Chapter 420: Lend Your Good Words
The Island Project Team did not directly respond to the challenge posed by IBM.
"The Island Quantum Computer is not a product at the same stage as IBM's Vulture. If IBM is willing, we can let our next-generation product compete with the Vulture in computing speed."
The Island quantum computer was released more than a year earlier than the Vulture.
Moreover, the "Vulture" is connected to a supercomputer, so it is obviously unfair to compare the two directly.
But the response from the island project team made IBM feel that we did not dare to take up the challenge.
"Sure enough, they know in their hearts that the performance of the island is not as good as that of the vulture. They have not yet found the real direction in the optimization of quantum computers."
IBM researchers believe that the combination of supercomputing and quantum computers is the best direction for the future development of quantum computers.
The island project team did not do so, most likely because the technology did not meet the corresponding requirements.
"Next generation product? It will take many years for them to develop the next generation of quantum computers. They know that if the island competes with the vulture, there is absolutely no chance of winning. This is just their delaying strategy."
"In their military tactics, this is called a 'delaying tactic'. In fact, it is normal for them to lag behind us in technology. There is nothing to be ashamed of."
With this idea in mind, IBM continued to reply to the island project team.
“We are very willing to compete with your next-generation products in computing. We look forward to the early release of your next-generation products!”
Our response to IBM's very official wish is very simple.
"I'll lend you my good words."
In fact, at this time, the quantum computer has been assembled and only the final testing phase remains.
Since the optical quantum chip for computing has passed the test, what needs to be done next is mainly some optimization work.
Xu You is also writing corresponding programs for this new quantum computer to build a new artificial intelligence.
Compared with superconducting quantum computers, optical quantum computers are more difficult to program and require solving many problems that have not been encountered in superconducting quantum computers.
Fortunately, Xu You began to think about and deal with these problems when he was developing Loongson G1.
With these foundations, Xu You will not seem powerless when facing these difficulties, and can even solve them with ease.
After repeated writing, testing and modification.
Xu You finally created a system language suitable for the optical quantum chip, allowing it to complete various programming tasks smoothly.
After solving the most critical problem, Xu You imitated Haidao's artificial intelligence and tried to make Suanjing have the ability to learn independently like Haidao.
" Let the island be your enlightenment teacher."
Although in the quantum random circuit sampling task, the computing speed of "Suanjing" is already a thousand times that of "Haisha".
But this does not mean that "Suanjing" has completely replaced "Haidao" and become a better quantum computer.
In terms of application, "mathematics" is still a newborn baby and there is still a lot to learn.
The "Haidao", which is relatively mature in technology, is very suitable to be a "mathematical" teacher.
Under the guidance of Haidao, Suanjing began to learn the various skills that Haidao already possessed.
Suanjing learned very quickly. In less than a day, he started from scratch and mastered many skills including Go, Chinese chess, international chess, medical diagnosis, etc.
In terms of deep learning capabilities, Suanjing is far superior to Haidao.
After only a few hours of study, Suanjing defeated Haidao at Go. Even when Suanjing gave Haidao two pieces, he could still win with a slight advantage.
Xu You previously thought that when his chess skills reached Haidao's level, even if he tried to improve further, there wouldn't be much room for improvement.
What Xu You didn't expect was that Suanjing could raise the ability to solve Go to a higher level in such a short period of time.
You should know that in Go, giving up two pieces does not just mean a difference of two points, but is roughly equivalent to giving up twenty points.
Such results show that humans are still a long way from finding the optimal solution to Go.
In addition, in the medical , Suanjing has also demonstrated a speed far faster than Haidao.
Once the technology matures, it is very likely that AI will be able to directly serve as an AI doctor in the hospital.
After completing the study and testing of these functions, the project team officially announced the news to the public.
"Suanjing, a quantum computer jointly developed by Jijing University, Huazhong University of Science and Technology, Jinghua University and other research institutions, was officially released today. Suanjing adopted the solution of photon quantum and successfully constructed 1,130 photons. It only took three milliseconds to perform the quantum random circuit sampling task, which is one thousand times faster than the Haidao quantum computer. This is also the world's first quantum computer that can solve complex task programming."
Because the "Suanjing" project was initiated by Xu You and was also led and completed by Xu You.
According to the ranking of scientific research institutions, Jijing University also ranks first in terms of project contribution.
Just to prevent Xu You from being too conspicuous, Xu You's name was not specifically highlighted in the announcement.
In fact, many people know that China has been researching the next generation of quantum computers.
However, no one expected that this generation of quantum computers would be released so soon and be so much better in performance than the previous generation.
"I thought it was Haidao No. 2, but it turned out to be a new series. But the speed is really too exaggerated. Haidao is so amazing, and the speed of the arithmetic is a thousand times faster than Haidao. Isn't this going to beat the vulture of the United States?"
"If my analysis is correct, Suanjing and Haidao are taking two completely different directions. Didn't we say before that optical quantum computers cannot be programmed because of the difficulty of logical operations between optical quantum bits? This time, has the problem of optical quantum being unprogrammable been solved?"
"Three milliseconds? That's about three thousandths of a second. The Vulture released by IBM previously took nearly a second to perform the same task. This means that the speed of the arithmetic is three hundred times that of the Vulture. There is no comparison between the two."
When IBM researchers learned of the news, they expressed serious doubts about it.
"Optical quantum computer? They actually chose this direction?"
This direction has always been considered by the industry to be a direction with no prospects because it has always been difficult to program.
