When Yu Zixian arrived at the factory, he heard bursts of shouts coming from the large conference room on the second floor, mixed with some swearing, and then there would be apologies and assurances from Director Kong of the Sifang Electron Tube Factory.

This should be the scene of the conflict between the large and small color TV remote control circuit board purchasers and the Sifang Factory to persuade them to re-sign the agreement.

Yu Zixian sighed and shook his head.

When he went to Luo Shouwu's office, he found that the door was locked, and Yu Zixian guessed that they were most likely still in the guest house.

Came to the guest house again, in the small conference room on the third floor, Yu Zixian saw the people from the factory office and inquired, and he was really in the conference room on the third floor.

Yu Zixian didn't know what your situation was, so he had to ask someone from the factory office to go in and report it.

After a while, the factory office asked him to go in and report.

The conference room was filled with smoke and the windows were opened, but they still could not circulate and replace the constant flow of smoke.

Several veteran comrades sat scattered in chairs, all with cigarette rolls in their hands, some of them unlit.

"Director Luo, Uncle Niu, Uncle Qi, Uncle Jiang!" When Yu Zixian was drinking last night, he unknowingly recognized several uncles.

"Well, Zixian is here. ”

"Quick, seat. Jiang Lao pointed to the seat next to him and said.

Luo Shouwu rolled his eyes angrily, "What do you mean, is this trying to pull a relationship and preparing to abduct someone?"

"Zixian, I'm still going to look for you. Qisheng said earlier that you went to the laboratory, how about it, is there any good news? We old men want to be stinkers, but we can't come up with a feasible way! You have a flexible head, think about it?!"

"Well, at present, I think of a way that is not a solution, but I don't know if it will be timely? ”

"It's okay, you say, it's already like this anyway, and you can't keep spending it like this. Otherwise, we'll be finished. ”

"That's right, at present, Toshiba is cutting off the core chip of TM50536, and other chips can be replaced or other manufacturers. Liu Jian and I reverse-engineered a chip based on Toshiba's TM50536, but it has not been verified by EDA, and our domestic software does not have it. I contacted Director Wei Xianglin of the joint venture factory and made sure that they had this software at their headquarters, so I had to go to Xiangjiang if I wanted to verify it. Once validated, it can be put into production. ”

"Where is it produced? How can it be guaranteed?" Jiang Lao nodded at first, but then frowned.

"As long as there is no problem with EDA, we can return to production, find a research institute with the most advanced process technology in China for trial production, as long as we can manufacture it, we will definitely fight back against Toshiba's conspiracy." "Toshiba is unkind, and Yu Zixian can't be blamed for being unjust. Patents and other things will be put aside for now.

"It is impossible in China, although the domestic manufacturing has been able to reach 1.5 microns in theory, but only in theory, at present, in the actual process, the whole process has not been completely opened, so you can't do it for the time being. ”

"Are you sure?"

"I'm sure. Yu Zixian didn't know why Jiang Lao would answer in a very positive way, but he shouldn't be wrong when he thought about it. After all, I just made plans according to some things in my memory, and even some assumptions.

"Aren't the 405 and the 408 ...... in this regard? Originally, I wanted to produce in China, but now the domestic technology cannot meet the mass production requirements of the 1 .5 micron process, so in the end I can only go through the Hong Kong and Taiwan channels, in that case, the time will really be too late. ”

"However, I can go and ask and see if there is a way!?"

"Excuse me? This thing is there, and if it doesn't, it doesn't. Is it enough to ask?"

At this time, it was noon, and it was estimated that everyone didn't have the heart to eat lunch.

Yu Zixian's stomach began to growl, Yu Zixian had no choice but to whisper to Luo Shouwu to eat first, Luo Shouwu gave up and beckoned everyone to lunch. He is still young, and it is okay to resist the first resistance, but I am afraid that these old comrades will not be able to bear it.

During the meal, Wang Qisheng came again, saying that the latest news from the joint venture factory was to ask the joint venture factory to contact Toshiba's headquarters in Yanjing, Huaxia.

It seems that the time to come is still coming, and Toshiba is going to take the opportunity to show its fangs.

After eating, everyone went their separate ways again.

Li Donghong and Wang Qisheng, the joint venture factory, came forward to talk to Toshiba, and Yu Zixian also had to find a way to see how to realize the tape-out of chips as soon as possible. Luo Shouwu went to negotiate with the major factories to negotiate and solve the problem on the price as much as possible, at least to achieve the purpose of delaying delivery, and give the joint venture factory the maximum response time.

Yu Zixian returned to the laboratory of the joint venture factory again and went to find Wei Xianglin.

Wei Xianglin finally gave Yu Zixian a not-so-bad news: "The Xiangjiang headquarters has agreed to conduct EDA verification. ”

"Director Wei, if you go to Wanwan tape-out, how long will it take?" At this time, Wanwan can already do some OEM services. So Yu Zixian thought of a way for Lu's Semiconductor to come forward to do tape-out. It's just a matter of whether the time can be compressed.

Although Jiang Lao went to contact the production unit, it may not be successful, in Yu Zixian's view, the opportunity cannot succeed; here, although it will take a little longer, but at least it is a safe choice.

"As little as 2 months, as much as 3 months. It also depends on whether you choose 'Full Mask' mode or shared wafer mode (MPW). ”

Yu Zixian knows one thing, the cost of the full mask mode is very high, and it is said that the tape-out will cost 5 million at a time, and the price will not be low even now. The shared wafer model is much cheaper, but Yu Zixian is now estimated to be no less than 100,000.

In the IC design and production process, the designer will generally hand over the final result of the integrated circuit design circuit layout data to the foundry for wafer production, commonly known as tape-out (wafer casting).

The foundry processes the semiconductor wafer according to the integrated circuit layout, and the processing process is very complex, including the manufacture of transistors and interconnects through oxidation, chemical etching, ion implantation doping, metal deposition and other methods. layers) into a sheet of semiconductors, the principle is similar to the imaging and processing process of a film camera, in which a large number of various shapes of hoods are used, which determine the shape of the graphics at each level on the semiconductor plane, which is called Mask in integrated circuit manufacturing, commonly known as "mask" or "mask".

The cost of a single production of a chip can be simply characterized by the number of masks used, and the more masks used, the larger and more complex the chip, and the higher the cost. Correspondingly, Full Mask means that all the masks in the entire wafer manufacturing process are used for a certain chip, and it is obvious that the cost of this chip is very high. The Full Mask method is only used when the design is fully assured of success and is ready for mass production and commercial use, because mass production can reduce costs.

However, chip design is often very difficult, especially for new research topics, which may have to go through multiple silicon projects before entering commercial use. For many designers or units, it is usually difficult to achieve successful mass commercial use in a single wafer production, so there is a need for small-batch sample production, and there are universities, scientific research institutions and independent designers who do not need to commercialize chips on a large scale, but only need to produce a small number of chips to verify their design results, so MPW is generated.