Jiang Bo's eyes swept around, and he only saw that in the huge workshop, hundreds of technicians were busy, and the sound of machinery and equipment was endless, but he didn't see the device in his mind.

Jiang Bo couldn't help but look at Ke Peng and wondered: "Didn't you say on the phone that the new fullerene device has been built, where is it, why didn't I see it?

The corners of Ke Peng's mouth twitched twice, he stared at Zheng Shouyi resentfully, and coughed lightly: "That boss, the thing has indeed been made, but Chief Engineer Zheng said that it was not qualified, and it was a defective product...... And let people tear it down. ”

Jiang Bo's brows moved slightly when he heard this, and he looked at Zheng Shouyi: "Professor Zheng, what's going on?

Zheng Shouyi's eyes were calm, and he replied: "There are some problems with the device on the chip, the output is low, and it is still far from the qualified indicators of the first generation, so I will let people dismantle it, and plan to re-examine the technology and doubts from beginning to end to see where it has been ignored." ”

"What's wrong with the chip?" Jiang Bo was slightly stunned.

Zheng Shouyi looked at Ke Peng, helpless, Ke Peng could only bite the bullet and explain the situation again.

Zheng Shouyi asked the chipset people in the workshop to do the manufacturing of ultra-microelectronic chips of fullerene devices, and the chipset people turned to Ke Peng for help.

After Ke Peng informed Jiang Bo of the matter, Jiang Bo asked him to go to the foundries of Ziguang Group or Huaxin for OEM.

Ke Peng didn't think it was so difficult, his plan was to let the Unigroup OEM the chip of the fullerene device, but sadly found that even if TSMC and Inter came, the chip of the fullerene device could not be made.

Because the process of the chip has reached 2nm, it cannot be made with the most advanced ASML extreme ultraviolet lithography machine.

Therefore, Ke Peng could only retreat and let the Unisplendour Group use the 7nm process to make the chip, although the chip can still be used, but the performance is much worse.

I didn't care about it at first, and I always felt that it wasn't the key, but as soon as I tested it today, the problem was revealed.

When Jiang Bo heard this, he said, "I still don't understand, the chip performance is not enough, and then, what's wrong?"

Ke Peng looked at Zheng Shouyi, and Zheng Shouyi explained: "The performance of the chip is insufficient, and it can only be tested in a specific software program.

This has to start with the numerical control system, the CNC system compiled by the headquarters is mainly divided into two parts.

The first part is the precise regulation and control of nine links, such as ultraviolet laser pulse equipment, vacuum generator, electromagnetic gauge field, and refrigeration system.

The second part is the control of all other drive units.

Among them, the first part is the regulation of ultraviolet laser pulse equipment and electromagnetic gauge field, which is the key among the key.

Ultraviolet laser pulse equipment aside, we use German goods, after a small modification, this thing except for power, other data are up to standard.

The CNC system only needs to control the appropriate output frequency according to its peak to prevent excessive power from damaging equipment parts. ”

Zheng Shouyi paused, and said in a deep voice: "But the electromagnetic gauge field is not easy to organize, a single pulse power of 100 femtoseconds laser pulses, after bombarding the raw materials, the molecular bonds of a large number of raw materials are directly destroyed, forming carbocations and carboaniones."

At this time, under the control of the numerical control system, the electromagnetic gauge field will accurately normalize this part of the charged carbon ions through an identification method.

Then, by bombardment with secondary laser pulses, they are blasted into molecules composed entirely of carbon with different structures, including fullerene C60.

Of course, this process also synthesizes many other useless carbon molecules when the pulse width of the laser pulse does not reach the specific synthetic pulse width of fullerene C60. ”

Jiang Bo bowed slightly, these are relatively basic knowledge in technical information, but he understood, and he didn't interject, and continued to listen.

Zheng Shouyi said: "The identification unit of the numerical control system for the identification of carbon ions is very costly, although it is only a rough identification and screening, but it is still 1Tflops per second, that is, 1 trillion times per second."

Originally, the chips manufactured by the 2nm process had dense integrated circuits, and the area of the entire chip was tens of square centimeters, the size of a palm.

However, because the chip was replaced with a 7nm process, the surface area of the chip was increased by nearly 20 times, and the performance fluctuated greatly.

Under high-speed computing, after the performance reaches its peak, the actual operation speed is also reduced by more than half, resulting in insufficient number of carbocations and carboaniones identified and standardized per unit time, which directly affects the synthesis of C60 molecules.

This is the reason why I asked people to analyze and find out after the quiz just now. ”

"Meaning, there is a problem with the operation of the chip?" Jiang Bo was stunned.

"Yes. ”

Jiang Bo breathed a sigh of relief and said with a smile: "I thought it was a problem, isn't the calculation speed trivial, since the computing speed is not enough, then increase the portrayal area of the integrated circuit in the chip and increase the number of transistors in the chip."

In other words, if multiple chips are directly connected together, isn't the technology in this area already mature, so that the computing speed can be improved?"

Zheng Shouyi nodded and said, "What you said, we have taken it into account, if you do this, the calculation speed can indeed reach the standard."

But when you didn't come just now, we found that even if the calculation speed is up to the standard, or even exceeded, it can only increase the output by two times at most. ”

Jiang Bo smiled: "Twice is already very good." Oh yes, with all that said, I don't know what the output of that device is like, so it shouldn't be too low, right?"

"The daily shift is 10 hours, the night shift is 10 kilograms, and if you don't take a break, you can produce up to 12 kilograms a day. ”

"Is it 10 kilograms per day...... Jiang Bo touched his chin and calculated, "Actually, it's not too little, what is the cost?"

Zheng Shouyi looked at Ke Peng when he heard this, Ke Peng was doing statistics.

Ke Peng hurriedly replied: "The cost of a device is about more than 10 million, less than 20 million, if you increase the scale of production, wait until the output is more, you can further share the fixed costs, and reduce the price of raw materials, so that the cost will be reduced." ”

It's a well-known fact that economies of scale can reduce costs.

Listening to Ke Peng's words, the cost of a device is less than 20 million, and it can be repaid in ten days at most.

All told, although the output of the plant did not meet expectations, it was actually very good.

It's just that since Zheng Shouyi said that he was unqualified and wanted to do it again, Jiang Bo naturally hoped that the output would be higher.

Thinking of this, Jiang Bo asked: "In addition to the computing speed of the chip, what other problems restrict the output of the device, have you found out?"