create images from sentences and text features, for example, at a human language level. Nvidia also wants to use this technology to create weather forecasts, Huang said. For this purpose, the company is developing a world simulation called Earth 2.

Huang emphasized that Nvidia has deployed Maxwell chips into computer systems that consist of approximately 600,000 components and weigh more than 1.3 tons. Originally developed for graphics cards, Nvidia technologies have long been proven in computing work for artificial intelligence applications. Competitors like Intel and AMD have so far been unable to catch up. This allows Nvidia’s business and stock market value to grow rapidly. Tech giants such as Microsoft, Google and Amazon are already planning to use Blackwell.

View with Apple Vision Pro

Nvidia’s new artificial intelligence super chip is named after American mathematician David Blackwell. Nvidia also wants to promote the use of digital twins, where companies can simulate their entire work on the computer. Huang emphasized that in the future, before building something in the real world, you will simulate it digitally.

With this concept, Nvidia makes Apple’s Vision Pro computer glasses more useful for use in companies. Nvidia’s Omniverse 3D environment, which allows companies to create digital twins of their products and factories, is coming to Apple devices. Technology is often used in the design of objects.

Huang demonstrated this at the internal developer conference GTC on Monday, using the example of South Korean automaker Hyundai, among others. This meant that a designer could look at different color variations of a car model in different environments and also virtually sit behind the steering wheel. With the $3,499 Vision Pro, Apple wants to build a computer called a spatial computer where digital content can be displayed in the real environment.

Nvidia also relies on robots. “Everything that moves will be robotic,” the company’s boss said. The aim is for robots to learn by simply observing people. “The ChatGPT moment for robotics may be around the corner,” Huang said. Simulations in which robots learn for the real world should play a central role.

(SDA/bot)

Source :Blick