Industry News

NVIDIA's "ambition" is too big: 1nm chips must be developed, and where there is AI, there is me

Update time : 2023-03-29 11:31:05

None of us thought that in 2022 we would say goodbye to the epidemic that has plagued us for 3 years, nor did we expect Ai to sweep our lives at the speed of supernova explosions, allowing us to glimpse the light of hope outside the door of the next generation of the information age, and break the curse of the evolution process, with Ai into the 1nm era.

Nvidia unveiled its new GPU architecture, NVIDIA Hopper, at GTC 2022, and later launched the NVIDIA H100, a product based on this architecture. Today, the H100 has become one of the most used GPUs for training artificial intelligence models. At the 2023 GTC conference, Nvidia founder Jen-Hsun Huang once again took the lead, bringing an innovative technology – NVIDIA cuLitho. NVIDIA cuLitho is a set of lithography libraries integrated into NVIDIA Hopper architecture that can improve GPU efficiency in chip manufacturing scenarios. Lao Huang announced that he has established partnerships with TSMC, ASML and other companies.
Nvidia predicts that replacing the CPU on the current lithography machine with NVIDIA cuLitho software library and Hopper GPUs can increase overall efficiency by 40 times and reduce computing workload by tens of billions of CPU hours. In addition, cuLitho technology can significantly reduce energy consumption and carbon emissions. As the chip process continues to move towards 3nm or even 2nm and 1nm nodes, computing demand is growing, and cuLitho's high efficiency and low energy consumption have brought important breakthroughs to the development of the chip manufacturing industry.
For the sake of AI computing, NVIDIA will naturally not let it go. DGX Cloud is an AI supercomputing service that gives businesses quick access to the infrastructure and software used to train advanced models for generative AI and other innovative applications. DGX Cloud provides NVIDIA DGX AI supercomputing clusters with 8 NVIDIA H100 or A100 80GB Tensor Core GPUs. Each node has 640GB of GPU memory for advanced AI training needs. Based on DGX Cloud's powerful computing power, NVIDIA has also developed a range of application models, such as NVIDIA AI Foundations, including NVIDIA NeMo language and NVIDIA Picasso image, video and 3D services.
For Ai computing enthusiasts, high-end RTX4080 and 4090 are generally used for computing, but these flagship graphics cards are not only expensive, but also explosive in power consumption, so at last night's GTC 2023, Lao Huang launched a small RTX 4000 graphics card, named RTX 4000 SFF, based on the Ada Lovelace architecture, in a smaller package size than the existing RTX 4000 series. This graphics card is primarily aimed at small businesses, powering CAD, graphics design, AI applications and software development.
In conclusion, Nvidia demonstrated its lofty ambitions for the AI market at GTC 2023, helping the development of various industries with innovative technologies, powerful Cloud as a Service, and targeted graphics cards. Nvidia is advancing Ai's innovation at an unprecedented rate, and hopes that the foreseeable sea of Ai will meet us as soon as possible.

Previous : For ruthless price increases such as Xiaomi/Qualcomm, some manufacturers have considered abandoning ARM chips: the opportunity for RSIC-V is here

Next : Renesas Electronics Announces New RZ/T2L Industrial MPU for Fast and Accurate Real-Time Control via EtherCAT Communication

How many chips does a car need?

Sep .19.2024

Automotive chips can be divided into four types according to their functions: control (MCU and AI chips), power, sensors, and others (such as memory). The market is monopolized by international giants. The automotive chips people often talk about refer to

Position and Function of Main Automotive Sensors

Sep .18.2024

The function of the air flow sensor is to convert the amount of air inhaled into the engine into an electrical signal and provide it to the electronic control unit (ECU). It is the main basis for determining the basic fuel injection volume. Vane type: The

Chip: The increasingly intelligent electronic brain

Sep .14.2024

In this era of rapid technological development, we often marvel at how mobile phones can run various application software smoothly, how online classes can be free of lag and achieve zero latency, and how the functions of electronic devices are becoming mo

LDA100 Optocoupler: Outstanding Performance, Wide Applications

Sep .13.2024

In terms of characteristics, LDA100 is outstanding. It offers AC and DC input versions for optional selection, enabling it to work stably in different power supply environments. The small 6-pin DIP package not only saves space but also facilitates install

About Us

About Us What we do Quality People

Products

CHIP/IC Active devices Passive device Components OTHER

China electronic components supplier----Wah Luen Shing Electronic.----www.wlschip.com

Leave a message