Huaguo does not know that in the future in the field of chips, the division of labor will become more and more detailed, and semiconductor manufacturers will gradually specialize in a certain field.

The original vertical manufacturers, semiconductor manufacturers single-handedly handle everything from chip design to chip manufacturing, and only Samsung and Intel remain.

Throughout the 90s, Huaguo's benchmarking learning objects in the semiconductor field were first neon companies, and then Korean enterprises.

Samsung to be exact.

The emergence of Zhou Xin gave Huaguo another option, that is, to separate chip design and chip foundry, and other technologies in the upstream and downstream of the design, manufacturing, and supply chain were supplied by different enterprises.

As a big country, Huaguo will not give up its original strategic line and cultivate a semiconductor company that takes into account both design and manufacturing. But they will not give up on exploring new models, especially Shenhai City, which hopes to replicate the success of Silicon Valley in Zhangjiang.

According to Zhou Xin's experience, there will be a technical advantage in one-handed design and manufacturing, and this advantage will become more and more obvious when the process comes below 22nm.

Intel has mastered both chip design and manufacturing processes in the field of 14nm, which is an important reason why they can make the most powerful CPU with a single core performance at 14nm.

Vertical manufacturing has technical advantages, but the development efficiency is not as good as the commercial division of labor model of chip foundry and chip design.

In the future, Intel's chip manufacturing department has always been in a leading position in the process, especially in the 22nm process in 2012, far surpassing TSMC and Samsung.

This vertical manufacturing model gave Intel strong confidence, but after 22nm, the 14nm process was delayed for half a year, and then Intel's process skipping became the norm, and then it was to outsource chip manufacturing to TSMC.

So much so that in the end, only Samsung was able to contract chip foundry and chip design in the most advanced process, and it still persisted.

Chip manufacturing has reached the 3nm era, and Intel needs TSMC's foundry capacity support.

As strong as Intel, which has taken advantage of technology in the first decade of the 21st century, it will also lag behind because of the path dependence of technology research and development. Therefore, Zhou Xin is not too sure about this path.

Intel's later search for TSMC foundry was not a sudden decision, but the result of long-term accumulation.

There are two main situations in which Intel is looking for TSMC foundry, one is the department it acquired at the beginning, such as Infineon's wireless department, which originally had some RF circuits using the older TSMC 28nm process, which was required by the original plan, and there was no need to transfer orders for the sake of transferring orders.

The other is a problem caused by Intel itself. Capacity planning is a long-term interlocking process, and problems in one of the links will have an impact on the future.

Build a factory, prepare raw materials, train engineers, adjust the machine, and after the technical department debugs the process formula, start small-scale trial production. In the process of trial production, on the one hand, the yield rate is adjusted, and on the other hand, the chip design team starts the design one to two years in advance.

Finally, at the agreed time between the chip foundry and the chip design, the design team hands over the design plan to the wafer factory, and begins to slowly do the mask layer by layer, and finally mass-production, packaging, testing, and shipment.

Under the original plan, Intel's 10nm process research and development went smoothly, the 14nm factory was mass-produced, and the 10nm factory was slowly tested and ready, and when the yield rate reached the standard, a large number of orders could be started. The chip design team has also basically completed the design on the 10nm process.

For Intel, after the 10nm process is realized, which chip design department will be prioritized? Graphics processors, because the architecture of graphics processors is consistent, the design cycle is short, and the design blueprint can be quickly produced. Then the CPU and Server began to produce 10nm chips one after another. 10nm production continues to increase, and 14nm slowly frees up capacity.

Among Intel's 100,000 employees, in addition to the core departments of TMG/CPU/Server, there are many peripheral organizations affiliated with Intel that also need chips.

The vast majority of these peripheral organizations are collectibles that Brian, the former CEO of Intel, squandered his old money to buy. Like Altera, an FPGA company that was originally a little famous but gradually forgotten, bought Infineon's wireless division in order to return to the mobile phone market, and caught up with the artificial intelligence Nervana, which was bought expensively, and autonomous driving Mobileye.

Excluding these large luxury goods, there are also some basic IPs, IOs, memory, and small CPU Atoms. Finally, there are some peripheral organizations that are not large and can be sent by an official car. Or a group that purely does research and publishes papers, and helps TMG to do the advanced chip design organization of test chips. In short, there are many mountains inside Intel, and there are no surprises.

