Chapter 605: It is not recommended to use too much?

When the first "miscalculated" came out, some people's hearts skipped a beat, and thought, "Oh no!"

But as more and more "it was correct" were heard, everyone looked at the few comrades who said "it was wrong" with a teasing look in their eyes.

Well, comrade, did you make a mistake in your calculation?
Several comrades looked embarrassed and very ashamed.

But at this moment, everyone's attention was no longer on whether their calculation was correct, but on the circuit board full of tubes.

Baby!
Although a 16-bit floating point number has only 3 to 4 significant digits and a small numerical range, this precision and numerical range are sufficient for many practical engineering applications.

Let’s take the most typical and widely used example, Pi.

Thanks to the famous poem "A Temple and a Pot of Wine on the Mountain Top", I believe many people can easily recite Pi to 5 decimal places. Those who have the leisure time can easily recite it to 20 digits and beyond. Of course, the joke of "He is reciting Pi backwards!" is impossible.

However, in actual use, no matter how long the number is, 3.14 is often used. The reason is very simple. The correlation results calculated using 3.14 and 3.14159 are very similar in engineering applications and are generally sufficient.

If you do the math, you'll find that even if you use Pi to the ten thousandth decimal place, the difference from 3.14 is only about ten-thousandths.

This is why Gao Zhendong first made a 16-bit floating-point hardware multiplier - it is sufficient and cheap.

"Chief Engineer Gao, what is this board called?" a comrade who was working on radar asked excitedly.

Gao Zhendong smiled and said, "Single-cycle 16-bit floating-point hardware multiplier."

This name was very direct, and the comrades understood it immediately. Some of the comrades who knew more about computers exclaimed: "This machine can complete multiplication operations in one cycle?"

Gao Zhendong nodded. "Yes, its instruction cycle is the same as that of DJS-59. Four clock cycles form one machine cycle. Theoretically, even if the computer code of DJS-4 is optimized, the time required to complete an 59-bit integer multiplication is more than 8 times that of this hardware multiplier to complete a 16-bit floating point multiplication. In fact, since the main frequency of this device is 25MHz, compared with the 1M DJS-0.6 and 59M DJS-0.8D, its speed is even faster than what I just said."

Better integrated circuit technology made the hardware multiplier's main frequency faster than the previous two general-purpose computers from the beginning, and its potential was greater.

To put it another way, the theoretical maximum computing power of this hardware multiplier is about 0.25 MIPS, or 25 times per second.

1/25. Even without considering the difference in main frequency, this ratio makes all the comrades working on radar excited.

"This means that we can calculate signals faster!"

"Great! I estimated that with this, the speed of our moving target indication algorithm will reach a usable level."

Although the moving target indication they developed used the delay line cancellation filter suggested by Gao Zhendong, which can be implemented with analog devices, it does not mean that they do not need mathematical calculations when performing target processing.

"This is very useful for our digital communications. Signal processing, encryption and decryption, the real-time voice signal processing and real-time encryption problems that have been bothering us, finally have hope."

This is a comrade who was called over by the Commission's Chief Engineer to work in communications.

However, the matter was not over yet. Gao Zhendong said something that made them even happier.

"In fact, the main frequency of this circuit has not reached its limit. The main frequency I am using now is relatively conservative. Under the existing integrated circuit technology, the main frequency that this device can achieve should be increased several times."

PMOS can reach a main frequency of several MHz, which is the limit under current conditions. Although PMOS is limited by its own principle defects and is 1000 times slower than Gao Zhendong's speed of several GHz in his previous life, it is still very fast in this era.

The happiest ones were the comrades who were working on the artillery position detection and calibration radar. The algorithm Gao Zhendong gave them involved real-time calculation of the artillery position, extended Kalman filtering based on the calculated position data set, as well as the required Taylor series expansion and weighted averaging. These all used a large number of multiplication operations. The DJS-60D that was originally used was a bit stretched for calculations due to bit width and architecture issues. In simple terms, it was a bit slow.

It is difficult to process the shell position data sent back by the radar in real time. They now use the method of first storing the radar signal data, then calculating the position data based on it, and finally filtering and positioning according to the point data group.

In fact, the whole process took several minutes, and it took several minutes for the counterattack artillery group to bind the data and send it back. It was not unusable, but it was obvious that it could not achieve the effect they wanted.

After all, according to the information they had, the most advanced enemies had already used self-propelled artillery, which was capable of running away within tens of seconds after firing several rapid rounds.

