Reborn in 2011, 985 missing from two copies

Chapter 208: Identity exposed, global sensation!

To develop EDA industrial software for chips, you can simply pay for the license.

Use it for a few years first, and Xingyi Technology will develop one by itself a few years later.

Or cooperate with Huawei on research and development, both are possible.

There is no fear of foreign blockade at all.

However, the ARM architecture is the foundation, so it is better to take advantage of the low price in 2011 and buy it out directly.

ARM is headquartered in England and has branches in the Empire and other countries.

President Norbert Bellomo met with Wang Yi in person:

"It turns out that the mysterious Eastern giant who caused a sensation in the empire is Mr. Wang!"

"Excuse me. Ordering so many chips at once puts me under a lot of pressure."

"Yes, it will be enough for you to use for the next year, hahaha." Norbert said with a smile: "But Director Wang came to us because he wanted to develop his own chips?"

"Yes, you have to dream, although it is difficult."

"Director Wang will definitely succeed!"

Norbert was overjoyed and wished everyone could develop chips so that ARM could sell more licenses and make more money!

"Whether it works or not, you have to try. I wonder how the licensing fee for the ARMV8 architecture is calculated?" Wang Yi asked.

V8, a 64-bit chip architecture just released this year, is much more powerful than V7, a 32-bit chip.

Wang Yi naturally chose the most advanced architecture and directly developed a quad-core 64-bit processor.

NVIDIA Tegra3 and Qualcomm's quad-core processor 8064 are still based on V7, 32-bit architecture, which lags behind.

Wang Yi entered chip research and development in 2011 and directly took over the V8 architecture, which can be regarded as a head start.

Norbert smiled slightly: "The V8 architecture has just been launched, there are few customers buying it, and the price is still very favorable."

Not long after V8 was released, chip manufacturers are waiting to see how its performance will be.

In addition to Apple buying it directly, other companies such as Qualcomm, NVIDIA, Texas Instruments, MediaTek, HiSilicon, etc. are continuing to use V7 and wait and see V8.

If V8 is good, they will follow up.

If the V8 crashes, the V7 is more cost-effective.

Norbert explained: "With the V8 architecture, the initial one-time fee is US$200 million. Later, you will pay a royalty fee of 30 cents for each chip, which is very cost-effective."

It can only be said that ARM is clear.

Regardless of whether the chip development succeeds or fails, $200 million will be paid first.

If successful, a royalty fee of 30 cents per chip will be paid based on chip production.

Of course, the initial $200 million is small money.

There will be a steady flow of royalties in the later period, and the royalties calculated on a per-unit basis are the bulk of it!

But the current royalties are very cheap, the same price as Apple.

Wang Yi murmured: "The royalty fee is 30 cents per chip. Then I will only do research and development in the next two years. If it is tape-out, it will only be a thousand chips at most, which is enough. A total of 300 US dollars in royalties?"

"..." Norbert was a little disappointed: "Theoretically, that's the case."

He thought he could make tens of millions of dollars in royalties when big clients came to visit him and spent a lot of money.

As a result, the operation was as fierce as a tiger, and the two-year royalties were only US$300...

Send away the beggar!

Wang Yi looked at the other party's ugly face and said quietly:

"It's okay. There is a high probability that we will be able to produce chips in the third year. It is estimated that we will produce five million chips a year, which is 150 million dollars in royalties. Of course, if the research and development fails, then forget it."

Norbert continued to remain silent: "..."

With a royalty fee of 150 million U.S. dollars in three years, do you really think that we, ARM, are beggars?

But for an emerging semiconductor company to successfully develop and produce five million chips in three years is pretty good...

It is even possible that Xingyi Technology will go bankrupt before three years. By then, they will not be able to earn a penny in royalties.

This is not possible!

Norbert was thinking a lot and changed the subject:

"Director Wang, in addition to this volume-based calculation method, our royalties also have a buyout system. For example, a one-time buyout of ten years only costs... 2000 million knives!"

Well, no matter how many years it takes for your chip to be successfully developed, I will earn the $2000 million in royalties in one go.

Even three years later, Xingyi Technology's chip development failed and went bankrupt, with $2000 million in pocket, and ARM would not retreat a cent!

Compared with earning 150 million U.S. dollars in royalties in three years, it is more cost-effective to earn 2000 million U.S. dollars in one go.

In Norbert's view, Wang Yi's self-developed chip has a high probability of failure.

If you charge royalties on a per-pixel basis, you may not be able to earn back 2000 million U.S. dollars in 20 years.

How can such an advanced thing like a chip be so easy to study?

Especially Xia Guo Enterprise, which can’t even make mobile phones well, why does it still make chips? What a joke.

It is as powerful as Huawei HiSilicon. After researching for so long, K3V1 was finally produced and tape-out was successful. However, the performance was too poor to be mass-produced, so it was considered a complete failure.

Whether it will continue to invest in the development of a second chip is a question.

Texas Instruments, the stronger chip overlord, is also about to give up its mobile chip business, and Nvidia may not be able to hold on for long.

These giants can't handle chip research and development, let alone making clothes by themselves?

Norbert even thinks that whether the Xingyi mobile phone can be made is a problem.

And Wang Yi’s chip research and development will fail 99% of the time!

It is more reliable to earn a large amount of royalties in one go.

Wang Yi shook his head: "Who can say for sure what will happen in the future? If it really fails, it will cost 2000 million dollars. It is more cost-effective to charge based on volume."

"..." Norbert's heart skipped a beat, this guy couldn't catch it!

"Let's do this, Director Wang. The ten-year royalty buyout only costs 1000 million knives, or million!"

Can the fat sheep delivered to your door allow you to go back unscathed?

joke!

Leave at least 1000 million dollars!

Even if the research and development of Wang Yi's chip failed, ARM still earned 1000 million U.S. dollars in royalties!

Wang Yi was secretly happy. The price was very good, but he could still cut it off a little more.

ARM never imagined that Wang Yi would continue to invest in chip research and development until it succeeds!

What’s more, many years later, Xingyi Semiconductor’s annual output is not 5 million chips, but 50 million chips, or even more!

At thirty cents a piece, the annual licensing fee is more than $1500 million.

The development of the smartphone market is so terrifying that Norbert could not have expected it, and neither could ARM!

Otherwise, I would not have given Apple the 30 cent unit price, and I would regret it until my intestines turned green.

But at this price, we can still negotiate!

Wang Yi frowned: "Ten-year buyout, 143 million U.S. dollars in royalties? If I succeed in research and development in three years, it will be equivalent to a total of million U.S. dollars in royalties in the next seven years, an average of million U.S. dollars a year. That is different from the current situation Pay royalties according to the amount, million dollars a year, there is no difference! It is not cost-effective."

Norbert was stunned. It seemed that this was true. It was indeed not cost-effective. So what should we do?

Just when Norbert was thinking about how to continue to trick Wang Yi into spending more money to buy out, Wang Yi spoke:

"Unless you give me the V8 architecture and buy out the license permanently!"

Permanent buyout authority? Does this make sense?