The National Day holiday chased the Douban score of 9.0 "Meritorious Service", which was very enjoyable and very touching.
"Meritorious Service" has eight units, the second unit is "The Unsung Hero Yu Min", Yu Min everyone knows, the father of China's hydrogen bomb.
There is such an interesting plot:
One night, Yu Min was working in the computer room when he suddenly heard a strange sound, and his colleague said that he was hungry and his stomach screamed.
Yu Min said no, and immediately got up to check the long paper tape spit out by the computer:
After looking at it for a while, he said: Here is wrong, from here on it is wrong, the machine is broken, let the technician come and visit.
Then he returned to his seat and continued to work, paying attention to the expression of a person next to him, who was full of disbelief: How could it be that the machine was broken?
The technician came to check, and sure enough, he found that the component of an adder was broken, and after replacing it, the calculation was normal immediately.
It's a godsend! How can the gap between people and people be so big!
The paper tape that Yu Min saw was not a number, nor a text, but something like this:
This feeling is equivalent to seeing a diagram like this:
Then judge that there is a problem with the data here, the computer is wrong, and you say that the cow is not a cow!
TV dramas may require artistic expression, according to the memories of Academician Du Xiangwan, the actual situation is like this:
The structure of nuclear weapons has many layers, and after the explosion of various materials, at each point in time and space, there are physical quantities such as temperature, speed, pressure, and acceleration. Yu Min suddenly found that a certain amount was suddenly abnormal from a certain point.
Everyone immediately went to check the reason. Du Xianghuan went to check the equations and parameters and found no errors; the people who did computational mathematics and programming went to check the reasons and did not find errors.
Finally, the inspection found that the original of the original adder was broken, and after replacing it, the physical quantity was normal immediately.
Yu Min's great point is that he understands the laws of physics very thoroughly, and in the vast sea of data, he can find that this trend is wrong, and the amount is wrong.
Then exclude equation and parameter errors, program errors, and finally suspects to the head of the computer.
Why should computers be suspected? How can a computer go wrong?
At that time, there were two fastest computers in China, which operated 50,000 times per second. One is the 119 developed by the Computing Institute, located in Beijing; the other is the J501, which is copied on the basis of the 119, located in Shanghai, running in one south and one north.
Don't look at the only 50,000 operations per second, but it was the fastest computer at that time, astronomy, meteorology, oil, nuclear bombs, etc. are all competing to use, and the machine time is very precious.
Yu Min's group led the team to use J501 because the 119 computer in Beijing was out of machine.
(J501 Computer)
But 119 and J501 are the first generation of tube computers, very unstable, their trouble-free running time I did not check, you can use Academician Xia Peisu to preside over the research and development of the 107 machine as a reference, 107 machine continuous trouble-free operation for more than 20 hours, which broke the record of the tube computer at that time.
When the machine jumped, the calculated result was not correct, so Yu Min suspected that the J501 computer was wrong.
When the computer makes a mistake, it will be abandoned, and when a lot of machines are wasted, Yu Min thought of a trick, every once in a while, the calculation results are temporarily stored in the magnetic drum, if there is a mistake, it will start over from the recent calculation results, which is called "taking the drum and redoing", using this method to reduce the loss of machine time.
But in this way, the machine is inseparable from the person, the machine works 24 hours, and the person also has to open his eyes and look at the calculation results printed on the paper tape.
For programmers, it may be curious, in what language is such a computer programmed?
Sorry, there is no assembly, let alone high-level languages such as Java and Python.
It all depends on hand-compiled machine instructions to implement the program, and a binary bit cannot be mistaken.
Little friends look at the machine instructions of the 119 computer, and program on such a computer, which is simply not a human job.
(Click to enlarge)
The computer's operation console only has two-input neon lamps and wrenches, and a number of buttons, which can only be used by one person, while the scheduled program debugging time is only 5-10 minutes, and everyone has to do their own tricks to improve programming and machine efficiency.
Under such a difficult circumstance, Yu Min and other scientific and technological workers of the older generation still sleep and forget to eat, and work day and night. On October 29, 1965, Yu Min opened the clouds and fog, put forward a new idea, made the direction of the hydrogen bomb configuration clearer and clearer, and formed a complete physical design scheme for the hydrogen bomb from the principle, material to the configuration.
The theoretical breakthrough makes people feel happy, and the bulky J501 computer has become cute! Cai Shaohui, then deputy director of the Scientific Research Office of the Theory Department of the Ninth Academy of Sciences, recorded the excitement at that time:
On the evening of November 1, 1965, in the J501 computer room of the East China Institute of Computing Technology, under the soft lighting, Tang Minjun, the leader of the computing team, played with the key buttons on the computer console, and the small neon lamp on the countertop was flashing happily. Meng Zhaoli and I squatted on the floor of the computer room and busied ourselves with the paper tape rolls, checking whether there were any errors in the input and output data. Yu Min took a slide rule and a pencil from time to time and wrote something.
The computer faithfully executes the master's instructions, sometimes emitting a beautiful melody, and sometimes emitting the heavy cyclic wheezing that is characteristic of the process of numerical solution superposition. Tang Minjun said: "Listen, now it has reached the most difficult stage, the energy equation can not be iterated, you have to quickly reduce the time step!" Only to see her fiddle a few more times on the operating table, and gradually the machine played a smooth rhythm, and people exchanged faces in joy. The paper tape rolls out exciting numbers slowly, and everything is so exciting!
Reading such a text now, it is really emotional, and I salute the older generation of scientists!