What's more, there are also some wandering samurai organizations that have been silently floating in Intel for decades for some reason. When Zhou Xin interned at Intel, a senior in the industry he knew collectively referred to these groups as children raised by stepmothers that no one cared for.

In fact, these itinerant samurai organizations used to be subordinate to regular organizations. It's just that these formal organizations have been disbanded, like Intel's whim-based wafer foundry, and the leaders of the design department have to help people work within Intel for the survival of their employees.

For example, the server department is not enough to make a memory control IC, so temporarily let the wandering samurai take over this job. If you face too much competition, such as Bangalore, India, who uses crowd tactics and likes to boast that everything is okay and snatches away the job, or encounters a Chinese engineer who works in Bengcheng, Malaysia, and dies, then you can only touch your nose and pick up some leftover jobs that others have picked up and have no future to see.

Intel also helped ZTE make chips, and these groups that helped ZTE make chips were wandering samurai organizations. In short, Intel is such a world of the jungle.

These various internal organizations decide whether to enter 10nm or continue to use 14nm excess capacity according to the importance of their contribution to the company's profits.

As a result, Intel's 10nm dystocia, and the chip design team idled. More importantly, what about CPU/Server/GT? had to continue to occupy 14nm capacity.

The worst situation is coming, the unit responsible for process research and development in TMG is TD (technology development). TD is already in the limit state of everyone's overwork, and in the past, 14nm was done, and the main force went to 10nm. Leave a handful of engineers to maintain 14nm, and then split off a small group of the elite to go to 7nm. Now 14nm will continue to be engaged, and 14nm+, 14nm++, 14nm+++ will be engaged in it, which is the squeezing toothpaste that Intel criticizes.

The number of people who are allocated to 10nm is less, and the number of people who are allocated to 7nm is even less. Forcefully let yourself fall deeper and deeper. Leads to a wrong step, a wrong step.

And the 14nm production capacity is given to the company's lifeblood CPU/Server, so what about other departments? The days still have to pass, the IC still has to be shipped, in order to survive, everyone has made difficulties, and the relatively powerful Infineon wireless department jumped out and said, we are using TSMC in the Infineon era, happy cooperation, let us continue to use it.

The AI team said that our AI chips can't wait, how much data is waiting for us to train, and Nvidia has thrown us several streets. So we must use the best process, and give me TSMC's 10nm without Intel's 7nm.

Some of the edge departments that help the server system make peripheral chips began to speak loudly, not shipping to us, and the server can't be shipped, everyone is going to die together.

At this point, Intel has no other choice, and opening orders to TSMC is the last resort. As a listed company, Intel has the obligation to make profits to shareholders, but making profits does not mean that it must rely on itself to produce chips.

As for Intel, after it began to place orders with TSMC, it had serious side effects on its own company system, and TSMC began to be used in the production of CPUs inexplicably.

Zhou Xin knows this period of history very well, Hu Zhengming was originally going to TSMC as chief technology officer, and later TSMC's chairman Liu Deyin was just Hu Zhengming's follower in the first decade of the millennium.

To be precise, Zhou Xin has a very good understanding of TSMC's history, and TSMC's 7nm process foundry was forcibly pushed on the horse at the beginning.

At the beginning, it was completely to grab orders with Samsung, and according to the OEM price and the yield rate at that time, even the cost could not be recovered, and it was completely relied on government subsidies to smooth out the losses.

Of course, the advantage is that after surviving the initial stage, relying on market monopoly, TSMC's gross profit margin is close to 50%, much higher than that of Samsung, which only does its own chip design. Samsung also has a small number of chip foundries from Qualcomm and Nvidia, but Nvidia and Qualcomm are particularly strong in price reduction.

TSMC and Samsung have government backing behind them, and the government would rather give high subsidies than let them fall.

Intel is even more so, even if Intel loses money in its own chip production chain, relying on TSMC and Samsung to achieve higher profits, they are unlikely to give up chip foundry.

Because Samsung and TSMC are located in Goryeo and Wanwan, which are too close to China, if there is a cold war in the future, these two regions are on the front line of confrontation, and the risk is too high, and Ameriken will lose the source of chips.

