Where is godson innovation? Where is the meaning?

Seeing that godson signed a contract with MIPS, he wrote an article titled failed to sign a contract with MIPS's proprietary CPU core strategy. He did not expect Sina to push it to a major problem, in fact, the old man just wants to sign the MIPs with Longson to demonstrate the mistake of the country's strategy of pushing its proprietary intellectual property right CPU core over the past decade. As for Longson, the old man believes that there is nothing wrong with the country's huge investment, the so-called experts and reporters on the Internet are blindly blaming longcore, and the old man is not angry with his thoughts.

1. Is there any infringement on godson?

Legally speaking, although godson has just recently obtained MIPS authorization, godson has not been infringed. In the early stage, godson mainly developed the cpu Of The mips32 architecture. As MIPS and other foreign companies did not pay much attention to the Chinese market in the past few years, many patents have not been filed in China. Therefore, the patents for the mips32 instruction set are not valid in mainland China and many countries in Asia and Africa. As long as longcore products are not exported, there will be no infringement issues, later, the CPU of the mips64 architecture developed by godson was indirectly authorized by St, so there was no infringement problem. It should be said that godson handled well in terms of intellectual property rights.

2. Why didn't godson be authorized until now?

Someone may ask why godson didn't sign the MIPs authorization at the beginning because of his instruction set. People who hold this question may not understand MIPS, I do not know about the instruction set authorization. As an expert, I was familiar with Cai Qingsheng, president of MIPS China and Hu weiwu, in fact, the computing Institute and MIPS have started to discuss authorization matters as early as a few years ago. For this reason, the global president of MIPS has also personally visited the computing Institute. Unfortunately, the authorization fee proposed by MIPs is as high as USD 5 million, it is dozens of times higher than arm. Should godson be authorized several years ago? Obviously, no. As for the specific cost of this authorization, the old hacker is not clear.

3. Is there any innovation in godson?

I am not familiar with the CPU. I suspect that the innovation of the dragon core is obviously an idiot for the knowledge of integrated circuits. The dragon core only uses the MIPs instruction set, and all the architecture design is self-developed, just as Intel and AMD use the x86 instruction set, can you say AMD has never been innovative? We all know that arm is dominant in mobile devices, but arm is authorized with a complete CPU kernel, and only one intel can modify the ARM core globally, even Intel is authorizing the ARM architecture, not just instruction sets. You can say that there is no innovation in Intel's arm cpu, obviously not.

4. Where is the National CPU strategy wrong?

From Ark, hand core, guoxin ..., Over the years, China has been vigorously promoting its proprietary intellectual property rights (CPU cores). Among these so-called proprietary rights, only godson uses the industry's popular instruction sets. Even so, the road to industrialization is already very difficult, several other CPUs with independent intellectual property rights even have no chance to survive, not to mention the so-called experts, over the past few years, college students in the Microelectronics curriculum can compile their own instruction sets through the open source code on the Internet. Is it useful?

5. Where is the significance of godson?

Some people say that since the industrialization of the dragon core is very difficult, since the dragon core is unlikely to replace intel into the mainstream CPU of the PC, the dragon core has no meaning of survival, this is more ridiculous, although the veteran did not agree with Li guojie's emphasis on the value of the dragon core on the grounds of national security, the national security is indeed part of the reason for the existence of the dragon core. The dragon core should not replace its own positioning with Intel, nor should it replace it, however, with godson, we can develop our own servers, routers, and even military products. Can you say these products have no value for national security and the existence of godson has no value? There is no such thing as a backdoor in the instruction set.

The other value of Godson is in the embedded field. It is better to promote the non-infringing and free mips32 compatible core in China than to spend a huge amount of money to purchase the ARM core. After all, it is in the embedded non-mobile field, MIPs has its own uniqueness. In the digital TV Field, MIPS's market share even exceeded arm. Unfortunately, godson has been emphasizing competing with intel on PC CPU over the past few years, the loss of a huge embedded market.

The other value of Godson is the cultivation of talents. Undoubtedly, compared with Europe and America, the local microelectronics in terms of cutting-edge integrated circuit design is far from being used as a R & D institution, even if it is operated by a commercial company, generally, companies involving high-tech products have to set up R & D centers in the United States. Only one reason is the lack of talents. Although longcore is not doing well in industrialization, at least many excellent talents have been trained for the country.

6. What is the fault of godson?

I have mentioned a lot about the value of the dragon core. The old man also talked about the mistakes of the dragon core over the years. First, the dragon core should not put too much energy into the PC CPU field, but should concentrate its resources on the server and embedded fields, long core cannot compete with intel in the PC field. Even if intel occasionally speaks a few flattering words out of courtesy, it even includes netbooks. Essentially, Netbooks are still computers, linux computers will never become popular in China, because Linux programs are not popular and ease of use is far more difficult than Windows. Another major mistake of Godson is that it is not enough to invest in software, in fact, the key to competition for processors is not the hardware design, but the software's support for CPU. Without stable software support, advanced CPU can only be junk, this is also the reason why the number of software and hardware employees in many major international factories is as high as or higher. With the strong software team and management, godson has no future.

Longson's mistakes in the embedded field can be seen from the rise of junzheng, Beijing. In recent years, the Institute of computing has even made great breakthroughs in the embedded field based on MIPS instruction sets, as well as Dr. Hu weiwu's Liu Qiang, the number of CPUs has exceeded 10 million. We can see that MIPs is not out of the market, but there is a problem with the focus of godson.

VII. Hu weiwu

I have my own thoughts on longcore and have communicated with Hu weiwu many times. Although many problems are rare to be consistent, I have always been very respectful to Hu weiwu for no reason, at least Hu weiwu is a very dedicated and innocent scientist. Compared with some so-called experts who are only fooled, Hu weiwu has much greater value to the country, the country not only needs an integrated circuit entrepreneur like Limin, CEO of ruixinwei, but also a scientist like Hu weiwu. (Author, old author)

Postscript: many of my friends commented that the old man shouldn't have defended the dragon core. Some friends even speculated that the old man was under pressure from some aspects, in the previous article, the old man pointed out that the signing of MIPS with godson means that the strategy of CPU core of proprietary intellectual property rights failed. Otherwise, godson's contribution to proprietary intellectual property rights was unfortunately solved by some reporters, the country team like longcore needs to do some research in the strategic field, even if longcore has such a problem, whether or not it is recognized by the entire industry.

Laoyaoba: IC community-semiconductor Dynamics

[Copyright statement]: This article is original to the old man. You are welcome to repost it, but you must specify "Source: Old Man ba" and other relevant information (author's name, etc.) and keep the original article link.