Video Game Empire

Since the mid-1980s, when the RISC architecture was technically mature and entered the stage of history, the battle between the complex instruction set (CISC) and the reduced instruction set (RISC) of processors has been fierce.

From an objective point of view, both have their own advantages and disadvantages. However, throughout the 1980s and 1990s, the academic community held that CISC was outdated and RISC was the development direction of advanced technology. The two most important leaders are Paterson from Berkeley and Hennessy from Stanford.

At that time, almost all universities in the United States used the textbooks co-edited by these two big cows, the two compulsory courses for computer majors, "Computer Principles" and "Computer System Architecture". For a long time, a lot of content in the book is based on the MIPS architecture invented by Hennessy himself as a case, and the RISC architecture is analyzed and explained in detail, and the CISC architecture is rarely mentioned.

If the 8086 processor comes out a few years later, I believe that Intel will never take the risk of the world and embark on the road of no return to CISC. But unfortunately, when Intel was developing the 8086, the RISC architecture had not yet appeared. In order to be compatible with 8086, the 286 and 386 developed by Intel can only continue to use the CISC instruction set, and go further and further on the road of fighting against the world.

Of course, the previous Intel in Li Xuan's memory was incomparably brilliant. Even if it fought against the world by itself, Intel still took advantage of IBM's PCs, and then a large number of PC-compatible machines that were listed in the market. Optimistic about war. It not only formed a monopoly in the desktop-level CPU market, but also counterattacked the server-level CPU market, grabbing a large share from various RISC processors.

But in this life, because of the emergence of the butterfly Li Xuan, there is a question mark as to whether Intel can still succeed. Li Xuan can guess that the proposal to enter the personal computer market must have been placed in front of the senior management of IBM's board of directors at this time, and he has already prepared a carefully prepared meal for the Big Blue.

The main course of this feast is the ERM architecture processor that Oriental Lab is developing with all its strength. In addition to avoiding the harassment of the media, the reason why Li Xuan has a regular schedule some time ago is actually because the research and development of ERM has entered the most critical moment, so he put aside all other complicated things, also Dedicated and dedicated to join the ERM research, only back and forth between home and studio every day.

Finally, a few days ago, the final good news came from the laboratory. The research and development of ERM has achieved initial success, and all designs have been completed. Next, it will be sent to the United States for tape-out. The processor enters the tape-out stage, which means that the design is basically completed. The next step is to test the sample CPU produced by tape-out, find errors and defects, and then make improvements.

It can be said that the ERM processor used by Li Xuan to prepare for the competition with IBM's PC has made a breakthrough. Because according to normal circumstances, it generally only takes half a year from the success of the tape-out to the official listing of the CPU.

In fact, Dongfang Electronics does not need to wait for the official launch of the CPU. As long as it is confirmed that the tape-out samples have no major defects, Aikang Computer and Dongfang Software and other branches will immediately start the next generation of personal computers according to the relevant technical data of the ERM architecture. Software development work.

When the ERM processor is officially put into production, the research and development of other hardware and software will be almost done. Orient Electronics could release its own new generation of personal computers before the PC hits the market, and take IBM by surprise.

But Li Xuan's mind is very clear. Although the ERM processor is a RISC architecture processor that represents the advanced direction of technology, its early performance is actually very mediocre. Even to say it's mediocre is a bit elevated. According to many industry insiders in later generations, until DEC joined the ARM camp and developed StrongARM, ARM was more like a microcontroller than a microprocessor.

The reason why Li Xuan decided to plagiarize the ARM architecture whose performance is not outstanding, and then "original" an ERM architecture, was actually the effect of the time difference. Using Tian Ji's method to deal with other middle-class horses in horse racing, he used 32-bit ERM1 to suppress Intel's 16-bit 8086 and 80286.

When Intel's 32-bit 80386 processor was launched, Dongfang Electronics had already established many barriers in the personal computer market by relying on its advantages in operating systems and application software. At that time, the 80386 processor did not have a fair chance to compete with the ERM processor at all, because no supplier on the market was willing to make great efforts to make their products compatible with Intel's new processors.

Li Xuan may be able to hack IBM and Intel by tricks, but in the workstation field, another rapidly rising microcomputer market, ERM is unlikely to be very competitive in the short term. If a personal computer is a microcomputer with ordinary performance, then a workstation is a more professional, higher-performance microcomputer, and it also has a name called a server.

Early workstations used Motorola's 32-bit MC68000 processor, and ERM had no advantage over it. The reason why the MC68000 failed to replace the 8086 as the mainstream CPU in the personal PC field is mainly because its price is too high.

In this regard, ERM has the advantage of simple RISC architecture design. The ERM that has just been tape-out only integrates 30,000 transistors, while Motorola's MC68000 integrates 70,000 transistors. Under the same process, the volume of the ERM chip will undoubtedly be much smaller, which means that a wafer can be divided into many more chips, and the manufacturing cost will naturally be cheaper.

However, the high-end workstation market is far less sensitive to price than personal computers, and customers first consider the performance of the computer. This is also the main reason why MC68000 was able to occupy the workstation market in the early 1980s.

Since Li Xuan can't use the existing ERM architecture to seize the workstation computer market that is about to enter the fast-growing period, he must make other layouts. Finding a new processor architecture is undoubtedly the best solution to the problem.

At present, Berkeley University and Stanford University ~www.readwn.com~ each have a research group that studies RISC architecture, led by two later generations, Professor Paterson and Professor Hennessy. When Li Xuan was in the United States, he specially asked Dongfang Electronics to communicate with two universities respectively. Professor Hennessy's MIPS research group was more interested in Li Xuan's business development plan.

Li Xuan also considered that he is more familiar with the MIPS architecture. After all, it is the mainstream architecture in the embedded field next to ARM. Li Xuan can also play the role of his golden finger, so he finally decided to cooperate with Professor Hennessy of Stanford University. Jointly develop commercial processors with MIPS architecture.

In fact, the architecture being developed by Professor Paterson of Berkeley University is also very good. The SPARC architecture of the later Sun Company (Sun) was developed on this basis. Of course, the Chinese are more familiar with the MIPS architecture, and later China's Godson adopted the relevant patent authorization of the MIPS architecture.

After reading the e-mail that Professor Hennessy just sent, Li Xuan quickly sent him an e-mail, giving his opinion on some of the questions he raised in the e-mail. Li Xuan is not unfamiliar with the MIPS architecture. By regenerating the golden finger, the Hennessy team that is exploring can avoid many detours.

Therefore, Professor Hennessy has recently recognized Li Xuan's strength more and more, and their email exchanges have become more frequent. To make the future president of Stanford University look at him with admiration, although Li Xuan knew that he was relying on cheating, he was still a little proud.

PS: Book-Friends Group 3-8-4-1-9-8-7-2-8, interested book friends can join in to chat!