This is also why Ameriken has to use various means to either force or induce TSMC to build a factory at Ameriken. It's just that TSMC was stunned for a while, and Samsung relied on procrastination to delay indefinitely.

From a long-term perspective, Zhou Xin prefers the mode of separating chip foundry and chip design. is as strong as Apple, and it also has to hand over the chip foundry to TSMC.

Intel, the only one that has persevered, has a strong advantage in the entire ecosystem, and most of the software in the server, such as operating systems, compilers, databases, etc., is all developed on the ecology of Intel chips.

These require a long period of running-in and debugging to reach a stable state.

If you change to another company's CPU, theoretically it should be directly compatible without any problems, but in practice, it will definitely need a lot of debugging to reach the production state. So unless the CPUs of other manufacturers, such as AMD, are too good, there is no incentive for the company to replace Intel's CPUs.

In the future, some countries will replace Intel for national security reasons, which will expand the infinite prospects for their own IT manufacturers.

Therefore, if you want to adopt Intel's model, Zhou Xin will face two problems, the first is that it is not as successful as Intel to take this road, and Zhou Xin also has no self-confidence.

Without a system, it is impossible to control all nodes from chip design to chip foundry by relying on the power of one person, let alone climbing upstream lithography machines and high-end materials.

There are too many nodes involved, and it is most feasible to adopt the method of innovative enterprise clusters.

The second problem is that no matter how good Zhou Xin is in the semiconductor field, he can't convince all Chinese companies to engage in domestic equipment and domestic operating systems to replace America's IT software and hardware at this point in time.

Therefore, the government appropriately tilts resources and cultivates a number of competitive enterprises in the field of integrated circuits by marketization, and these enterprises can occupy a certain market share in the upstream and downstream of the semiconductor industry chain by their own capabilities before 18 years.

Then, after 18 years, under the changes in the macro environment, these enterprises will rely on the Chinese market to grow rapidly and replace the foreign industrial chain.

This is also why Zhou Xin values Zhang Jiang very much.

In the original time and space, Zhangjiang did not play a planning role, and there was a gap between the planning and the planning when the Zhangjiang Hi-Tech Park was established.

Zhou Xin hopes to rely on the model project of Xinxin Technology and financial advantages to attract talents and industries to form a scale effect.

Xinxin Semiconductor and Xinxin lithography machine are just the first step.

Matrix is the second step, and Apple can improve performance in the field of mobile phone chips regardless of the cost by relying on the high profit margins of mobile phones. Because in any case, they were able to make a profit through the brand premium of the iPhone.

Qualcomm needs to consider the cost of designing chips, and even Android phones with the most advanced manufacturing process cannot compete with iPhones.

In addition, in the field of chip manufacturing, wage costs account for only a fraction of revenue, and the most expensive cost comes from equipment depreciation.

Only by relying on the success of matrix and forming high profits, we can rely on high profits to improve performance in the chip field.

At this stage, SMIC will not be stuck by foreign countries, and can catch up with the most advanced processes abroad under the tilt of orders.

After SMIC catches up with the most advanced foreign manufacturing process, it gradually allows the domestic chip supply chain to enter the SMIC supply chain, starting with testing.

In this way, a positive cycle can be formed.

When Zhou Xin went to Zhangjiang for research, he thought that after he did this, a large number of semiconductor companies located in Zhangjiang would definitely cause Zhangjiang's housing prices to soar faster than expected.

In later generations, even people like Zhou Xin who have achieved a small business will look at the housing prices of the magic capital and grit their teeth, feel that it is not worth it when they buy it, and feel that something is missing if they don't buy it.

The progress of Xinxin Technology in Zhangjiang is smoother than Zhou Xin's expectations.

Different from the smooth progress of Xinxin Technology, Guan Jianying was very unsmooth in the process of promoting cooperation with SMIC.

According to the original time and space, SMIC's 250nm process will be trial-produced in September 2001 and mass production will begin in the first quarter of 2002.

In other words, the current 220nm prototype of the new chip lithography machine can fully meet SMIC's requirements.

Guan Jianying, as the general manager of Xinxin Technology, will also be responsible for part of the business of Xinxin lithography machine, in his opinion, the emergence of a new chip foundry enterprise in Huaguo, or the industry's famous Zhang Rujing foundry.