Even if it is a towed artillery, in the hands of our skilled soldiers, it is actually possible to put it on a trailer and retreat within a few minutes after firing.

Therefore, compressing the solution time is their biggest wish. It is unrealistic to pile up DJS-60D, which is too expensive and complicated to program. Now, this wish has the possibility of being realized.

"Chief Engineer Gao, do you have any extra of this? Give us one to try."

Gao Zhendong smiled and said, "There are no extra ones now, but don't worry, this thing can be completely made in China, and you can have as much as you want. But I don't recommend that you use too much."

It is not recommended to use too much? What does that mean?
All the comrades who are working on radar are ready to go and are thinking about how to use this thing.

Convolution, series expansion, Fourier transform, Laplace transform, z transform, curve fitting...

There are so many places where it can be used. Just this small piece of circuit board can improve the back-end performance of the radar by at least an order of magnitude!

Then you told us that it is not recommended to use too much of this stuff? Keeping us on suspense, right?

Even the cheerful leader of the Defense Work Committee was confused by Gao Zhendong's words.

"Zhendong, is this thing difficult to produce? Or is it very expensive?"

In his opinion, these two reasons are not a big problem. If there are difficulties in production, then expand the scale. It is very expensive. This thing looks very similar to DJS-60D. If 60D is affordable, this thing is actually no problem.

So he was very curious about why Gao Zhendong did not recommend its use. In any case, it couldn't be because Gao Zhendong was reluctant to do so.

Gao Zhendong shook his head. "Neither. As I said earlier, there is no problem with self-production. As for the price, it is definitely expensive, but compared with radar, it is not too expensive. Generally speaking, it is still worth it. What I mean is that in my plan, this thing is just a semi-finished product, not yet in its final form. If you use too much, you will have to change the design when the final form is produced. It is okay to use a little, at least it is suitable for designing prototypes and the like."

Everyone then breathed a sigh of relief. So this is what was going on. I was scared to death.

"Chief Engineer Gao, you may think this thing is not perfect, but in our opinion, it is good enough. We really can't wait any longer."

"Yes, yes, we think this semi-finished product is good enough to use. Let us use it first."

The comrades present at the meeting were surprised at how high Gao Zhendong's demands were on himself, and at the same time they strongly requested that if you want to revise this thing, you can continue to revise it, but as for the current version, you should first make a batch for us, as time waits for no one.

"Well, Mr. Gao, we must have a batch. The electronic countermeasures you once told us about are really needed. We went back and carefully studied the countermeasures you gave us. They are all good, but we can't do most of them because the back-end processing technology is too backward. We can only drool at the good things. Now it's good. You have solved the biggest problem of high-speed processing devices. I don't care how imperfect you think this thing is. I think it can be used, haha."

——I don’t want to know what you think, I want to know what I think.

Electronic countermeasures have very high real-time requirements. No matter how many DJS-60Ds are piled up, they are useless in many situations. It is not a question of whether it is expensive or not.

I never expected that Chief Engineer Gao was not only responsible for providing solutions, but also quietly solving basic technical problems! The comrades who were engaged in electronic warfare looked very happy.

Next, many comrades made similar speeches, saying, "Mr. Gao, you are awesome. We think this thing is not good enough, but we are not. We think it is already very good and we strongly request that it be finalized and promoted."

Seeing the "excited" look of the comrades, the leader of the Defense Work Committee was very happy and said with a smile: "Comrade Zhendong, the comrades' voices still need to be addressed. They are very anxious. Let's finalize this thing and produce a batch first so that the comrades can use it first. This will not delay the development of technology and allow everyone to accumulate experience first. As for better ones, they can follow your improvements after you make them."

Everyone nodded: "Yes, yes, yes, we will ensure that the improvements will keep up."

Gao Zhendong couldn't say that this thing was in his plan. In fact, it was just a chip in the end. Since everyone strongly requested it, it was not impossible.

He nodded: "In that case, let's produce a batch..."

Before he finished speaking, the leader interrupted and said, "Give this thing a name. It cannot be placed in the 12th Ministry of Machinery for the time being. Just hang it under the Defense Work Committee. We will organize production here when the time comes."

He was wearing glasses and had good eyesight, so he could clearly see the bumps and grooves on the surface of the board.

He also realized that this thing should be built with the SZ61 series chips that Gao Zhendong mentioned. Once the circuit diagram and component list are obtained, it can be easily copied.

So he did not hesitate and directly intercepted the results. According to the rules of the Defense Work Committee, the confidentiality work should be easier than that of the 12th Ministry of Machine Building.

(End of this chapter)