Chapter 607 Desert’s Request
Real-time encryption of voice data streams generated in real time is simply not something that a computer with specifications like the DJS-60D can handle.
But today Gao Zhendong's hardware multiplier gave them a hint, whether they could make a dedicated encryption hardware and string it in the voice circuit to use the advantages of high-speed hardware to encrypt the voice data stream.
"Well, that's fine. Please sort out your encryption requirements and come to me when the time comes. By the way, if it's convenient, you can also tell me about voice digitization. Maybe I can help. You.”
The comrade working in communications was stunned for a moment. Do you understand this?
The chief engineer of the National Defense Work Committee was overjoyed. How could he have forgotten this? This guy is leading comrades from the Communications Institute of the Postal and Communications Commission to develop program-controlled switches. I heard that the progress is very good. He must also be doing research in this area.
"That's great!" He said with a smile, turning to the comrades engaged in communications: "You go back, sort out the relevant requirements for voice digitization and voice encryption, and go back to Comrade Gao Zhendong."
The chief engineer who met with the committee said so. They knew that Chief Engineer Gao must be very good in this area, so they happily agreed.
To be honest, the single-cycle hardware multiplier only allowed them to see the dawn of digitalization, but they were still blank on how to do it. They had no idea. It would be great to have technical guidance.
For Gao Zhendong, this was originally the professional content of his previous life - PCM processing technology for voice calls. As a university with a communications background, automation majors also need to learn this.
PCM is a set of time division multiplexing technology for digital program-controlled switches, which naturally includes voice coding technology.
Thinking of this, Gao Zhendong suddenly realized that he should also pull the program-controlled switch upward.
I didn't expect that the comrades who made photolithography machines were so capable that they could make the process to 10 microns all of a sudden, so I temporarily put aside the idea of digital switches and started to engage in air separation switching.
Unexpectedly, the plan has changed. Now it seems that the conditions are basically met, at least for the time exchange, the conditions are already met.
However, the research on space division switching is not in vain. Time division and digital switches will originally use space division switching.
Well, we can carefully analyze this matter, and then bring together the comrades from the Communications Institute and those who provide communications for the troops, and talk to them about the PCM.
——
Deep in the desert, several comrades were writing at desks in front of computers.
Computers are already much faster than humans, but they are still a bit slow for the amount of calculations they do. They are doing other work while waiting.
The calculation time of these tasks is measured in hours or even days. If you just wait, it will be a waste of time.
"I wish I could finish it all at once. Isn't that a bit greedy? Hahaha." A comrade raised his head and twisted his wrist.
Whenever we do research, one of the main tools is a pen. Many comrades who do scientific research do not have very healthy wrists.
"Haha, you are indeed a bit greedy. Even when I was at Citi, I never thought that I would be able to work in such a condition." Another comrade raised his head and said with a smile.
"But our calculation load is still a bit large. After counting the small ones, we have to count the big ones; after counting the ones lying on the ground, we still have to count the ones flying in the sky."
For complex calculations, although the 8-bit DJS-59 and 60D can achieve 30-bit data calculation through clever programming, their speed is definitely not very high.
Especially when there are a lot of floating point calculations, the time consumption skyrockets.
So while scientists are glad that working conditions are good now, they can't help but hope that it will be faster.
If we could finish it one day earlier and devote our energy to other equally urgent tasks one day earlier, would it be possible to bring greater changes to society?
What they don't know is that their work is actually much ahead of schedule compared to before.
"Yes, it's a pity that these calculations are in order. If the previous result cannot be calculated, the next one cannot be calculated, and you may not even know how to calculate it. In this case, it is useless no matter how many computers you have, all of them can be calculated at the same time. There are only a few routes to take.”
Computers are only auxiliary tools. When routes are limited, the overall scale of the heap's computing power is of little significance.
Several people were working when a scientist rushed in with a piece of paper.
"Old X, look and see if this thing is useful to you!"
He was holding a piece of printing paper in his hand. Judging from the format, it was a scientific and technological intelligence bulletin issued by the National Defense Work Committee.
Knowing that his comrade would never target someone without aim, a scientist stood up and took the paper from the other person's hand.
"Hahahaha, I want to doze off, here is the pillow! Take a look, come, take a look." Looking at the content on the paper, he burst out laughing.
Several other comrades stretched their heads curiously and looked at the report.
Then they all laughed.
"Haha, it's true! The comrades outside are working so hard."
"This speed is incredible!"
"That's 250,000 16-bit multiplications per second! Even if we count 32-bit multiplications by half, that's still at least 120,000 times!"