There are naturally a lot of complementarities between Xinxin and SMIC, and Xinxin lithography machine can fully cooperate with SMIC, and the testing of Xinxin lithography machine can be carried out in parallel with SMIC's factory construction.

And at that time, Zhou Xin and Zhang Rujing also had a good conversation when they were in Shenhai, and both of them said that they would strengthen cooperation in the future, especially in the field of lithography machines.

As a result, after Zhou Xin returned to San Francisco, Guan Jianying wanted to push the matter, but SMIC played all kinds of tai chi and did not accept it at all.

Zhou Xin lamented that there were still too few chips, and even SMIC couldn't handle it.

Unlike the outside world thinks, SMIC relies on high salaries to poach people from TSMC, and the treatment given by SMIC back then was only about 70% of TSMC's. Most of the engineers who chose to go to SMIC were marginalized engineers at TSMC.

Zhou Xin roughly knows why SMIC is reluctant to cooperate with Xinxin lithography machines, and the fundamental reason is that Xinxin lithography machines are not mature.

For SMIC, which wants to put into production as soon as possible, where can I have the time and energy to help Xinxin Semiconductor test prototypes?

Zhou Xin thought to himself, it is still necessary to find an opportunity to control SMIC, although it is divided into different companies to do this, but the chip foundry and chip design are separated, and now the impact is not great, and when the 22nm process is below, it will affect the performance.

After the 22nm process, the chip design must be done together with the chip foundry.

The quality of the chip does not depend only on the process. The most typical case is that Nvidia defeated AMD's TSMC 7nm chip (GTX 2080 vs Radeon VII) with TSMC's 14nm process, with higher performance and even lower power consumption, which is the charm of architecture design.

After 22nm, the design rules of chip manufacturers will not tell chip design companies, TSMC helps HiSilicon foundry, and it is impossible to tell HiSilicon how to set the layout of the 7nm chip and NAND gate, and where to place the MOS of the inverter on the clock line.

This belongs to the unique secret that chip manufacturers have piled up by experience, and it belongs to TSMC's housekeeping skills. Therefore, Huaguo's chip design company will only give TSMC a general style, and then add a logical sequence.

These logic sequences need to be taped out multiple times to be realized, HiSilicon is responsible for the general direction of the chip, TSMC is responsible for the general direction to make adjustments, how to design the specific layout, and how to do the mask.

The effect is naturally not as good as Intel and Samsung, and it can start from the bottom.

This is also why Qualcomm and AMD's chips are not as good as Samsung and Intel's chips.

70 billion seems like a lot, but in fact, it is not the same thing at all when you really want to operate.

When the Nasdaq bubble burst, Zhou Xin acquired ARM and Nvidia, in fact, Zhou Xin coveted Applied Materials the most.

Applied Materials has a monopoly on almost all equipment except lithography machines, but these equipment are not as well-known as lithography machines.

But even if the bubble bursts, Applied Materials has a market value of $100 billion, and it would have to prepare at least twice as much capital to hold it.

Of course, the acquisition is not very useful, and it is impossible for Ameriken to watch you move the R&D and production of applied materials to China.

"Newman, I recently came to the school, and I have some details about your paper that I need to communicate with you." Chenming。

"Okay, Professor, I'll get there as soon as I can." Newman。

In addition to the promotion of enterprise qq in Riot Games, matrix and NewPay, academic organizations such as CAAS have also purchased enterprise qq.

When the enterprise QQ was first launched, members thought: What is the difference between this thing and ICQ?

However, considering that most of CAAS's funds are from Zhou Xin, the purchasing company qq has no objections, and it is also Zhou Xin's own money anyway.

Moreover, in order to expand Tencent's business, the enterprise qq was sold to CAAS at a very low price.

At the beginning, most members were not accustomed to using corporate QQ to communicate, and everyone was still used to email.

One is that you have developed a habit after sending and receiving emails, and the other is that you can write back to emails intermittently, and you don't need to reply to others right away.

Instant messaging apps are different, and communication on instant messaging apps is a continuous conversation, and it's not good to stop for too long in the middle of a conversation before replying to someone.

