After listening to Tang Shengwu's introduction.
Chen Jin nodded his head.
Suddenly, he thought of something else.
He asked Ling Jun Dong, who was sitting on his left, "Old Ling, did Intel release a desktop CPU product last month? It's said that its performance surpassed our 'Green Dragon Second Gen.' Is this true?"
Some time ago, he vaguely remembered that there were many provocative comments under his personal Weibo, many of which were written in English.
His English was not very good, but after noticing the words "F * ck," he knew that it was not something good.
"Yes, that's true."
Ling Jun Dong nodded his head. "The performance of the Green Dragon Second Gen CPU was surpassed by Intel's iX- 10000 CPU. Their counterattack was very swift.
"However, the method they used was very extreme. They doubled the area of the chip wafer and increased the operating frequency to 8GHZ. In the end, in terms of running points, they surpassed our Azure Dragon Second Gen. It could be said that they had turned the tables.
But there was not much meaning to it. The maximum power consumption of this CPU was more than 400 watts. It could be said to be a fire dragon, so hot that you could fry an egg on it. Its commercial value was not great. It was only to prove that they did not lose. This product could not be sold in the market. "
Intel had only won back their pride. They were still losing ground in the market.
Chen Jin frowned and asked, "Wouldn't the CPU be smoking if it was operating at 8GHZ?"
Chen Jin asked with a frown. The operating frequency that silicon chips could withstand was basically 5GHZ. How did they manage to achieve 8GHZ without burning the chips?
"They used the second generation of graphene heat dissipation technology. The heat dissipation capacity has been greatly improved. If liquid nitrogen is used for cooling, it can even be overclocked to 10GHZ. But it can still quickly eliminate the heat and allow the CPU to work normally.
"From this point of view, Intel's product does have some advantages. Our Suzaku second-generation CPU, the highest domestic DIY players can only overclock to 6.8GHZ. We still have some areas to continue to catch up."
Ling Jun Dong introduced.
In addition, Intel planned to conquer the 7-nanometer process CPU within 18 months. At that time, the performance would double.
They also launched research and development of a new architecture, striving to come up with an architecture that was more advanced than X86 within two years, reaching a level comparable to the "Weaver Architecture".
In short, in order to deal with the challenge posed by Xing Hai Technologies, Intel had mobilized all its forces. They had completely abandoned the "squeezing out the toothpaste strategy." Instead, they would throw out a toothpaste every year. They had to suppress Xing Hai Technologies!
However, in Chen Jin's view, these actions of Intel were just a last-ditch struggle.
Xing Hai Technologies would not dance along with them and follow them into the 7 nanometer.
Instead, they drove directly into another track and left Intel behind forever.
This was because there was a CPU made of a material that had a much brighter future than a CPU made of silicon crystal!
Graphene CPU!
The silicon chip had come to an end. It was the real future.
It just so happened that in addition to the Materials Research Center, there was a graphene CPU R&D team in Xing Hai Technologies' CPU Research Institute. The team had been established for more than two years. Chen Jin had also invested a lot of funds and resources into it. He had also provided a lot of high-value information.
In addition, the Materials Research Agency had completely mastered the graphene production technology.
The graphene CPU that he had been looking forward to finally came out with some laboratory products. However, Chen Jin did not have a detailed understanding of the specific performance of the product.
Therefore, he asked Liang Ying, the vice president, to contact the research center. During the video conference, Chen Jin consulted the project leader, Rong Weizhi, on a few questions.
"Team Leader Rong, you said that you have produced a few laboratory products. How are the performance of these products?"
Rong Weizhi was in his early thirties. He was very young. Moreover, he was a top student from a famous school. He was quite arrogant and a perfectionist. Not only did he have high requirements for himself, but he also had high requirements for others. Therefore, he was not very sociable. Many people ridiculed him behind his back. They said that he had an "extremely high IQ, but zero EQ."
