NVIDIA's success is not unnoticed. Almost all major chip makers are suddenly starting to pursue AI dreams, and a large number of startups are beginning to develop new deep learning chips. In the future of technology, deep learning is of paramount importance, so the most important customer of NVIDIA, Google, is now also making chips and becoming a competitor. 　 From video games to artificial intelligence, NVIDIA’s attack has never stopped. Nvidia is ahead of other technology giants in deep learning, and its advantages are self-evident. In the increasingly fierce competition, if you want to maintain its status, you can't relax. Founded in Taiwan, the founder Huang Renxun immigrated to the United States with his family at the age of 9. At the age of 30, he founded Nvidia with two other partners. The share price has more than doubled in the past five years. The Forbes writing describes how Nvidia creates miracles and constantly changes the world. In Denny's shop near a flyover in San Jose Berri Essa, Nvidia's co-founder Chris Malachowsky ate sausages and omelettes and sipped coffee with burnt flavor. In April 1993, in this dilapidated restaurant, three electrical engineers - Malakowski, CurTIs Priem and Nvidia's current CEO, Huang Renxun, founded a company. . The company aims to make special chips that are faster and produce a more realistic picture. At that time in San Jose, the law and order was not very good. People shot at the police car that was parked, and the front of the restaurant was full of bullets. No one would have thought that these three young people who are drinking unlimited cups of coffee in this dilapidated shop will build a company that changes the world. The impact of this change and Intel brought to the world in the 1990s. The same is true of change. Malakovsky said: "In 1993, such chips had no market at all, but we saw the upcoming wave. There is a five-month surfing competition in California, when the organizers observed some surges in Japan. Or when the storm, they will inform the surfers to come to California, because the wave will come in two days. This is what we did at the time, we were in the stage of watching the tide to the tide or the storm." The wave of NVIDIA co-founders saw the emerging image processing (GPU) market. Most of the chips they make are sold in card form for gamers to plug into the PC motherboard for an ultra-fast 3D picture experience. The company's chips are named after male hormones such as "TItan X" or "GeForce GTX 1080", which are priced at up to $1,200. Today, 20 years later, Nvidia’s annual revenue has reached $5 billion, but sales of these chips still account for more than half of the company’s total revenue. Although NVIDIA has shown unexpected adaptability in PC games (when the entire PC industry suffered from a cold winter, NVIDIA increased 63% year-on-year in the most recent quarter), but it is not their game that makes Wall Street drool over the company. The business is the company's artificial intelligence (AI) business. This small one-byte silicon can change more magic, it can summon a different landscape, you can also draw a perfect explosion map. The most suitable and hottest application for AI is deep learning. Deep learning allows computers to learn autonomously, programmers do not need to manually code all programs, and AI offers unparalleled advantages in image recognition and speech recognition. In order to build a data center, technology giants such as Google, Microsoft, Facebook and Amazon are buying a large number of NVIDIA chips. Institutions such as the Massachusetts General Hospital use NVIDIA's chips to detect abnormalities in medical images such as CT scans. Tesla also recently announced that in order to achieve autonomous driving, NVIDIA's GPUs will be installed in all of its vehicles. In Facebook and HTC's VR products, NVIDIA can also drive VR helmets as a base device. Nvidia is headquartered in Santa Clara, California. Huang Renxun is dressed in his iconic black dress – black leather shoes, black jeans, black Polo shirts, black leather jackets at the company headquarters. He said, “In the development of our company, we have never been so big like this. The market center. This is due to the fact that we are doing the right thing, this is the graphics calculation (GPU compuTIng)." There are 3,000 AI startups around the world, most of which are based on NVIDIA's platform. They use NVIDIA's GPU to develop apps that include stock trading, online shopping, and drone navigation. There is even a company called June that uses NVIDIA chips to make AI ovens. “We have invested in a large number of startups and applied deep learning to many areas, each of which is based on a NVIDIA platform,†added Mark Anderson of venture capital firm a16z. “It’s like In the 1990s, everyone started to develop around Windows, and in the first decade of the 21st century, they all developed around the iPhone." Anderson joked: "There is a game inside our company called 'If we are a hedge fund, which company will be invested', everyone chooses to invest in Nvidia." NVIDIA is the leader in the GPU field with a market share of 70%. NVIDIA's expansion into these new markets has caused its stock to soar. In the past 12 months, its stock market value has almost doubled, almost five times that of five years ago, and its market value has increased by more than 40 times to reach 50 billion US dollars, becoming the company with the highest market value in this field. Nvidia's good performance also made Huang Renxun's personal worth worth 2.4 billion US dollars (co-founder Malakovsky has been semi-retired, another founder Primm left NVIDIA in 2003). Soaring stocks put Nvidia in the top of the semiconductor industry's first "Just 100 Best Corporate Citizenship" list. This list was jointly released by Forbes and Just Capital, which was founded by billionaire hedge fund investor and philanthropist Paul Dusit Jones II. "Just 100" interviewed 50,000 Americans and asked them to evaluate about 1,000 listed companies. The scope of the survey covered the company's attitude toward employees, customers and shareholders to evaluate the company. Among the 10 indicators surveyed, NVI's scores on employee benefits, benefits, product contributions and environmental impact far exceeded the average. The company's employee-friendly policies, such as long vacations, flexible working hours and stress management courses, have enabled NVIDIA to leave rivals behind Glassdoor (Glassdoor is an anonymous evaluation of the work environment that is popular among Silicon Valley technology professionals in the job-hopping period. website). In such a well-known and homogenized industry, NVIDIA has carried out various projects to promote the more core positions of women and ethnic minorities. Huang Renxun said, "I see the company as a person and see it as a living individual. Corporate culture is the company's genes, or the company's operating system. If you want to say that I have any corporate development experience, then that is the corporate culture is the most An important part." Huang Renxun has been well aware that NVIDIA's image processing chips are not only used in games but also have more potential. But he did not think that NVIDIA's GPU can be applied to deep learning. Deep learning techniques (also known as neural networks) draw on the workings of brain neurons and synapses. Deep learning technology emerged in the academic world in the 1960s and made significant progress in the 1980s and 1990s. However, there are always two factors that restrict the development of deep learning: 1. Big data that can train deep learning computing programs; 2. Cheap, efficient computing power. The emergence of the Intel network solved the first problem - suddenly everyone's fingertips can reach a huge amount of data. But the second problem is still not solved. This problem continued until 2006, when NVIDIA released a programming tool called CUDA that makes it easier for programmers to edit every pixel on the display. The GPU simulates thousands of tiny computers simultaneously processing pixels. These tiny computers run low-level operations to generate shadows, reflections, light, and transparency. Before the release of CUDA, GPU programming was very hard for programmers, and they needed to write a lot of low-level machine code to complete programming. NVIDIA has spent years developing CUDA, bringing high-level programming languages ​​such as Java and C++ to CUDA to make programmers' coding easier. With CUDA, researchers can develop deep learning models in a faster and cheaper way. “Deep learning is almost the same as the brain,†says Huang Renxun. “It’s very useful. You can teach him almost everything. But it has a very big barrier, that is, it requires a lot of calculations. And our GPU has such calculations. Capabilities are ideal for deep learning computing models." Formation System,Quality Formation System,Formation Automatic Line System,Automatic Battery Waterbath Formation System Zhejiang Baishili Battery Technology Service Co,.Ltd. , https://www.bslbatteryservice.com