When enterprise qq was first launched within CAAS, few people used it. It wasn't until the first anniversary of the establishment of CAAS that Zhou Xin used enterprise qq to pull a few groups that this software began to be used slowly.

The most used function is to create a group, and then upload some papers in the group for everyone to discuss.

The members of CAAS called this a discussion group, which was mainly a discussion group for the topic, and later there were various eating, drinking, and entertainment groups, and local dating groups were established.

Sometimes the promotion of software needs an opportunity.

This does not include Hu Zhengming, who has liked corporate QQ very much from the beginning, because he can finally communicate with others at any time without thinking about the email format.

Therefore, the most frequently used enterprise qq is not Zhou Xin, but Hu Zhengming, too many friends who join CAAS can communicate their feelings through enterprise qq.

At present, only enterprise QQ is a social software with real-name system for acquaintances.

After the enterprise QQ entered the Amerit, the personal QQ was also menacing in Americon.

NewPay's second round of financing, the main participants are Wells Fargo and Citibank, the investment banking divisions of two mainstream banks in Ameriken, in addition to banks, Walmart.

As the only non-financial institution, it invested in NewPay, which became the biggest positive for Walmart's millennium capital market. It jumped 7% on the day the news was officially announced.

After Wal-Mart invested in NewPay, Wal-Mart purchased enterprise QQ from Tencent, and Wal-Mart deployed local enterprise QQ on Americon's large and medium-sized city servers.

"Walmart can treat the group as an online release board, invite users to join the group, and regularly distribute some preferential information and limited-time discount information.

Not only can you send you special offers, but you can also send you information about the latest items on sale. ”

This is also the key point for enterprise QQ to attract Walmart to purchase enterprise QQ.

Compared to offline flyers, it is much more convenient to rely on group chats to send messages.

Of course, there will be a problem, that is, the cost of communication is too high, because there are too many customers in Walmart stores, you can't help customers join the group one by one, and now there is no QR code to scan the code to join the group.

In addition, the number of computers is limited, and it is impossible to put some computers in each store to help users join the group, which is very costly.

At present, Wal-Mart can only print the group number under the receipt and hand it to the user, and the cashier will remind it.

At the same time, advertisements are placed in the media to remind users to join the group to receive benefits.

This strength is too weak and the conversion rate is limited.

Even after users join a group, they rarely follow the group messages in real time.

Therefore, the outbreak of the use of enterprise QQ in the user group had to be established on Thanksgiving Day in December, and the situation was improved after the online supermarket built by NewPay and Walmart was launched.

After selecting a specific offline store on Walmart Online Supermarket, the user completes the online payment with NewPay after purchasing the product, and will receive a text message on the mobile phone number registered with NewPay after the online payment.

The content of the SMS contains the order number, mobile phone number, and SMS verification code.

The user takes the mobile phone to the offline store, reports the order number to the special online service store staff, and they will verify the mobile phone number and verification code before directly packaging and handing the product to the user.

That is, online shopping and offline pickup.

This model of online ordering and offline pick-up did not become popular in Europe until 2015.

Walmart was the first to introduce this model, and the new e-commerce model made Amazon feel like a big enemy.

As a result, their vigilance was correct, and after Walmart's online ordering and offline pickup were launched, Amazon was the first to be impacted.

Because they will choose to shop online, they are going to save trouble, and the online shopping of Americon in the millennium is not cheap, and it is even more expensive than offline on many goods.

Online shopping, on the other hand, is also hassle-free, and there is no need for shipping, you can buy it before you go out, and make a trip to Walmart on the way home from work, and you don't need to buy what you want to buy in five minutes.

In the millennium, Walmart's offline stores can radiate much wider than express sites.

"Newman, the idea you gave Walmart is worth at least 5% of the shares, and I'm a little jealous of Walmart.

This plan is enough to support a 10 billion Internet company, and before 2000, 100 billion is possible. After Walmart's online shopping function was launched, Bill Gates complained when he was having dinner with Zhou Xin.

"For Wal-Mart, they have a new growth point, which gives users more choices in terms of business form, and also allows them to open up the gap with other supermarket brands.

The advantage of online Walmart will last for at least half a year. If other supermarket brands can't access NewPay, the advantage will be longer than half a year.