Fortunately, this did not delay the progress of the project. Chen Jin did not particularly care.
"No, the performance of the products is not good enough. They are all rubbish."
Rong Weizhi shook his head. "We have determined the technical route of the graphene CPU a long time ago. There are not many problems. What we are doing wholeheartedly is to find ways to reduce the size of the graphene transistor.
In August last year, we successfully produced the first graphene transistor. Its size was 100 microns. It was very far from the commercial standard.
We continued to improve and reduce the size of the transistor. After a month, the size was compressed to 10 microns.
We continued to improve and reduce the size of the transistor. It was compressed to 1 micron, 600 nanometers, 500 nanometers, 400 nanometers, 300 nanometers … Recently, the size can be reduced by 100 nanometers every month.
In August this year, we finally reached the size of the graphene transistor to the level of 100 nanometers.
But it suddenly became difficult to go further. We have not achieved any significant results in the past two months. The size of 100 nanometers is absolutely not enough. It completely does not meet the commercial standard. It must be compressed to 50 nanometers and below. Only then can it be possible to completely replace the traditional silicon CPU. "
Rong Weizhi shook his head repeatedly. He was still extremely dissatisfied with the current progress.
100 nanometers?
Chen Jin's heart skipped a beat. He asked, "What is the operating frequency of the graphene CPU after you have compressed it to 100 nanometers?"
“100GHZ。”
Rong Weizhi replied, "It's easy to reach this operating frequency. The heat generated by the graphene transistor is only one-thousandth of that of the traditional silicon transistor. Its thermal conductivity is also far better than that of silicon. In theory, the operating frequency of the graphene CPU can be increased to 1 THZ, but the heat generated is still less."
Chen Jin was quite surprised. 1 THZ. This operating frequency could be said to be extremely high to an extremely exaggerated degree.
It meant that the efficiency of one graphene transistor was equal to 1,000 silicon transistors.
And for the same area, the number of transistors in a 10 nanometer process chip was about 100 times that of a 100 nanometer process chip.
But in terms of operating frequency, the graphene transistor could be thousands of times higher.
Overall, it could not be said that the 100 nanometer process graphene CPU had no commercial value at all. On the contrary, its commercial value was very high.
For example, the processor used in supercomputers. Even if the manufacturing process was relatively backward, if the operating frequency was set to 100 gigahertz, the computing performance would not be much different, but the electricity cost would be reduced by at least 90%.
Based on the annual electricity cost of more than 300 million yuan for a supercomputer, after using the graphene CPU, the annual electricity cost saved would be more than 270 million yuan.
The electricity cost saved would almost be enough to cover the cost of purchasing the CPU.
Due to Rong Weizhi's perfectionist character, the commercially available graphene CPU was delayed by two months.
Chen Jin immediately criticized him. "The 100 nanometer process is already enough. Why did you only tell us now? You've wasted dozens of days of time! "
"I'm sorry, Mr. Chen. This is my fault. I didn't fully realize its value."
Rong Weizhi realized his mistake. He scratched his head to express his apology.
Chen Jin shook his head and did not continue to criticize him. Instead, he issued two instructions. First, continue to explore the graphene CPU with a smaller process.
Second, immediately commercialize the 100 nanometer process graphene CPU for the development of a supercomputer with lower power consumption!
After all, supercomputers consumed too much power. Xing Hai Technologies had a lot of large servers and supercomputers. Among them, there were more than 30 supercomputers. They had reached the standard of opening a supercomputer internet cafe.
What followed was naturally the extremely exaggerated electricity cost.
Even a nouveau riche like Chen Jin could not completely ignore the electricity cost of at least 3 to 4 billion a year. It would be great if he could save some of the electricity costs.
Therefore, the use of graphene CPU could be said to be urgent.
You've already exceeded your reading limit for today. If you want to read more, please log in.
Login
Select text and click 'Report' to let us know about any bad translation.