This statement is actually not correct. For single-cycle hardware multipliers, 16-bit to 32-bit cannot be easily converted. In other words, it is difficult to expand a formed 16-bit single-cycle multiplier to a 32-bit multiplier.
But they are not in this business, so they don't know this situation. They only know that they have new weapons.
"It's a pity that what we have to calculate are relatively large numbers with relatively high precision requirements. This 16-bit multiplier is still almost useless." The computer that originally hit the squirrel was a 31-bit computer.
"It's almost interesting, it's almost interesting. I'm not greedy. It's better than not interesting at all. The half-point can always be solved using mathematical methods, haha."
Mathematics, science and engineering despise the one at the top of the chain, or the most transcendent one.
"Old X, now that we have this technology, can you please report to the National Defense Work Committee and ask the comrades who are doing this to build us a 32-bit hardware multiplier?" A scientist was so blessed that he made a suggestion.
"Isn't it bad? Does it affect the work of other comrades too much?"
"Try it, there's no problem if you try it."
The National Defense Work Committee quickly received a request from Da Mo, asking about the possibility of a 32-bit hardware multiplier.
Looking at this request, the National Defense Working Committee has mixed feelings.
The good news is that this thing seems to be of great use over there.
What worries me is whether Gao Zhendong is too busy, not whether he can do it, but whether he is too busy.
But no matter what, we must try our best to meet the requirements there. This is not just a matter of "living up".
When Gao Zhendong received a call from his leader in the confidential room of the No. 3 Factory, he learned this was such a request.
"Use the fastest speed to create a 32-bit single-cycle hardware multiplier."
The leader's words were very direct, there was no room for discussion, and he was not afraid that Gao Zhendong wouldn't be able to do it. He believed that Gao Zhendong would be fine.
Gao Zhendong was also unambiguous: "Resolutely complete the task and we will see the result within 20 days."
For floating point numbers, the operation is actually divided into three parts.
The sign bit is XORed and the exponent part is added, both parts are very simple.
The remaining troublesome thing is the multiplication of the exponent part. This is the most troublesome. In addition to multiplication, there are also rounding and other issues to consider.
Fortunately, the algorithm principles are roughly the same. From 16-bit to 32-bit, the main reason is the increase in scale. Some logical operation structures can expand the number of operations through series and parallel connections, which saves him a lot of work.
After listening to Gao Zhendong's answer, the leader put down the phone with confidence. Well, if Gao Zhendong said 20 days, then it must be 20 days. It will only be advanced, not postponed.
Gao Zhendong did not engage in the IC-based work of the originally prepared hardware multipliers, but directly focused on the 32-bit hardware multipliers.
He could vaguely guess where this request came from.
Even if he doesn't come from there, the urgent attitude of the National Defense Work Committee shows that this matter is very important and worthy of his own attention.
Gao Zhendong not only wants to make it 32-bit, he also plans to maximize the frequency as much as possible. Speed is always not enough no matter how fast it is.
——
While Gao Zhendong was busy working on the 32-bit multiplier, the Guidance Research Institute had completed the personnel organization for portable air defense missiles, and several comrades came directly to come here.
They are all acquaintances, some of them even very familiar.
Gao Zhendong was a little confused when he saw the person in charge of the guidance system, Lin Lianwei.
Seeing Gao Zhendong's appearance, Lin Lianwei laughed loudly: "Old Gao, you didn't expect it to be me, did you?"
In fact, it is not surprising. Lin Lianwei came from an automation major and was the person most familiar with Gao Zhendong and the Guidance Research Institute. He learned things the fastest and most conveniently, and the design of the two generations of seekers, Thunderbolt-1 and Thunderbolt-3, He participated in the whole process, worked hard and has rich experience.
Therefore, it is perfect for him to be responsible for the guidance system of this small missile.
Gao Zhendong smiled and said: "I didn't expect that you were going to be the leader of Thunderbolt-3."
Lin Lianwei shook his head: "With your help, that thing was completed very quickly. It is almost finished. Mr. Lu felt that it was a waste for me to be here, so he recommended me to the hospital and gave me more responsibility."
< br>
Gao Zhendong smiled: "Well, that's good, this thing, the most important thing is this seeker."
Everyone exchanged pleasantries, and the chief engineer of the model selected by the Institute smiled and said: "Mr. Gao, we have also made a preliminary plan. Do you want to take a look first?"
Gao Zhendong nodded: "Okay, let's communicate."
Regarding infrared air-to-air bombs, the Guidance Research Institute, which has rich experience, shows typical path dependence.
The plan is basically a smaller version of the Thunderbolt-3, the same, with built-in refrigerated infrared orthogonal four-element detector, continuous rod warhead, gas rudder...
Gao Zhendong laughed when he saw it.
(End of chapter)