If I were you, I would choose to make this product myself, and cooperate with all supermarkets, you are responsible for the purchase and payment on the Internet, and the supermarket is responsible for the offline supply. ”

In order to obtain this idea, Walmart signed a long-term exclusive cooperation agreement with NewPay, which will be Walmart's only electronic payment channel for the next five years.

If it weren't for this condition, Zhou Xin would not have helped them design this mechanism.

Because Walmart is a listed company, this kind of exclusivity agreement will have to be announced sooner or later, so Zhou Xin directly explained it to Bill Gates, and then continued:

"The business model of online ordering and offline pick-up is very simple, and the only threshold is electronic payment, but at present, in addition to NewPay, there are several electronic payment options on the market.

And even if they choose NewPay as a partner, NewPay will not refuse, so that the advantages of electronic payment are gone.

This business model requires strong control and influence, I need to control the update of goods in each supermarket, I need to connect with different supermarkets, and I need to ensure that users can pick up the goods offline, which is related to warehouse management and goods supply.

And then there are 51 states in the United States, countless supermarkets, my reputation is too big, to do this, I'm afraid that even the San Francisco Bay Area has not been occupied, and a bunch of Internet companies doing the same business model have popped up.

Not to mention that traditional supermarkets can also do this business. It's like an English dish that looks like a delicious cake but is actually hard to swallow.

Only a company like Wal-Mart has the control and confidentiality to make the whole United States go online at the same time.

And before the launch of the commercial, only the wind came out, and there was no more specific news. ”

Before the launch of online Walmart, there were rumors that Newman and Walmart had reached an in-depth cooperation to help Walmart design an app.

This news is especially popular among Walmart investors, who swear: "If Newman does not cooperate deeply with Wal-Mart, why would NewPay accept Wal-Mart's investment?" Apparently Newman saw the potential at Walmart. ”

Later, it was confirmed that Zhou Xin did help Walmart design an online Wal-Mart, and sometimes the news of the secondary market is true, but the timing of retail investors getting the news will lag behind.

In an interview with reporters, the CEO of Wal-Mart claimed: "Yes, the online Wal-Mart was designed by Newman for us, he is a genius in the product field, Wal-Mart is honored to cooperate with Newman, and the online Wal-Mart achieved a revenue of 1 billion in the United States in the first month after its launch, which is a very exaggerated figure." ”

Wal-Mart's revenue in 2000 was only $165 billion, and only about $150 billion in the Ameriken area.

As Ameriken's largest e-commerce website, Amazon's annual sales in 2000 were only 5.8 billion.

After the announcement of the interview with the CEO of Walmart, Amazon's stock price continued to plummet, and Amazon, which was originally affected by the bursting of the Nasdaq bubble and lost 80% of its stock price, shrank further.

Online Walmart's revenue in the first month was 1 billion, and the annual revenue of the first launch is expected to far exceed Amazon, so why do investors still invest in Amazon?

Walmart felt that it had made money, not only acquiring nearly 10% of NewPay's shares, but also successfully finding a second growth point for the company.

It is very difficult for traditional enterprises to integrate with the Internet, and it is even more difficult for traditional retailers.

It is not a loss for Zhou Xin and NewPay, with the two payment scenarios of Walmart and Riot Games, NewPay can ensure that NewPay is invincible.

When the mobile Internet matures, NewPay will firmly occupy the first place with the help of hardware.

Since today's mobile phones do not have a browser function, they can only confirm the payment through the verification code, and there is no way to scan the QR code.

If you have a browser, you can jump to the webpage to display the QR code through the SMS link.

2G networks can do this.

Therefore, Zhou Xin is extremely eager to look forward to the launch of Matrix, and when Matrix is launched, it will be more convenient to shop online and pick up goods offline.

The success of online Walmart is just another verification of Zhou Xin's aura.

This has led to Bill Gates' obsession with NewPay, and with NewPay's power in various channels, Microsoft is bound to accept the electronic payment function after the Xbox is launched.

Bill Gates has fully recognized the importance of NewPay, which is by no means as simple as online banking.

I know the progress is slow, I can only add more to make up for the progress, today it is more than 8K!

Within the crow's ability, the crow will try to update as much as possible.

(End of chapter)